Social Psychology

SocialPsychologybySaulKassinStevenFeinHazelRoseMarkusz-lib.org.pdf

Home >Psychology homework help >Social Psychology

80220_fm_ptg01_hr_i-001.indd 4 30/01/16 12:37 PM

Want to turn C’s into A’s? Obviously, right? But the right way to go about it isn’t always so obvious. Go digital to get the grades. MindTap’s customizable study tools and eTextbook give you everything you need all in one place. Engage with your course content, enjoy the flexibility of studying anytime and anywhere, stay connected to assignment due dates and instructor notifications with the MindTap Mobile app...

and most of all…EARN BETTER GRADES.

TO GET STARTED VISIT WWW.CENGAGE.COM/STUDENTS/MINDTAP

80220_ifc_ptg01_hr.indd 1 13/01/16 8:12 AM

Copyright 2017 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.

Saul KaSSin John Jay College of Criminal JustiCe

Steven Fein Williams College

Hazel RoSe MaRKuS stanford university

Australia • Brazil • Mexico • Singapore United Kingdom • United States

Social Psychology

TenTh ediTion

80220_fm_ptg01_hr_i-001.indd 1 30/01/16 12:37 PM

80220_fm_ptg01_hr_i-001.indd 4 30/01/16 12:37 PM

This is an electronic version of the print textbook. Due to electronic rights restrictions, some third party content may be suppressed. Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. The publisher reserves the right to remove content from this title at any time if subsequent rights restrictions require it. For valuable information on pricing, previous editions, changes to current editions, and alternate formats, please visit www.cengage.com/highered to search by ISBN#, author, title, or keyword for materials in your areas of interest.

Important Notice: Media content referenced within the product description or the product text may not be available in the eBook version.

Social Psychology, Tenth Edition Saul Kassin, Steven Fein, and Hazel Rose Markus

Product Director: Jon Goodspeed

Product Manager: Melissa Gena

Content Developer: Tangelique Williams-Grayer

Product Assistant: Kimiya Hojjat

Marketing Manager: Melissa Larmon

Content Project Manager: Michelle Clark

Art Director: Vernon Boes

Manufacturing Planner: Karen Hunt

Production Service: Graphic World Inc

Text and Photo Researcher: Lumina Datamatics

Copy Editor: Graphic World Inc

Illustrator: Graphic World Inc

Text Designer: Marsha Cohen

Cover Designer: Irene Morris

Cover Image: Jorge Lizana Photo / Getty Images

Compositor: Graphic World Inc

Printed in Canada Print Number: 01 Print Year: 2016

ALL RIGHTS RESERVED. No part of this work covered by the copyright herein may be reproduced, transmitted, stored, or used in any form or by any means graphic, electronic, or mechanical, including but not limited to photocopying, recording, scanning, digitizing, taping, Web distribution, information networks, or information storage and retrieval systems, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without the prior written permission of the publisher.

Library of Congress Control Number: 2015959689

Student Edition: ISBN: 978-1-305-58022-0

Loose-leaf Edition: ISBN: 978-1-305-86313-2

Cengage Learning 20 Channel Center Street Boston, MA 02210 USA

Cengage Learning is a leading provider of customized learning solutions with employees residing in nearly 40 different countries and sales in more than 125 countries around the world. Find your local representative at www.cengage.com.

Cengage Learning products are represented in Canada by Nelson Education, Ltd.

To learn more about Cengage Learning Solutions, visit www.cengage.com.

Purchase any of our products at your local college store or at our preferred online store www.cengagebrain.com.

For product information and technology assistance, contact us at Cengage Learning Customer & Sales Support, 1-800-354-9706.

For permission to use material from this text or product, submit all requests online at www.cengage.com/permissions.

Further permissions questions can be e-mailed to permissionrequestcengage.com.

80220_fm_ptg01_hr_i-001.indd 2 30/01/16 12:37 PM

WCN: 02-200-203

We dedicate this book to our children: Briana, Marc, Andrew, Malin,

Jordan, Elle, Alina, Hannah, and Krysia.

80220_fm_ptg01_hr_i-001.indd 3 30/01/16 12:37 PM

80220_fm_ptg01_hr_i-001.indd 4 30/01/16 12:37 PM

Preface xvii

about the authors xxv

Part 1 introduction Chapter 1 What Is soCIal psyChology? 2

Chapter 2 DoIng soCIal psyChology researCh 24

Part 2 Social Perception Chapter 3 the soCIal self 52

Chapter 4 perCeIvIng persons 104

Chapter 5 stereotypes, prejuDICe, anD DIsCrImInatIon 154

Part 3 Social influence Chapter 6 attItuDes 208

Chapter 7 ConformIty 262

Chapter 8 group proCesses 308

Part 4 Social Relations Chapter 9 attraCtIon anD Close relatIonshIps 350

Chapter 10 helpIng others 408

Chapter 11 aggressIon 452

Part 5 Applying Social Psychology Chapter 12 laW 504

Chapter 13 BusIness 554

Chapter 14 health anD Well-BeIng 600

Glossary G1

references r1

name Index I1

subject Index I19

Brief Contents

80220_fm_ptg01_hr_i-001.indd 5 30/01/16 12:37 PM

80220_fm_ptg01_hr_i-001.indd 6 30/01/16 12:37 PM

vii

Contents

Preface xvii about the author xxv

Part 1 introduction

What is social Psychology? | 6 Defining Social Psychology 6 Social Psychological Questions and Applications 7 The Power of the Social Context: An Example of a Social Psychology Experiment 8 Social Psychology and Related Fields: Distinctions and Intersections 9 Social Psychology and Common Sense 11

a Brief history of social Psychology | 12 The Birth and Infancy of Social Psychology: 1880s–1920s 12 A Call to Action: 1930s–1950s 13 Confidence and Crisis: 1960s–Mid-1970s 15 An Era of Pluralism: Mid-1970s–2000s 15

social Psychology today: What is trending now? | 16 Integration of Emotion, Motivation, and Cognition 17 Genetic and Evolutionary Perspectives 17 Cultural Perspectives 18 Behavioral Economics, Political and Moral Issues, and Other Interdisciplinary Approaches 19 The Social Brain and Body 20 New Technologies and the Online World 21

revieW | 23

toP 10 Key Points in ChaPter 1 | 23

Key terms | 23

Why should you learn about research methods? | 27

developing ideas: Beginning the research Process | 27 Getting Ideas and Finding Out What’s Been Done 28 Hypotheses and Theories 28 Basic and Applied Research 29

refining ideas: defining and measuring social Psychological variables | 29 From the Abstract to the Specific: Conceptual Variables and Operational Definitions 30 Measuring Variables: Using Self-Reports, Observations, and Technology 31

1. What is Social Psychology? | 2

2. doing Social Psychology Research | 24

vii

80220_fm_ptg01_hr_i-001.indd 7 30/01/16 12:37 PM

viii Contents

testing ideas: research designs | 33 Descriptive Research: Discovering Trends and Tendencies 33 Correlational Research: Looking for Associations 36 Experiments: Looking for Cause and Effect 38 Meta-Analysis: Combining Results Across Studies 45 Culture and Research Methods 45

ethics and values in social Psychology | 46 Institutional Review Boards and Informed Consent 47 Debriefing 47 Ethics and Consent Online 48 Values and Science: Ongoing Debates and New Controversies 48

revieW | 50

toP 10 Key Points in ChaPter 2 | 50

Key terms | 51

Part 2 Social Perception

3. The Social Self | 52 Putting Common sense to the test | 54

the self-Concept | 55 Rudiments of the Self-Concept 55 Introspection 58 Self-Perception 60 Influences of Other People 65 Autobiographical Memories 68 Culture and the Self-Concept 70

self-esteem | 76 The Need for Self-Esteem 76 Are There Gender and Race Differences? 79 Self-Discrepancy Theory 80 The Self-Awareness “Trap” 80 Self-Regulation and Its Limits 84 Ironic Mental Processes 87 Mechanisms of Self-Enhancement 88 Are Positive Illusions Adaptive? 94 Culture and Self-Esteem 95

self-Presentation | 96 Strategic Self-Presentation 97 Self-Verification 99 Self-Monitoring 100

reflections: the multifaceted self | 101

revieW | 102

toP 10 Key Points in ChaPter 3 | 102

Key terms | 103

80220_fm_ptg01_hr_i-001.indd 8 30/01/16 12:37 PM

ix Contents

Putting Common sense to the test | 107

observation: the elements of social Perception | 107 A Person’s Physical Appearance 108 Perceptions of Situations 110 Behavioral Evidence 111 Detecting Truth and Deception 117

attribution: from elements to dispositions | 120 Attribution Theories 120 Attribution Biases 123 Culture and Attribution 128 Motivational Biases 131

integration: from dispositions to impressions | 134 Information Integration: The Arithmetic 134 Deviations From the Arithmetic 135 Perceptions of Moral Character 140

Confirmation Biases: from impressions to reality | 142 Perseverance of Beliefs 142 Confirmatory Hypothesis Testing 145 The Self-Fulfilling Prophecy 146

social Perception: the Bottom line | 149

revieW | 152

toP 10 Key Points in ChaPter 4 | 152

Key terms | 153

4. Perceiving Persons | 104

5. Stereotypes, Prejudice, and discrimination | 154 Putting Common sense to the test | 156

the nature of the Problem: Persistence and Change | 157 Defining Our Terms 157 Racism: Current Forms and Challenges 158 Sexism: Ambivalence, Objectification, and Double Standards 165 Beyond Racism and Sexism: Age, Weight, Sexuality, and Other Targets 170 Being Stigmatized 171 Stereotype Threat: A Threat in the Air 172

Causes of the Problem: intergroup, motivational, Cognitive, and Cultural factors | 176 Social Categories and Intergroup Conflict 176 Social Identity Theory 182 Culture and Social Identity 185 Culture and Socialization 186 How Stereotypes Distort Perceptions and Resist Change 190 Automatic Stereotype Activation 193 The Shooter Bias 193

80220_fm_ptg01_hr_i-001.indd 9 30/01/16 12:38 PM

x Contents

Putting Common sense to the test | 210

the study of attitudes | 210 How Attitudes Are Measured 211 How Attitudes Are Formed 216 The Link Between Attitudes and Behavior 218

Persuasion by Communication | 223 Two Routes to Persuasion 223 The Source 227 The Message 234 The Audience 240 Culture and Persuasion 244

Persuasion by our own actions | 245 Role Playing: All the World’s a Stage 246 Cognitive Dissonance Theory: The Classic Version 247 Cognitive Dissonance Theory: A New Look 251 Alternative Routes to Self-Persuasion 254 Ethical Dissonance 257 Cultural Influences on Cognitive Dissonance 259

Changing attitudes | 259

revieW | 260

toP 10 Key Points in ChaPter 6 | 260

Key terms | 261

6. Attitudes | 208

reducing the Problem: social Psychological solutions | 196

Intergroup Contact 196 Intergroup Friendships and Extended Contact 198 The Jigsaw Classroom 199 Shared Identities 199 Trust, Belonging, and Reducing Stereotype Threat 200 Exerting Self-Control 201 Changing Cognitions, Cultures, and Motivations 204

revieW | 206

toP 10 Key Points in ChaPter 5 | 206

Key terms | 207

Part 3 Social influence

80220_fm_ptg01_hr_i-001.indd 10 30/01/16 12:38 PM

xi Contents

Putting Common sense to the test | 265

social influence as “automatic” | 265

Conformity | 268 The Early Classics 268 Why Do People Conform? 271 Majority Influence 275 Minority Influence 278 Culture and Conformity 281

Compliance | 283 Mindlessness and Compliance 284 The Norm of Reciprocity 285 Setting Traps: Sequential Request Strategies 286 Assertiveness: When People Say No 290

obedience | 291 Milgram’s Research: Forces of Destructive Obedience 293 Milgram in the Twenty-First Century 298 Lingering Questions 300 Defiance: When People Rebel 301

the Continuum of social influence | 303 Social Impact Theory 304 Perspectives on Human Nature 305

revieW | 306

toP 10 Key Points in ChaPter 7 | 306

Key terms | 307

7. Conformity | 262

8. Group Processes | 308 Putting Common sense to the test | 310

fundamentals of groups | 311 What Is a Group? Why Join a Group? 311 Key Features of Groups: Roles, Norms, and Cohesiveness 312 Culture and Cohesiveness 316

individuals in groups: the Presence of others | 316 Social Facilitation: When Others Arouse Us 317 Social Loafing: When Others Relax Us 320 Culture and Social Loafing 321 Deindividuation 322

80220_fm_ptg01_hr_i-001.indd 11 30/01/16 12:39 PM

xii Contents

group Performance: Problems and solutions | 325 Losses and Gains in Groups 325 Brainstorming 326 Group Polarization 328 Groupthink 329 Communicating Information and Utilizing Expertise 332 Goals and Plans in Groups 335 Training and Technology 336 Virtual Teams 337 Culture and Diversity 338 Collective Intelligence: Are Some Groups Smarter Than Others? 339

Conflict: Cooperation and Competition Within and Between groups | 339 Mixed Motives and Social Dilemmas 340 Negotiation 343 Culture and Negotiation 344 Finding Common Ground 347

revieW | 348

toP 10 Key Points in ChaPter 8 | 348

Key terms | 349

Part 4 Social Relations

Putting Common sense to the test | 352

need to Belong: a fundamental human motive | 352 The Thrill of Affiliation 354 The Agony of Loneliness 355

the initial attraction | 358 Familiarity: Being There 359 Physical Attractiveness: Getting Drawn In 362 First Encounters: Getting Acquainted 369 Mate Selection: The Evolution of Desire 376

Close relationships | 382 The Intimate Marketplace: Tracking the Gains and Losses 384 Types of Relationships 387 How Do I Love Thee? Counting the Ways 389 Culture, Attraction, and Close Relationships 395 Relationship Issues: The Male–Female Connection 396

revieW | 406

toP 10 Key Points in ChaPter 9 | 406

Key terms | 407

9. Attraction and Close Relationships | 350

80220_fm_ptg01_hr_i-001.indd 12 30/01/16 12:39 PM

xiii Contents

10. helping others | 408

11. Aggression | 452

Putting Common sense to the test | 410

evolutionary and motivational factors: Why do People help? | 412 Evolutionary Factors in Helping 412 The Evolution of Empathy 415 Rewards of Helping: Helping Others to Help Oneself 418 Altruism or Egoism: The Great Debate 420 Helping as a Default? 424

situational influences: When do People help? | 424 The Bystander Effect 424 Time Pressure 431 Moods and Helping 432 Prosocial Media Effects 434 Role Models and Social Influence 435

Personal influences: Who is likely to help? | 437 Are Some People More Helpful Than Others? 438 What Is the Altruistic Personality? 438 Culture and Helping 441

interpersonal influences: Whom do People help? | 443 Perceived Characteristics of the Person in Need 443 A Little Help for Our Friends, and Others Like Us 444 Gender and Helping 446 Culture and Who Receives Help 447 The Helping Connection 449

revieW | 450

toP 10 Key Points in ChaPter 10 | 450

Key terms | 451

Putting Common sense to the test | 454

What is aggression? | 455

Culture, gender, and individual differences | 456 Culture and Aggression 457 Gender and Aggression 463 Individual Differences 464

origins of aggression | 466 Evolutionary Psychology 466 Genes, Hormones, and the Brain 468 How Is Aggression Learned? 470 Culture and Honor 474

80220_fm_ptg01_hr_i-001.indd 13 30/01/16 12:39 PM

xiv Contents

situational influences on aggression | 477 The Frustration–Aggression Hypothesis 477 The Frustration–Aggression Hypothesis: Does the Evidence Support It? 478 Negative Affect 478 Arousal 480 Thought: Automatic and Deliberate 481 The Struggle for Self-Control: Rumination, Alcohol, and Other Factors 482 Situational Influences: Putting It All Together 485

media effects | 486 Violence in Popular Media: Does Life Imitate Art? 486 Pornography 492 Objectification and Dehumanization 494

reducing aggression and violence | 494 Thoughts, Feelings, and Self-Control 495 Sociocultural Approaches 498 Multiple-Level Approaches: Programs to Prevent Violence and Bullying 499

revieW | 502

toP 10 Key Points in ChaPter 11 | 502

Key terms | 503

Part 5 Applying Social Psychology

Putting Common sense to the test | 506

eyewitness testimony | 508 Perceiving the Crime 510 Storing the Memory 512 Identifying the Culprit 514 Testifying in Court 517 Improving Eyewitness Justice 519 The Alibi: Eyewitness to Innocence 520

Confessions | 521 Suspect Interviews: The Psychology of Lie Detection 521 Police Interrogations: Social Influence Under Pressure 523 False Confessions: Why Innocent People Confess 524 Confessions in the Courtroom 527 Pleading Guilty in the Shadow of Trial 528

Jury decision making | 529 Jury Selection 530 The Courtroom Trial 536 Jury Deliberation 540

Posttrial sentencing and Prison | 545 The Sentencing Process 545 The Prison Experience 546

12. Law | 504

80220_fm_ptg01_hr_i-001.indd 14 30/01/16 12:40 PM

xv Contents

13. Business | 554

Perceptions of Justice | 549 Justice as a Matter of Procedure 549 Culture, Law, and Justice 550

Closing statement | 552

revieW | 552

toP 10 Key Points in ChaPter 12 | 552

Key terms | 553

Putting Common sense to the test | 556

Personnel selection | 559 The Typical Job Interview 559 “Scientific” Alternatives to Traditional Interviews 562 Affirmative Action 566 Culture and Organizational Diversity 568

Performance appraisals | 571 Supervisor Ratings 571 Self-Evaluations 572 New and Improved Methods of Appraisal 573 Due-Process Considerations 574

leadership | 575 The Classic Trait Approach 575 Contingency Models of Leadership 576 Transactional Leadership 577 Transformational Leadership 578 Leadership Among Women and Minorities 580 Cultural Influences on Leadership 582

motivation at Work | 583 Economic Reward Models 584 Bonuses, Bribes, and Intrinsic Motivation 585 Equity Considerations 587 The Progress Principle 589

economic decision making | 590 The Symbolic Power of Money 590 Social Influences in the Stock Market 592 Commitment, Entrapment, and Escalation 595

revieW | 597

toP 10 Key Points in ChaPter 13 | 597

Key terms | 599

80220_fm_ptg01_hr_i-001.indd 15 30/01/16 12:40 PM

xvi Contents

14. health and Well-Being | 600 Putting Common sense to the test | 602

stress and health | 603

What Causes stress? | 604 Crises and Catastrophes 605 Major Life Events 607 Microstressors: The Hassles of Everyday Life 608

how does stress affect the Body? | 610 The General Adaptation Syndrome 610 What Stress Does to the Heart 611 What Stress Does to the Immune System 614 The Links Between Stress and Illness 616

Processes of appraisal | 618 Attributions and Explanatory Styles 618 The Human Capacity for Resilience 619 Pollyanna’s Health 623

Ways of Coping With stress | 624 Problem-Focused Coping 625 Emotion-Focused Coping 628 Proactive Coping 632 Culture and Coping 635

treatment and Prevention | 637 Treatment: The “Social” Ingredients 637 Prevention: Getting the Message Across 638

the Pursuit of happiness | 641 Does Money Buy Happiness? 642 Emerging Science on How to Increase Happiness 646

revieW | 649

toP 10 Key Points in ChaPter 14 | 649

Key terms | 650

glossary g1

referenCes r1

name index i1

suBJeCt index i19

80220_fm_ptg01_hr_i-001.indd 16 30/01/16 12:40 PM

xvii

The opening years of the twenty-first century have proved to be an exciting and tumultuous time—more so, it seems, than any in recent memory. On the one hand, thanks to the rise of Facebook, Twitter, Skype, YouTube, Instagram, Snapchat, and other social media, all of which are available on mobile apps, wherever we are, it has never been easier to be “social”—to talk to others or share opinions, pictures, music, and footage of live events as they occur with people from all corners of the world. On the other hand, deep social and po- litical divisions; religious and ethnic conflicts all over the world; economic disparities; and an ever-present threat of terrorism surround us. As Charles Dickens (1859) said in A Tale of Two Cities, “It was the best of times, it was the worst of times.”

Encircled by its place in science and world events, social psychology—its theories, research methods, and basic findings—has never been more relevant or more important. We used to think of social psychology as a discipline that is slow to change. As in other sciences, we thought, knowledge builds in small in- crements, one brick at a time. Social psychology has no “critical” experiments, no single study can “prove” a theory, and no single theory can fully explain the complexities of human social behavior. While all this remains true, the process of revising this textbook always seems to show us how complex, dynamic, and responsive our field can be. As the world around us rapidly changes, so too does social psychology. Whether the topic is world news, politics, business, health, education, law, travel, sports, or entertainment, social psychology has weighed in.

Despite the promise that it has fulfilled and brings to the future, social psy- chology has recently been rocked by scandal and controversy. Three events in particular have weighed on the field. First, in 2011, a social psychologist in the Netherlands was found to have falsified data that were published in some fifty articles. That case was followed by two other instances of fraud and a paper that survived peer review at JPSP purporting to prove ESP. Second, after an exhaustive multiyear effort to replicate 100 published studies, a group of social psycholo- gists reported in Science, in 2015, that more than half of the findings they sought to replicate failed when retested. This finding was heavily reported in the news media, as seen in The New York Times article, “Many Psychology Findings Not as Strong as Claimed, Study Says.” Third, a “political” controversy has erupted over the question of whether social psychology research is inherently biased by a liberal ideology. This debate—in terms of how ideology can influence what researchers choose to study and how they interpret the results—continues un- abated as we revise this book.

It is clear that social psychology is undergoing a process of self-examination. This has led the field to adopt new, more rigorous methods, statistical practices, and safeguards, and it has led us to raise the bar in the standards we use to decide which new findings to report. What has not changed in this reassessment is the enthusiasm with which we present classic and contemporary social psychology in each and every page of this textbook.

Preface

xvii

80220_fm_ptg01_hr_i-001.indd 17 30/01/16 12:40 PM

xviii Preface

In the competitive college textbook business, it is a rare and special milestone to publish a tenth edition and span thirty years in print. Being the brainchild and inspiration of Sharon Brehm, our first lead author, the inaugural edition of this book was published in 1990—before any of us had access to the Internet or a cell phone and before e-books were an option. In countless ways, the world was a simpler and far different place. Yet human nature—our fundamental need to belong, and to be accepted; our deeply rooted and profound vulnerability to social influence, to satisfy these needs; and the range of settings in which social psychology is on display—has remained very much the same. The continuity of social psychology over time, its contributions to the field of psychology as a whole, and its acceptance within other professions and popular culture, has never been so clear.

We had three main goals for this revision.

1. Our first goal was to present the most important and exciting perspectives in the field as a whole. To communicate the breadth and depth of social psychology, we have self-consciously expanded our coverage to include not only the classics but also the most recent developments in the field—developments that cap- ture new thinking about social neuroscience, evolutionary theory, nonconscious and implicit processes, effects of social media and technology, and cultural influences.

2. In light of questions that have surfaced concerning replicability, our second goal was to try to vet brand new findings in an effort to ensure that the discipline we present will prove accurate over time. No method of vetting is perfect. But as a departure from past practice, we have chosen to exclude any research presented at professional conferences or reported in the news that has not been published in a peer-reviewed journal. For articles newly published, we sought to determine if the findings were consistent with other research.

3. Finally, we want this book to serve as a good teacher outside the classroom. While speaking the student’s language, we always want to connect social psychology to current events in politics, sports, business, law, music, travel, entertainment, the use of social networking sites, and other life domains. We will say more about this later in the preface, in the section, “Connections With Current Events.”

What’s new in this edition

As in the past, we have tried both to capture the essence of social psychology from its inception as well as to reflect the shifts within the field over time and culture. It is our hope that the reader will feel the pulse of our field today in each and every page.

j the Content Comprehensive, up-to-date scholarship Like its predecessors, the tenth edi- tion offers a broad, balanced, mainstream look at social psychology. Thus, it includes detailed descriptions of classic studies from social psychology’s historical warehouse as well as the latest research findings from hundreds of new references. In particular,

goals for this edition

80220_fm_ptg01_hr_i-001.indd 18 30/01/16 12:40 PM

xix Preface

we draw your attention to the following topics, which are either new to this edition or have received expanded coverage:

d The social brain and body (Chapter 1)

d The challenges of doing research across cultures (Chapter 2)

d Ethics and consent in online research (Chapter 2)

d Facebook as a venue for social comparison (Chapter 3)

d Social class as a cultural influence (Chapter 3)

d Attributing mind to machines (Chapter 4)

d Perceptions of moral character (Chapter 4)

d Racial tensions sparked by police shootings and ensuing protests (Chapter 5)

d New research and discussion of dehumanization (Chapter 5)

d Ethical dissonance (Chapter 6)

d Engaged followership model of obedience (Chapter 7)

d Collective intelligence: Are some groups smarter than others? (Chapter 8)

d Uses of technology to train real decision-making groups (Chapter 8)

d New research on online dating (Chapter 9)

d Mate selection and conspicuous consumption (Chapter 9)

d Neuroscience of empathy (Chapter 10)

d Social influences on helping in philanthropy (Chapter 10)

d Evolutionary psychology approaches to aggression (Chapter 11)

d Effects of genes, hormones, and brain functioning on aggression (Chapter 11)

d Alibis as eyewitnesses to innocence (Chapter 12)

d Pleading guilty in the shadow of trial (Chapter 12)

d Cybervetting in personnel selection (Chapter 13)

d Cultural influences on leadership (Chapter 13)

d The link between social class and health (Chapter 14)

d Cultural differences in social support seeking (Chapter 14)

As this nonexhaustive list shows, this tenth edition contains new—and newsy— material. In particular, you may notice that we have zeroed in on new developments in social neuroscience, evolutionary theory, implicit and nonconscious processes, effects of social media and technology, and cultural perspectives—including social class and racial and ethnic groups within cultures. As to this latter point, as social psychology is now a truly international discipline, this book routinely cites new research conducted throughout North and South America, Europe, Asia, Australia, and other parts of the world. We believe that the study of human diversity—from the perspectives of researchers who themselves are a diverse lot—can help students become better informed about social relations as well as about ethics and values.

Connections With Current events To cover social psychology is one thing; to use its principles to explain events in the real world is quite another. Fifteen years

80220_fm_ptg01_hr_i-001.indd 19 30/01/16 12:40 PM

xx Preface

ago, the events of 9/11 changed the world. In different ways not fully discernible, so did the severe economic recession—and slow recovery—in the United States and Europe; the rise of China as an economic power; the changes that have swept through the Middle East, as seen in the sudden rise of ISIS; the increasing threat of violence, as seen in mass shootings in the United States and terrorism in Paris and throughout the world; the racial tensions between police and citizens, which has spawned the “Black Lives Matter” movement; and the ease with which people can now meet, interact, and share services—as seen in the global rise of Uber and Airbnb—through online social networking sites on mobile devices. More than ever, connecting basic theory to real life is the best way to heighten student interest. Over the years, teachers and students have told us how much they value the “newsy” features of our book.

The tenth edition, like its predecessors, is committed to making social psy- chology relevant. Almost every page includes a passage, a quote, a figure, a table, a photo, or a cartoon that refers to people, places, events, music, social trends, and issues that are prominent in contemporary culture. The reader will find stories about the events in Ferguson, Missouri, and elsewhere; Edward Snowden, and his disclosures about mass surveillance; ongoing political debates over same-sex marriage and immigration; cases in the news—such as the South Africa trial of Olympian Oskar Pistorius; and the role of Facebook and other social media in bringing people together—for good purposes and bad.

As in past editions, you will also find in the margins various quotations, song lyrics, public opinion poll results, “factoids,” and links to relevant websites. These high-interest items are designed to further illustrate the connectedness of social psychology to a world that extends beyond the borders of a college campus.

social Psychology and Common sense Several years ago, we introduced a feature that we remain excited about. Building on a discussion in Chapter 1 about the links (and lack thereof) between social psychology and common sense, each substantive chapter opens with Putting Common Sense to the Test, a set of true– false questions designed to assess the student’s intuitive beliefs about material later contained in that chapter. Some examples: “Sometimes the harder you try to control a thought, feeling, or behavior, the less likely you are to succeed,” “People often come to like what they suffer for,” and “Opposites attract.” The answers to these questions are revealed in a marginal box after the topic is presented in the text and then explained at the end of each chapter. We think that students will find this exercise engaging. It will also enable them, as they read, to check their intui- tive beliefs against the findings of social psychology and to notice the discrepancies that exist.

j the organization Of all the challenges faced by teachers and textbooks, perhaps the greatest is to put information together in a way that is both accurate and easy to understand. A strong organizational framework helps in meeting this challenge. There is nothing worse for a student than having to wade through a “laundry list” of names, dates, and studies whose interconnections remain a profound mystery. A strong struc- ture thus facilitates the development of conceptual understanding.

But the tail should not wag the dog. Since organizational structure is a means to an end, not an end in itself, we want to keep it simple and unob- trusive. Look through the Table of Contents, and you will see that we present social psychology in five major parts—a heuristic structure that teachers and

80220_fm_ptg01_hr_i-001.indd 20 30/01/16 12:40 PM

xxi Preface

students have found sensible and easy to follow through nine editions. The book opens with two Introduction chapters on the history, subject matter, and research methods of social psychology (Part I). As before, we then move to an intraindividual focus on Social Perception (Part II), shift outward to Social Influ- ence (Part III) and Social Relations (Part IV), and then conclude with Applying Social Psychology (Part V). We realize that some instructors like to reshuffle the deck to develop a chapter order that better fits their own approach. There is no problem in doing this. Each chapter stands on its own and does not require that others be read first.

j the Presentation Even when the content of a textbook is accurate and up to date, and even when its organization is sound, there is still the matter of presentation. As the “teacher outside the classroom,” a good textbook should facilitate learning. Thus, every chapter contains the following pedagogical features:

d An abstract, chapter outline, and common-sense quiz (beginning with Chapter 3).

d Key terms highlighted in the text, defined in the margin, listed at the end of the chapter, and reprinted in an alphabetized glossary at the end of the book.

d Bar graphs, line graphs, tables, sketches, photographs, flow charts, and cartoons that illustrate, extend, enhance, and enliven material in the text. Some of these depict historic images; others, more contemporary, are new to the tenth edition and often “newsy.”

d At the end of each chapter, a numbered list of “Top 10 Key Points” designed to provide students with a shorthand summary the major takeaway messages.

MindTap is a personalized teaching experience with relevant assignments that guide students to analyze, apply, and improve thinking, allowing you to measure skills and outcomes with ease.

d Personalized Teaching: Becomes yours with a Learning Path that is built with key student objectives. Control what students see and when they see it. Use it as-is or match to your syllabus exactly—hide, rearrange, add and create your own content.

d Guide Students: A unique learning path of relevant readings, multimedia, and activities that move students up the learning taxonomy from basic knowledge and comprehension to analysis and application.

d Promote Better Outcomes: Empower instructors and motivate students with analytics and reports that provide a snapshot of class progress, time in course, engagement and completion rates.

In addition to the benefits of the platform, MindTap for Kassin, Fein, and Markus’s Social Psychology features:

d Videos and animations, all based on key social psychology topics and concepts.

d Chapter-opening assignments, including choose-your-own-activity style exercises, videos, animations, and polling questions which all integrate supporting social psychology research.

mindtap for Kassin, fein, and markus’s social Psychology

80220_fm_ptg01_hr_i-001.indd 21 30/01/16 12:40 PM

xxii Preface

Textbooks are the product of a team effort—more today than ever before. We are grateful to Cengage Learning for its commitment to quality as the first priority. First, we want to thank Tangelique Williams-Grayer, our content developer. We also want to express our gratitude to Roman Barnes, our photo researcher, who has helped to make this book so photographically interesting. Finally, we want to thank all those whose considerable talents and countless hours of hard work can be seen on every page: Cassie Carey, Senior Project Manager, Graphic World; Michelle Clark, Senior Content Production Manager; and Kimiya Hojjat, Prod- uct Assistant. We also thank Publisher Jon-David Hague and Product Manager, Melissa Gena.

Over years, several colleagues have guided us through their feedback on prior editions. Every one of these teachers and scholars has helped to make this a better

acknowledgments

j supplements instructor’s resource manual In each chapter of the Online Instructor’s Resource Manual ISBN: 9781305968004, we provide:

d Learning Objectives. A listing of what students should be able to do after reading.

d A Detailed Overview. A comprehensive review, with key points highlighted and key terms listed.

d Lecture/discussion ideas. Many ways to introduce and discuss topics with students, including video suggestions and extended examples.

d Class activity ideas. Helpful prompts, including hand-out suggestions, for in-class activities. Each idea includes a “What if this bombs?” section for when executing an idea does not go as expected.

d Multimedia resources. An annotated list of related videos, websites, YouTube clips, computer programs, and video clip suggestions from the Research in Action collection from Cengage Learning.

d Handouts. Helpful handouts for students that correlate with suggested activities and homework.

d MindTap Instructor’s Resource Center. Descriptions of assets and tools to inte- grate into your course in MindTap.

Cognero Cengage learning testing Powered by Cognero is a flexible, online system that allows you to author, edit, and manage test bank content from multiple Cengage Learning solutions, create multiple test versions in an instant, and deliver tests from your Learning Management System (LMS), your classroom, or wherever you want. ISBN: 9781305856950

PowerPoint, online, features lecture outlines and key images from the text. ISBN: 9781305967755

80220_fm_ptg01_hr_i-001.indd 22 30/01/16 12:40 PM

xxiii Preface

William Adler, Collin County College Craig Anderson, Iowa State University Dan Batson, University of Kansas Lorraine Benuto, University of Nevada, Las Vegas Bryan Bonner, The University of Utah Rebecca Carey, Stanford University Kimberly Coffman, Florida International University John Dovidio, Yale University Jean Egan, Asnuntuck Community College Rebecca Francis, West Virginia State University Rowell Huesmann, University of Michigan Steven Karau, Southern Illinois University John Levine, University of Pittsburgh Mike Mangan, University of New Hampshire, Durham Margo Monteith, Purdue University Richard Moreland, University of Pittsburgh Paul Paulus, University of Texas at Arlington

William Pedersen, California State University, Long Beach

Michele Reich, Drexel University Todd Shackelford, Oakland University Nicole Shelton, Princeton University Jennifer Shibley, Columbia College, Chicago Charles Stangor, University of Maryland Nicole M. Stephens, Northwestern University, Kellogg

School of Management Arlene Stillwell, SUNY College at Potsdam Eric Stocks, University of Texas at Tyler Courtney von Hippel, University of Queensland William von Hippel, University of Queensland Gregory Walton, Stanford University Kipling Williams, Purdue University Elizabeth Williford, Belhaven University

Saul Kassin Steven Fein

Hazel Rose Markus

book. For their invaluable insights, comments, and suggestions, we thank the following reviewers of our more recent editions:

80220_fm_ptg01_hr_i-001.indd 23 30/01/16 12:40 PM

80220_fm_ptg01_hr_i-001.indd 24 30/01/16 12:40 PM

xxv

saul Kassin is Distinguished Professor of Psychology at John Jay College of Criminal Justice, in New York, and Massachusetts Professor of Psychology at Williams College in Williamstown, Massachusetts. Born and raised in Brooklyn, he received his Ph.D. from the University of Connecticut followed by fellowships at the University of Kansas and U.S. Supreme Court and a visiting professorship at Stanford University. In addition to authoring textbooks, he has co-authored and edited Confessions in the Courtroom, The Psychology of Evidence and Trial Procedure, and The American Jury on Trial. Interested in using social psychology to prevent wrongful convictions, Kassin pioneered the scientific study of police interrogations and false confessions several years ago, an interest that continues to this day. Kassin is past president of the American Psychology-Law Society and recipient of its Lifetime Contribution Award. He is also a Fellow of APS and APA. He has testified as an expert witness in state, military, and federal courts and has appeared as a media consultant in documentaries and on national news programs.

steven fein is Professor of Psychology at Williams College, Williamstown, Massachusetts. Born and raised in Bayonne, New Jersey, he received his A.B. from Princeton University and his Ph.D. in social psychology from the University of Michigan. He has been teaching at Williams College since 1991, with time spent teaching at Stanford University in 1999. His edited books include Emotion: Inter- disciplinary Perspectives, Readings in Social Psychology: The Art and Science of Research, and Motivated Social Perception: The Ontario Symposium. He has served on the executive committee of the Society of Personality and Social Psychology and as the social and personality psychology representative at the American Psy- chological Association. His research interests concern stereotyping and prejudice, suspicion and attributional processes, social influence, and self-affirmation theory.

haZel rose marKus is the Davis-Brack Professor in the Behavioral Sciences at Stanford University. She co-directs Stanford SPARQ (Social Psychological Answers to Real-world Questions) and also the Research Institute of the Stanford Center for Comparative Studies in Race and Ethnicity. Before moving to Stanford in 1994, she was a professor at the University of Michigan, where she received her Ph.D. Her work focuses on how the self-system, including current conceptions of self and pos- sible selves, structures and lends meaning to experience. Born in England to English parents and raised in San Diego, California, she has been persistently fascinated by how nation of origin, region of the country, gender, ethnicity, race, religion, and social class shape self and identity. With her colleague Shinobu Kitayama at the Uni- versity of Michigan, she has pioneered the experimental study of how culture and self influence one another. Markus was elected to the American Academy of Arts and Sciences in 1994, is a Fellow of APS, APA, and Division 8, and received the 2008 APA award for Distinguished Scientific Contribution. Some of her recent co-edited and co- authored books include Culture and Emotion: Empirical Studies of Mutual Influence, Engaging Cultural Differences: The Multicultural Challenge in Liberal Democracies, Just Schools: Pursuing Equal Education in Societies of Difference, and Doing Race: 21 Essays for the 21st Century, and Clash! How to Thrive in a Multicultural World.

About the Authors

xxv

80220_fm_ptg01_hr_i-001.indd 25 30/01/16 12:40 PM

80220_fm_ptg01_hr_i-001.indd 26 30/01/16 12:40 PM

Social Psychology

80220_fm_ptg01_hr_i-001.indd 1 30/01/16 12:40 PM

What Is Social Psychology?

This chapter introduces you to the study of social psychology. We begin by defining social psychology and identifying how it is distinct from but related to some other areas of study, both outside and within psychology. Next, we review the history of the field. We conclude by looking forward, with a discussion of the important themes and perspectives that are propelling social psychology today and in the years to come.

80220_ch01_ptg01_hr_002-023.indd 2 27/01/16 11:18 AM

What Is Social Psychology?

1 What Is Social Psychology? | 6

Defining Social Psychology

Social Psychological Questions and Applications

The Power of the Social Context: An Example of a Social Psychology Experiment

Social Psychology and Related Fields: Distinctions and Intersections

Social Psychology and Common Sense

A Brief History of Social Psychology | 12

The Birth and Infancy of Social Psychology: 1880s–1920s

A Call to Action: 1930s–1950s

Confidence and Crisis: 1960s–Mid-1970s

An Era of Pluralism: Mid-1970s–2000s

Social Psychology Today: What Is Trending Now? | 16

Integration of Emotion, Motivation, and Cognition

Genetic and Evolutionary Perspectives

Cultural Perspectives

Behavioral Economics, Political and Moral Issues, and Other Interdisciplinary Approaches

The Social Brain and Body

New Technologies and the Online World

Review

3© Y

ur i A

rc ur

s/ Sh

ut te

rs to

ck .c

80220_ch01_ptg01_hr_002-023.indd 3 27/01/16 11:18 AM

4 Chapter 1 What Is Social Psychology?

We’re certainly not one of the largest animals. But compared to the rest of the animal world, the size of the human brain, relative to the size of our bodies, is massive. Why is this? The most obvious explanation is that we’re smarter, that we’ve mastered our environments to a degree no other animal has. But recent evidence suggests that the relatively huge size of the human brain— and particularly of the neocortex, at its outermost layer—may be due to some- thing more specific, and rather surprising: We have such large brains in order to socialize (Dunbar, 2014; Spunt et al., 2015).

The remarkable success of our species can be traced to humans’ ability to work together in groups, to infer others’ intentions, to coordinate with extended networks of other people. Our brains needed to be able to handle the incredibly complex challenges associated with these tasks. Long ago Aristotle famously ob- served, “Man is by nature a social animal.” Even Aristotle couldn’t have imagined the degree to which that is true, that the social nature of humans seems to be written into our very DNA. Indeed, recent studies of brain activity have found that when the brain is basically at rest, not engaging in any active task, its default pattern of activity seems to involve social thinking, such as thinking about other people’s thoughts and goals. Social neuroscientist Matthew Lieberman puts it, “Evolution has made a bet that the best thing for our brains to do in any spare moment is to get ready for what comes next in social terms” (E. E. Smith, 2013).

The social nature of the human animal is what this book, and the field of social psychology, is all about. The ways in which we are social animals are countless, and they can be obvious or incredibly subtle. We work, play, and live together. We hurt and help each other. We define happiness and success for each other. We forge our individual identities not alone but in the context of other people. We visit family, make friends, have parties, build networks, go on dates, pledge an enduring commitment, decide to have children. We watch others, speculate about them, and predict who will wind up with whom, whether in real life or in popu- lar culture as we keep up with the Kardashians or watch The Bachelorette. Many of us text or tweet each other about everything we’re up to, or we spend lots of time on social networking sites, interacting with countless peers from around the world, adding hundreds or even thousands of “friends” to our social networks. Our moods can fluctuate with the number of virtual friends who “like” our lat-

est posted photo. Even being ignored by a stranger we don’t really care about can be as painful as the experience of real physical pain (Eisenberger, 2015).

You’ve probably seen or at least heard about the movie It’s a Wonderful Life. When the hero, George Bailey, is about to kill himself, the would-be angel Clarence doesn’t save him by showing him how much personal happiness he’d miss if he ended his life. Instead, he shows George how much his life has touched the lives of others and how many people would be hurt if he was not a part of their world. It was these social relationships that saved George’s life, just as they define our own.

Precisely because we need and care so much about social interactions and

A woman eyes her many suitors on a summer 2015 episode of The Bachelorette, as viewers wondered which man the featured bachelorette might choose for possible marriage. The enormous popularity of shows like this illustrates part of the appeal of social psychology—people are fascinated with how we relate to one another.

Ri ck

R ow

el l/

D is

ne y

A BC

T el

ev is

io n

G ro

up /G

et ty

Im ag

“We take our bearings, daily, from others. To be sane is, to a great extent, to be sociable.”

– John Updike

80220_ch01_ptg01_hr_002-023.indd 4 27/01/16 11:18 AM

5 What Is Social Psychology?

relationships, the social contexts in which we find ourselves can influence us profoundly. You can find many examples of this kind of influence in your own life. Have you ever laughed at a joke you didn’t get just because those around you were laughing? Do you present yourself in one way with one group of people and in quite a different way with another group? The power of the social context can also be much more subtle than in these examples, as when others’ unspoken and inaccurate expectations about you may have real and enduring impact on your own behavior and sense of who you are.

The relevance of social psychology is evident in everyday life, of course, such as when two people become attracted to each other or when a group tries to co- ordinate its efforts on a project. Dramatic events can heighten its significance all the more, as is evident in people’s behavior during and after war, terrorist attacks, or natural disasters. In these traumatic times, a spotlight shines on how people help or exploit each other, and we witness some of the worst and best that human relations have to offer. These events invariably call attention to the kinds of ques- tions that social psychologists study—questions about hatred and violence, about intergroup conflict and suspicion, as well as about heroism, cooperation, and the capacity for understanding across cultural, ethnic, racial, religious, and geo- graphic divides. We are reminded of the need for a better understanding of social psychological issues as we see footage of death and destruction in the Middle East or Congo or are confronted with the reality of an all-too-violent world as nearby as our own neighborhoods and campuses. We also appreciate the majesty and power of social connections as we recognize the courage of a firefighter, read about the charity of a donor, or see the glow in the eyes of a new parent. These are all—the bad and the good, the mundane and the extraordinary—part of the fascinating landscape of social psychology.

Social networking plays a key role in social movements today. The “Ice Bucket Challenge” was a viral sensation that spanned the globe in the summer of 2014 (people taking the challenge in Sydney, Australia, are seen in the photo on the left), raising tremendous amount of awareness and money in the fight against the debilitating disease, ALS. The “Black Lives Matter” hashtag spread dramatically on Twitter in 2014 and 2015 in response to the killings of African Americans in several cities in the United States, inspiring a variety of protests and rallies, as in this rally in Charleston, South Carolina, in June 2015 (right).

Sc ot

t B ar

bo ur

/G et

ty Im

ag es

N ew

s/ G

et ty

Im ag

Br en

da n

Sm ia

lo w

sk i/

A FP

/G et

ty Im

ag es

80220_ch01_ptg01_hr_002-023.indd 5 27/01/16 11:19 AM

6 Chapter 1 What Is Social Psychology?

We begin by defining social psychology and mapping out its relationship to sociology and some other disciplines within the field of psychology.

j Defining Social Psychology Social psychology is the scientific study of how individuals think, feel, and be- have in a social context. Let’s look at each part of this definition.

Scientific Study There are many approaches to understanding how people think, feel, and behave. We can learn about human behavior from novels, films, history,

and philosophy, to name just a few possibilities. What makes social psychology different from these artistic and humanistic endeavors is that social psychology is a sci- ence. It applies the scientific method of systematic obser- vation, description, and measurement to the study of the human condition. How, and why, social psycholo- gists do this is the focus of Chapter 2.

How Individuals Think, Feel, and Behave Social psychology concerns an amazingly diverse set of top- ics. People’s private, even nonconscious beliefs and attitudes; their most passionate emotions; their heroic, cowardly, or merely mundane public behaviors—these all fall within the broad scope of social psychology. In this way, social psychology differs from other social sciences such as economics and political science. Research on attitudes offers a good illustration. Whereas economists and political scientists may be interested in people’s eco-

nomic and political attitudes, respectively, social psychologists investigate a wide variety of attitudes and contexts, such as individuals’ attitudes toward particular groups of people or how their attitudes are affected by their peers or their mood. In doing so, social psychologists strive to establish general principles of attitude formation and change that apply in a variety of situations rather than exclusively to particular domains.

Note the word individuals in our definition of social psychology. This word points to another important way in which social psychology differs from some other social sciences. Sociology, for instance, typically classifies people in terms of their nationality, race, socioeconomic class, and other group factors. In contrast, social psychology typically focuses on the psychology of the individual. Even

What Is Social Psychology?

Our social relationships and interactions are extremely important to us. Most people seek out and are profoundly affected by other people. This social nature of the human animal is what social psychology is all about.

IT S

to ck

/J up

ite r I

m ag

social psychology The scientific study of how individuals think, feel, and behave in a social context.

Not only will you learn interesting and relevant research findings throughout the book, you also will learn how social psychologists have discovered this evi- dence. It is an exciting process and one that we are enthusiastic about sharing with you. The purpose of this first chapter is to provide you with a broad overview of the field of social psychology. By the time you finish it, you should be ready and (we hope) eager for what lies ahead.

80220_ch01_ptg01_hr_002-023.indd 6 27/01/16 11:19 AM

7 What Is Social Psychology?

when social psychologists study groups of people, they usually emphasize the behavior of the individual in the group context.

A Social Context Here is where the “social” in social psychology comes into play and how social psychology is distinguished from other branches of psychol- ogy. As a whole, the discipline of psychology is an immense, sprawling enterprise, concerned with everything from the actions of neurotransmitters in the brain to the actions of dancers in a crowded club. What makes social psychology unique is its emphasis on the social nature of individuals.

However, the “socialness” of social psychology varies. Social psychologists sometimes examine nonsocial factors that affect people’s thoughts, emotions, motives, and actions. For example, they may study whether hot weather causes people to behave more aggressively (Anderson, 2012; Ranson, 2014). What is social about this is the behavior: people hurting each other. In addition, social psy- chologists sometimes study people’s thoughts or feelings about nonsocial things, such as people’s attitudes toward Nike versus Adidas basketball shoes. How can attitudes toward basketball shoes be of interest to social psychologists? One way is if these attitudes are influenced by something social, such as whether LeBron James’s endorsement of Nike makes people prefer that brand. Both examples— determining whether heat causes an increase in aggression or whether LeBron James causes an increase in sales of Nike shoes—are social psychological pursuits because the thoughts, feelings, or behaviors either (a) concern other people or (b) are influenced by other people.

The “social context” referred to in the definition of social psychology does not have to be real or present. Even the implied or imagined presence of others can have important effects on individuals (Allport, 1985). For example, if people imagine receiving positive or nega- tive reactions from others, their self-esteem can be affected significantly (Libby et al., 2011; Smart Richman & Leary, 2009). If students imagine having contact with a stranger from another country, their attitudes toward people from that country and their experiences visiting that country can become more positive (Vezzali et al., 2015). And if college students imagine living a day in the life of a professor, they are likely to perform better later on an analytic test; if they imagine instead being a cheerleader, however, they perform worse (Galinsky et al., 2008)!

j Social Psychological Questions and Applications

For those of us fascinated by social behavior, social psychol- ogy is a dream come true. Just look at Table 1.1 and con- sider a small sample of the questions you’ll explore in this textbook. As you can see, the social nature of the human animal is what social psychology is all about. Learning about social psychology is learning about our- selves and our social worlds. And because social psychology is scientific rather than anecdotal, it provides insights that would be impossible to gain through intuition or experience alone.

A well-liked celebrity such as Oprah Winfrey can influence the attitudes and behaviors of millions of people. When Oprah recommends a book, for example, sales of the book are likely to skyrocket.

D an

ny M

ol os

ho k/

La nd

80220_ch01_ptg01_hr_002-023.indd 7 27/01/16 11:19 AM

8 Chapter 1 What Is Social Psychology?

The value of social psychology’s perspective on human behavior is widely recognized. Courses in social psychology are often required or encouraged for students interested in careers in medicine, busi- ness, law, education, and journalism, as well as in psychology and sociology. Although many gradu- ates of social psychology programs hold faculty ap- pointments in colleges or universities, others work in medical centers, law firms, government agen- cies, the military, and a variety of business settings involving management, investment banking, mar- keting, advertising, human resources, negotiating, and social networking.

The number and importance of these appli- cations continue to grow. Judges are drawing on social psychological research to render landmark decisions, and lawyers are depending on it to se- lect juries and to support or refute evidence. Busi- nesses are using cross-cultural social psychological research to operate in the global marketplace, and they are consulting research on group dynamics to foster the best conditions for their work forces. Health care professionals are increasingly aware of the role of social psychological factors in the pre- vention and treatment of disease. Indeed, we can think of no other field of study that offers expertise that is more clearly relevant to so many different career paths.

m T a b l e 1 . 1

examples of Social Psychological Questions

Social Perception: What Affects the Way We Perceive Ourselves and Others?

d Why do people sometimes sabotage their own performance, making it more likely that they will fail? (Chapter 3)

d How do people in East Asia often differ from North Americans in the way they explain people’s behavior? (Chapter 4)

d Where do stereotypes come from, and why are they so resistant to change? (Chapter 5)

Social Influence: How Do We Influence Each Other?

d Why do we often like what we suffer for? (Chapter 6)

d How do salespeople sometimes trick us into buying things we never really wanted? (Chapter 7)

d Why do people often perform worse in groups than they would have alone? (Chapter 8)

Social Interaction: What Causes Us to Like, Love, Help, and Hurt Others?

d How similar or different are the sexes in what they look for in an intimate relationship? (Chapter 9)

d When is a bystander more or less likely to help you in an emergency? (Chapter 10)

d Does exposure to TV violence or to pornography trigger aggressive behavior? (Chapter 11)

Applying Social Psychology: How Does Social Psychology Help Us Under- stand Questions About Law, Business, and Health?

d Can interrogators really get people to confess to serious crimes they did not commit? (Chapter 12)

d How can business leaders most effectively motivate their employees? (Chapter 13)

d How does stress affect one’s health, and what are the most effective ways of coping with stressful experiences? (Chapter 14)

j The Power of the Social Context: An Example of a Social Psychology Experiment

The social nature of people runs so deep that even very subtle clues about our social connection with

others can have a profound effect on our lives. Think about your first weeks of high school or college. If you’re like most students, there probably were times when you felt insecure and wondered if you fit in there. For some groups of students, however, these fears are especially frequent and strong. Students from underrepresented racial or ethnic minority groups, or women entering programs specializing in science, technology, engineering, or mathematics (STEM), are es- pecially vulnerable to such doubts. Social psychological research has found that these concerns can interfere with the academic performance.

Gregory Walton and his colleagues (2015) are among the social psychologists who have studied this issue. For example, they wondered if they could improve the academic success of women at a prestigious engineering school who were taking courses in majors in which the very large majority of students were men. Women in these majors often report that they feel unwelcome and disrespected. To try to counter these concerns, the researchers conducted an experiment in which they provided some students early in their first semester at the engineer- ing school with a brief bit of information suggesting how typical it is that most

80220_ch01_ptg01_hr_002-023.indd 8 27/01/16 11:19 AM

9 What Is Social Psychology?

students—regardless of their gender—go through periods of social stress, disrespect, and feeling that they don’t belong during their freshman year, and that these struggles tend to go away soon after their first year. After reading this information the students completed a pair of brief writing activities based on this information.

Walton and his colleagues wanted to assess how these students would do by the end of their first year compared to other students who did not receive any information about social belonging concerns. The re- sults can be seen in d Figure 1.1. The bars in this graph illustrate the students’ first-year GPAs, as mea- sured on a scale from 0 to 100. The first pair of bars on the left show the GPAs of men and women who did not receive the belongingness intervention. As the difference in height of these two bars indicates, the GPAs of the women were far below that of the men. Compare this to the pair of bars on the right. These are the GPAs of the men and women who received the brief information designed to counter concerns about belonging. With this brief intervention, the women earned GPAs as high as the men. The large gender difference was eliminated, therefore, by just a small amount of social information received early in their first year. This study illustrates the power of the social context—or more accurately, of the perceived social context—on a critically important real-world outcome. You’ll learn more about this and related research by Walton and others in Chapter 5.

j Social Psychology and Related Fields: Distinctions and Intersections

Social psychology is sometimes confused with certain other fields of study. Before we go on, we should clar- ify how social psychology is distinct from these other fields, and we will also illustrate that interesting and significant questions can be addressed through inter- actions between social psychology and these other fields (see Table 1.2 on page 10).

Social Psychology and Sociology Sociologists and social psychologists share an interest in many is- sues, such as violence, prejudice, cultural differences, and marriage. As noted, however, sociology tends to focus on the group level, whereas social psychology tends to focus on the individual level. For example, so- ciologists might track the racial attitudes of the middle class in the United States, whereas social psychologists might examine some of the specific factors that make individuals more or less likely to behave in a racist way toward members of some group.

According to social psychological research described in this chapter and elsewhere in the book, how socially connected students feel with their fellow students can have a significant effect on their academic success.

A nd

er se

n Ro

ss /B

le nd

Im ag

es /G

et ty

Im ag

d F I g U R E 1 . 1

Social Belonging and gPA

Some students early in their first semester at an engineering school were given information designed to reduce their doubts about fitting in and belonging at their school, and other students were not given this information. The bars on the left represent the average first-year grades of the men (red) and women (green) who did not receive the information about belonging. Women’s average first-year grades were far worse than the men’s in this no-treatment condition. The bars on the right illustrate that the women’s underperformance was completely eliminated if they received the information designed to reduce their uncertainty about belonging. Adapted from Walton et al., 2015.

G PA

(0 -1

00 s

ca le

)

Condition

No Treatment Treatment 60

Men Women

80220_ch01_ptg01_hr_002-023.indd 9 27/01/16 11:19 AM

10 Chapter 1 What Is Social Psychology?

In addition, although there are many ex- ceptions, social psychologists are more likely than sociologists to conduct experiments in which they manipulate some variable and de- termine the effects of this manipulation using precise, quantifiable measures.

Despite these differences, sociology and social psychology are clearly related. Indeed, many sociologists and social psychologists share the same training and publish in the same journals. When these two fields inter- sect, the result can be a more complete un- derstanding of important issues. For example, interdisciplinary research on stereotyping and prejudice has examined the dynamic roles of both societal and immediate factors, such as how particular social systems or institutional norms and beliefs affect individuals’ attitudes and behaviors (Foels & Pratto, 2015; Koenig & Eagly, 2014; Rosenthal & Levy, 2013; D. Sanchez et al., 2014).

Social Psychology and Related Areas of Psychology Tell people not very familiar with psychology that you are taking a social psychol- ogy class, and they may say things like “Are you going to start psychoanalyzing me?” or “Great, maybe you can tell me why everyone in my fam-

ily is so messed up.” The assumption underlying these reactions, of course, is that you are studying clinical, or abnormal, psychology. If you base your impressions of psychology primarily by how it’s portrayed in popular culture, you’re likely to miss how incredibly broad and diverse the field is. Although these various areas of psy- chology are related to each other, each has a very different focus.

Clinical psychologists, for example, seek to understand and treat people with psychological difficulties or disorders. Social psychologists do not focus on disor- ders; rather, they focus on the more typical ways in which individuals think, feel, behave, and influence each other. Personality psychology is another area that is often confused with social psychology. However, personality psychology seeks to understand stable differences between individuals, whereas social psychology seeks to understand how social factors affect most individuals regardless of their different personalities. In other words, a personality psychologist may ask, “Is this person outgoing and friendly almost all the time, in just about any setting?” A social psychologist may ask, “Are people in general more likely to seek out friends when they are made anxious by a situation than when they are made to feel relaxed?”

Cognitive psychologists study mental processes such as thinking, learning, remembering, and reasoning. Social psychologists are often interested in these same processes, but they are concerned with these processes more specifically in a social context.

These examples show the contrast between the fields, but, in fact, social psychological theory and research often intersect with these other areas quite a bit. For example, both clinical and social psychology may address how people

m T a b l e 1 . 2

Distinctions between Social Psychology and Related Fields: The Case of Research on Prejudice

To see the differences between social psychology and related fields, consider an example of how researchers in each field might conduct a study of prejudice.

Field of Study Example of How a Researcher in the Field Might Study Prejudice

Sociology Measure how prejudice varies as a function of social or economic class

Clinical psychology Test various therapies for people with antisocial personalities who exhibit great degrees of prejudice

Personality psychology Develop a questionnaire to identify men who are very high or low in degree of prejudice toward women

Cognitive psychology Manipulate exposure to a member of some category of people and measure the thoughts and concepts that are automatically activated (A study of prejudice in this field would, by definition, be at the intersection of cognitive and social psychology.)

Social psychology Manipulate various kinds of contact between individuals of different groups and examine the effect of these manipulations on the degree of prejudice exhibited

80220_ch01_ptg01_hr_002-023.indd 10 27/01/16 11:19 AM

11 What Is Social Psychology?

cope with anxiety, emotion, or pressure in social situations, or how being bullied or stereotyped by others can affect individuals’ physical and mental health (Cheung et al., 2015; M. Knowles et al., 2015; Rosenthal et al., 2015; Young-Jones et al., 2015).

Personality and social psychology are espe- cially closely linked because they complement each other so well. For example, some social psychologists examine how receiving negative feedback or experiencing conflict (social factors) can have different effects on people as a func- tion of their self-esteem (a personality factor), or whether people who are high in empathy (a per- sonality factor) are especially likely to go out of their way to help other people (a social factor) (Krizan & Johar, 2015; Morelli et al., 2015).

Cognitive and social psychology are also closely connected. The last few decades have seen an explosion of interest in the intersection of cognitive and social psychol- ogy. The study of social cognition is discussed in more detail later in this chapter, and it is a focus throughout this text, especially in Part II on Social Perception.

Social Psychology and Other Fields of Study Social psychologists today are doing research that spans across traditional boundaries between fields more than ever. The intersections of social psychology with disciplines such as neuroscience, biology, economics, political science, public health, environmental studies, law, and medicine are increasingly important to contemporary social psychology. We will dis- cuss a bit more about some of these intersections later in this chapter, but you should see these connections throughout this book, most especially in Part V on Applying Social Psychology.

j Social Psychology and Common Sense After reading about a theory or finding of social psychology, you may sometimes think, “Of course. I knew that all along. Everyone knows that.” This “knew-it- all-along” phenomenon often causes people to question how social psychology is different from common sense, or traditional folk wisdom. After all, would any of the following social psychological findings be surprising to you?

d Beauty and brains don’t mix. Physically attractive people tend to be seen as less smart than physically unattractive people.

d People will like an activity more if you offer them a large reward for doing it, caus- ing them to associate the activity with the positive reinforcement.

d People think that they’re more unique than they really are. They tend to underes- timate the extent to which others share the same opinions or interests.

d Playing contact sports or violent video games releases aggression and makes peo- ple less likely to vent their anger in violent ways.

In a minute, we will have more to say about each of these statements. Common sense may seem to explain many social psychological findings after

the fact. The problem is distinguishing commonsense fact from commonsense

Do provocative, sexualized images of women in advertising, such as on the billboard seen here, make people more sexist or prone to treat women in more objectified ways? Do they affect women’s self-esteem or body issues? These are among the questions that social psychology addresses.

Ri ch

ar d

Le vi

ne /D

em ot

ix /C

or bi

s N

ew s/

Co rb

80220_ch01_ptg01_hr_002-023.indd 11 27/01/16 11:19 AM

12 Chapter 1 What Is Social Psychology?

myth. After all, for many commonsense notions, there is an equally sensible- sounding notion that says the opposite. Is it “Birds of a feather flock together” or “Opposites attract”? Is it “Two heads are better than one” or “Too many cooks spoil the broth”? Which are correct? We have no reliable way to answer such ques- tions through common sense or intuition alone.

Social psychology, unlike common sense, uses the scientific method to put its theories to the test. How it does so will be discussed in greater detail in the next chapter. But before we leave this section, one word of caution: Those four “find- ings” listed above? They are all false. Although there may be sensible reasons to believe each of the statements to be true, research indicates otherwise. Therein lies another problem with relying on common sense: Despite offering very com- pelling predictions and explanations, it is sometimes wildly inaccurate. And even when it is not completely wrong, common sense can be misleading in its simplic- ity. Often there is no simple answer to a question such as “Does absence make the heart grow fonder?” In reality, the answer is more complex than common sense would suggest, and social psychological research reveals how such an answer depends on a variety of factors.

To emphasize these points and to encourage you to think critically about so- cial psychological issues before as well as after learning about them, this textbook contains a feature called “Putting Common Sense to the Test.” Beginning with Chapter 3, each chapter opens with a few statements about social psychological issues that will be covered in that chapter. Some of the statements are true, and some are false. As you read each statement, make a prediction about whether it is true or false and think about why this is your prediction. Marginal notes through- out the chapter will tell you whether the statements are true or false. In reading the chapter, check not only whether your prediction was correct but also whether your reasons for the prediction were accurate. If your intuition wasn’t quite on the mark, think about what the right answer is and how the evidence supports that answer. There are few better ways of learning and remembering than through this kind of critical thinking.

A Brief History of Social Psychology People have probably been asking social psychological questions for as long as humans could think about each other. Certainly early philosophers such as Plato offered keen insights into many social psychological issues. But no systematic and scientific study of social psychological issues developed until the end of the nineteenth century. The field of social psychology is therefore a relatively young one. Recent years have marked a tremendous interest in social psychology and an injection of many new scholars into the field. As social psychology is now in its second century, it is instructive to look back to see how the field today has been shaped by the people and events of its first century.

j The Birth and Infancy of Social Psychology: 1880s–1920s

Like most such honors, the title “founder of social psychology” has many poten- tial recipients, and not everyone agrees on who should prevail. Over the years, most have pointed to the American psychologist Norman Triplett, who is credited with having published the first research article in social psychology at the end of

“Psychology has a long past, but only a short history.”

— Herman Ebbinghaus, Summary of Psychology

80220_ch01_ptg01_hr_002-023.indd 12 27/01/16 11:19 AM

13 A Brief History of Social Psychology

the nineteenth century (1897–1898). Triplett’s work was noteworthy because after observing that bicyclists tended to race faster when racing in the presence of oth- ers than when simply racing against a clock, he designed an experiment to study this phenomenon in a controlled, precise way. This scientific approach to studying the effects of the social context on individuals’ behavior can be seen as marking the birth of social psychology.

A case can also be made for the French agricultural engineer Max Ringel- mann. Ringelmann’s research was conducted in the 1880s but wasn’t published until 1913. In an interesting coincidence, Ringelmann also studied the effects of the presence of others on the performance of individuals. In contrast to Triplett, however, Ringelmann noted that individuals often performed worse on simple tasks such as pulling rope when they performed the tasks with other people. The issues addressed by these two early researchers continue to be of vital interest, as will be seen later in Chapter 8.

Some scholars (Haines & Vaughan, 1979; Stroebe, 2012) suggest a handful of other possible examples of the first social psychology studies, including research that Triplett himself cited. These studies also were conducted in the 1880s and 1890s, which seems to have been a particularly fertile time for social psychology to begin to spring roots.

Despite their place in the history of so- cial psychology, these late-nineteenth-century studies did not truly establish social psychol- ogy as a distinct field of study. Credit for this creation goes to the writers of the first three textbooks in social psychology: the English psychologist William McDougall (1908) and two Americans, Edward Ross (1908) and Floyd Allport (1924). Allport’s book in particular, with its focus on the interaction of individuals and their social context and its emphasis on the use of experimentation and the scientific method, helped establish social psychology as the discipline it is today. These authors an- nounced the arrival of a new approach to the social aspects of human behavior. Social psy- chology was born.

j A Call to Action: 1930s–1950s What one person would you guess has had the strongest influence on the field of social psychology? Various social psychologists, as well as psychologists outside of social psychology, might be mentioned in response to this question. But someone who was not a psychologist at all may have had the most dramatic impact on the field: Adolf Hitler.

Hitler’s rise to power and the horrendous consequences that followed caused people around the world to become desperate for answers to social psychological questions about what causes violence, prejudice, genocide, conformity and obedi- ence, and a host of other social problems and behaviors. In addition, many social psychologists living in Europe in the 1930s fled to the United States and helped es- tablish a critical mass of social psychologists who would give shape to the rapidly maturing field. The years just before, during, and soon after World War II marked an explosion of interest in social psychology.

Racers from around the world compete in the Tour de France in July 2014. Would these cyclists have raced faster or slower if they were racing individually against the clock rather than racing simultaneously with their competitors? More generally, how does the presence of others affect an individual’s performance? The earliest social psychology experiments ever done sought to answer questions such as these. Chapter 8 on group processes brings you up to date on the latest research in this area.

To lg

a A

km en

/A na

do lu

A ge

nc y/

G et

ty Im

ag es

80220_ch01_ptg01_hr_002-023.indd 13 27/01/16 11:19 AM

14 Chapter 1 What Is Social Psychology?

In 1936, Gordon Allport (younger brother of Floyd, author of the 1924 textbook) and a number of other social psychologists formed the Society for the Psychologi- cal Study of Social Issues. The name of the society illustrates these psychologists’ concern for making important, practical contributions to society. Also in 1936, a social psychologist named Muzafer Sherif published groundbreaking experimental research on social influence. As a youth in Turkey, Sherif had witnessed groups of Greek soldiers brutally killing his friends. After immigrating to the United States, Sherif drew on this experience and began to conduct research on the powerful influences groups can exert on their individual members. Sherif’s research was cru- cial for the development of social psychology because it demonstrated that it is pos- sible to study complex and important social issues in a rigorous, scientific manner.

Another great contributor to social psychology, Kurt Lewin, fled the Nazi on- slaught in Germany and immigrated to the United States in the early 1930s. Lewin was a bold and creative theorist whose concepts have had lasting effects on the field (e.g., Lewin, 1935, 1947). One of the fundamental principles of social psy- chology that Lewin helped establish was that behavior is a function of the interac- tion between the person and the environment. This position, which later became known as the interactionist perspective (Blass, 1991; Snyder, 2013), emphasized

the dynamic interplay of internal and external factors, and it marked a sharp contrast from other major psycho- logical paradigms during his lifetime: psychoanalysis, with its emphasis on internal motives and fantasies; and behaviorism, with its focus on exter- nal rewards and punishments.

Lewin also profoundly influenced the field by advocating for social psy- chological theories to be applied to important, practical issues. Lewin re- searched a number of practical issues, such as how to persuade Americans at home during the war to conserve ma- terials to help the war effort, how to promote more economical and nutri- tious eating habits, and what kinds of leaders elicit the best work from group members. Built on Lewin’s legacy, ap-

plied social psychology flourishes today in areas such as advertising, business, education, environmental protection, health, law, politics, public policy, religion, and sports. Throughout this text, we draw on the findings of applied social psy- chology to illustrate the implications of social psychological principles for our daily lives. In Part V, three prominent areas of applied social psychology are dis- cussed in detail: law, business, and health. One of Lewin’s statements can be seen as a call to action for the entire field: “No research without action, no action without research.”

During World War II, many social psychologists answered Lewin’s call as they worked for their government to investigate how to protect soldiers from the pro- paganda of the enemy, how to persuade citizens to support the war effort, how to select officers for various positions, and other practical issues. During and after the war, social psychologists sought to understand the prejudice, aggression, and conformity the war had brought to light. The 1950s saw many major contributions

What determines whether people are likely to volunteer to help others, such as dedicating one’s time and energy to work with others to construct a house for those in need? Built on the legacy of Kurt Lewin, applied social psychology contributes to the solution of numerous social problems.

H er

o Im

ag es

In c/

H er

o Im

ag es

/C or

bi s

interactionist perspective An emphasis on how both an individual’s personality and environmental characteristics influence behavior.

80220_ch01_ptg01_hr_002-023.indd 14 27/01/16 11:19 AM

15 A Brief History of Social Psychology

to the field of social psychology. For example, Gordon Allport (1954) published The Nature of Prejudice, a groundbreaking book that continues to inspire research on stereotyping and prejudice more than six decades later. Solomon Asch’s (1951) demonstration of how willing people are to conform to an obviously wrong ma- jority amazes students even today. Leon Festinger (1954, 1957) introduced two important theories—one concerning how people try to learn about themselves by comparing themselves to other people, and one about how people’s attitudes can be changed by their own behavior—that remain among the most influential theo- ries in the field. These are just a sample of a long list of landmark contributions made during the 1950s. With this remarkable burst of activity and impact, social psychology was clearly, and irrevocably, on the map.

j Confidence and Crisis: 1960s–Mid-1970s In spectacular fashion, Stanley Milgram’s research in the early and middle 1960s linked the post–World War II era with the coming era of social revolution. Milgram’s research was inspired by the destructive obedience demonstrated by Nazi officers and ordinary citizens in World War II, but it also looked ahead to the civil disobedience that was beginning to challenge institutions in many parts of the world. Milgram’s experiments, which demonstrated individu- als’ vulnerability to the destructive commands of authority, became the most famous research in the history of social psychology. This research is discussed in detail in Chapter 7.

With its foundation firmly in place, social psychology entered a period of expansion and enthusiasm. The sheer range of its investi- gations was staggering. Social psychologists considered how people thought and felt about themselves and others. They demonstrated why people fail to help others in distress. They examined aggres- sion, physical attractiveness, and stress. For the field as a whole, it was a time of great productivity.

Ironically, it was also a time of crisis and heated debate. Many of the strong disagreements during this period can be understood as a reaction to the dominant research method of the day: the labora- tory experiment. Critics of this method asserted that certain prac- tices were unethical, that experimenters’ expectations influenced their participants’ behavior, and that the theories being tested in the laboratory were historically and culturally limited (Gergen, 1973; Kelman, 1967; Rosenthal, 1976). Those who favored labo- ratory experimentation, on the other hand, contended that their procedures were ethical, their results were valid, and their theoreti- cal principles were widely applicable (McGuire, 1967). For a while, social psychology seemed split in two.

j An Era of Pluralism: Mid-1970s–2000s Fortunately, both sides won. As we will see in the next chapter, more rigorous ethical standards for research were instituted, more stringent procedures to guard against bias were adopted, and more attention was paid to possible cross-cultural differences in behavior. Laboratory experiments continued to dominate, but often with more precise methods. Laboratory experiments did, however, get some com- pany. A pluralistic approach emerged as a wider range of research techniques and questions became established.

This World War II poster featuring Rosie the Riveter was part of the U.S. government’s campaign to encourage women to take jobs in traditionally male- dominated occupations, such as welding. When the war was over and the men who had served in the military returned to the workforce, new advertisements were designed to encourage women to leave these jobs and concentrate on raising families.

Co ur

te sy

o f t

he N

at io

na l A

rc hi

ve s

an d

Re co

rd s

A dm

in is

tr at

io n

80220_ch01_ptg01_hr_002-023.indd 15 27/01/16 11:19 AM

16 Chapter 1 What Is Social Psychology?

Pluralism in social psychology extends far beyond its methods. There are also important variations in what aspects of human behavior are emphasized. For example, social psychologists became more and more interested in processes rel- evant to (as well as to adapt methods from) cognitive psychology. A new subfield was born called social cognition, the study of how we perceive, remember, and interpret information about ourselves and others. Social cognition research con- tinues to thrive today and examines issues important to virtually every major area in social psychology.

Another source of pluralism in social psychology is its development of interna- tional and multicultural perspectives. Although individuals from many countries helped establish the field, social psychology achieved its greatest professional rec- ognition in the United States and Canada. At one point, it was estimated that 75% to 90% of social psychologists lived in North America (Smith & Bond, 1993; Tri- andis, 1994). However, this aspect of social psychology began to change rapidly in the 1990s, reflecting not only the different geographic and cultural backgrounds of its researchers and participants but also the recognition that many social psycho- logical phenomena once assumed to be universal may actually vary dramatically as a function of culture. You can find evidence of this new appreciation of the role of culture in every chapter of this book.

Social psychologists are becoming increasingly interested in cross-cultural research, which helps us break out of our culture-bound perspective. Many of our behaviors differ across cultures. In some cultures, for example, people are expected to negotiate about the price of the products they buy, as in this market in Tunisia. In other cultures, such bargaining would be highly unusual and cause confusion and distress.

Lo rn

e Re

sn ic

k/ Th

e Im

ag e

Ba nk

/G et

ty Im

ag es

Social Psychology Today: What Is Trending Now? The field today continues to grow in the number and diversity of researchers and research topics, areas of the world in which research is conducted, and industries that hire social psychologists and apply their work.

Throughout this text, we emphasize the most current, cutting-edge research in the field, along with the classic findings of the past. In the remainder of the chap- ter, we focus on a few of the exciting themes and perspectives emerging today.

social cognition The study of how people perceive, remember, and interpret information about themselves and others.

80220_ch01_ptg01_hr_002-023.indd 16 27/01/16 11:19 AM

17 Social Psychology Today: What Is Trending Now?

j Integration of Emotion, Motivation, and Cognition In the earlier days of social cognition research in the 1970s and 1980s, the dominant perspective was called “cold” because it emphasized the role of cognition and de- emphasized the role of emotion and motivation in explaining social psychological issues. This was contrasted with a “hot” perspective, focusing on emotion and moti- vation as determinants of our thoughts and actions. Today there is growing interest in integrating both “hot” and “cold” perspectives, as researchers study how individuals’ emotions and motivations influence their thoughts and actions, and vice versa. For example, researchers examine how motivations we aren’t even consciously aware of (such as being motivated to treat others fairly, or being motivated to feel superior to others) can bias how we interact with or interpret information about another person (Dunning, 2015; Higgins & Scholer, 2015; Moskowitz, 2014; Spencer et al., 2005).

One issue illustrating the integration of “hot” and “cold” variables concerns the conflict between wanting to be right and wanting to feel good about oneself, and how these different motivations influence how we process information. On the one hand, we want to be accurate in our judgments about ourselves and oth- ers. On the other hand, we don’t want to be accurate if it means we will learn something bad about ourselves or those closest to us. These goals can pull our cognitive processes in very different directions. How we perform the required mental gymnastics is an ongoing concern for social psychologists (Balcetis et al., 2014; Hardin & Larsen, 2014; Stroessner & Dweck, 2015; Wojcik & Ditto, 2014).

Another theme running through many chapters of this book is the growing in- terest in distinguishing between automatic and controllable processes and in under- standing the relationship between them. How much do we have control over our thoughts and actions, and how vulnerable are we to in- fluences beyond our awareness or control? Are we sometimes influenced by stereotypes even if we don’t want to believe them? Can we train ourselves to regulate ourselves against auto- matic impulses? These are among the ques- tions that social psychologists are studying today (Bargh et al., 2012; Monteith et al., 2013; J. W. Sherman et al., 2014).

j genetic and Evolutionary Perspectives

Recent advances in behavioral genetics—a sub- field of psychology that examines the effects of genes on behavior—has triggered new research to investigate such matters as the extent to which political attitudes are at least partially inherited and the roles that genes play in individuals’ sexual orienta- tion or identity (Bell & Kandler, 2015; Burri et al., 2015; Sanders et al., 2015). The role of genes in an ever-growing array of social behaviors is being bet- ter understood as breakthroughs continue to drive interdisciplinary research involving genetics.

Evolutionary psychology, which uses the principles of evolution to understand human behavior, is another growing area that is sparking new research in social psychology. According to this perspective, to understand a social psychological

Our desire to be accurate in our judgments can sometimes interfere with our desire to feel good about ourselves.

he N

ew Y

or ke

r Co

lle ct

io n

19 97

M an

ko ff

f ro

ca rt

oo nb

an k.

co m

. A ll

rig ht

s re

se rv

ed .

behavioral genetics A subfield of psychology that examines the role of genetic factors in behavior.

evolutionary psychology A sub- field of psychology that uses the principles of evolution to understand human social behavior.

“On the one hand, eliminating the middleman would result in lower costs, increased sales, and greater consumer satisfaction; on the other

hand, we’re the middleman.”

80220_ch01_ptg01_hr_002-023.indd 17 27/01/16 11:19 AM

18 Chapter 1 What Is Social Psychology?

issue such as jealousy, we should ask how tendencies and reactions underlying jealousy today may have evolved from the natural-selection pressures our an- cestors faced. Evolutionary psychological theories can then be used to explain and predict gender differences in attraction, the situational factors most likely to trigger jealousy, and so on (Brase et al., 2014; Perilloux & Kurzban, 2015; Shack- elford & Hansen, 2015). This perspective is discussed in many places throughout the textbook, especially in Part IV on Social Relations.

j Cultural Perspectives Because of the fantastic advances in communication technologies in recent years and the globalization of the world’s economies, it is faster, easier, and more neces- sary than ever before for people from vastly different cultures to interact with one another. Thus, our need and desire to understand how we are similar to and differ- ent from one another are greater than ever as well. Social psychology is currently experiencing tremendous growth in research designed to give us a better under- standing and appreciation of the role of culture in all aspects of social psychology.

What is meant by “culture” is not easy to pin down, as many researchers think of culture in very different ways. Broadly speaking, culture may be considered to be a system of enduring meanings, beliefs, values, assumptions, institutions, and practices shared by a large group of people and transmitted from one generation to the next. Whatever the specific definition, it is clear that how individuals perceive and derive meaning from their world are influenced profoundly by the beliefs, norms, and practices of the people and institutions around them.

Increasing numbers of social psychologists are evaluating the universal gen- erality or cultural specificity of their theories and findings by conducting cross- cultural research, in which they examine similarities and differences across a variety of cultures. More and more social psychologists are also conducting multi- cultural research, in which they examine racial and ethnic groups within cultures.

These developments are already profoundly influencing our view of human behavior. For example, cross-cultural research has revealed important distinctions between the collectivist cultures (which value interdependence and social har- mony) typically found in Africa, Asia, and Latin America and the individualist cultures (which value independence and self-reliance) more commonly found in North America and Europe. The implications of these differences can be seen throughout the textbook.

Consider, for instance, a recent experiment by Alyssa Fu and Hazel Markus (2014) in which Asian-American and European-American high school students worked on a task that involved trying to make words out of scrambled letters. The researchers designed the task to be extremely difficult, and they told each student that he or she did worse than average for the school. Fu and Markus then gave the students a chance to work on this task a second time. How much effort would they put into this task after having experienced the initial failure?

Before they were given the second task, half of the students were asked to describe their mother in a few sentences, whereas the other students were asked to describe themselves. The researchers predicted that subtly getting the students to think about their moms would be particularly motivating for Asian-American students, but not for the European-American students. Why? As will be discussed in Chapter 3, social psychological research has shown that Asian Americans are more likely than European Americans to have an interdependent model of self, one in which motivation and action stem in large part from the influences of close others, especially of one’s mother. Although European Americans also certainly are influenced by close others such as their mother, they tend to have a relatively

culture A system of enduring meanings, beliefs, values, assumptions, institutions, and practices shared by a large group of people and transmitted from one generation to the next.

cross-cultural research Research designed to compare and contrast people of different cultures.

multicultural research Research designed to examine racial and ethnic groups within cultures.

80220_ch01_ptg01_hr_002-023.indd 18 27/01/16 11:19 AM

19 Social Psychology Today: What Is Trending Now?

independent model of self, one emphasizing personal preferences and goals. On the basis of this cultural difference, Fu and Markus reasoned that thinking about their moms would be more motivating for Asian-American than for European-American students as they were about to try the task a second time. (As one example of this kind of reaction, see the quote by Amy Chua in the margin.)

As can be seen in d Figure 1.2, the results supported this pre- diction. When made to think about themselves, European-American students worked on the task slightly more than the Asian-American students, but this difference was not significant. When made to think about their mothers, in contrast, the Asian-American students worked harder than the European-American students. More gener- ally, results such as these demonstrate that recognizing cultural dif- ferences can be vital in understanding what motivates individuals most effectively.

In this text, we describe studies conducted in dozens of coun- tries, representing every populated continent on earth. As our knowledge expands, we should be able to see much more clearly both the behavioral differences among cultures and the similarities we all share. But it is important to note that even within a particular society, people are often treated differently as a function of social categories such as gender, race, physical appearance, and economic class. They may be raised differently by their parents, confronted with different expectations by teachers, exposed to different types of advertising and marketing, and offered different kinds of jobs. In a sense, then, even people within the same town or region may develop and live in distinct subcultures, and these differences can have profound effects on people’s lives.

Some social psychology textbooks devote a separate chapter to culture or to culture and gender. We chose not to do so. Because we believe that cultural influences are inherent in all aspects of social psychology, we chose instead to integrate discussions of the role of culture and gender in every chapter of the textbook.

j Behavioral Economics, Political and Moral Issues, and Other Interdisciplinary Approaches

A rapidly growing number of social psychologists today are asking questions and using methodologies that cross traditional academic boundaries. We’ve already discussed intersections of social psychology with neuroscience, evolutionary the- ory, and cultural psychology. Other topics are beginning to trend in increasing numbers.

For example, one relatively new area of study that has received a great deal of attention is known as behavioral economics. This subfield focuses on how psychology—particularly social and cognitive psychology—relates to economic decision making. Behavioral economics research has revealed that the tradi- tional economic models were inadequate because they failed to account for the powerful—and often seemingly irrational—role that psychological factors have on people’s economic behavior. For example, Jiang Jiang and others (2015) con- ducted a series of experiments with high school students in China in which some students were made to either experience a rejection by a peer while playing a sim- ple game or to recall a time when they were rejected by peers. Compared to other

“[My daughter Lulu] had looked at a question and had drawn a blank. Lulu said, ‘Then I heard your annoying voice in my head, saying, ‘Keep thinking! I know you can do this”—and the answer just came to me.’”

— Amy Chua, about her influence on her daughter. Time, 1/20/2011.

N um

be r

of P

ro bl

em s

A tte

m pt

Condition

Mother Self 3

Asian American European American

d F I g U R E 1 . 2

Motivated by Mom? A Cultural Difference

Asian-American and European-American high- school students attempted a difficult letter- unscrambling task after having previously done poorly on it. The students worked equally hard on the task if they first were asked to think about themselves (the two bars on the right), but the Asian-American students worked harder than the European-American if they were first asked to think about their mothers (the two bars on the left). Based on Fu & Markus, 2014.

behavioral economics An interdisciplinary subfield that focuses on how psychology—particularly social and cognitive psychology— relates to economic decision making.

80220_ch01_ptg01_hr_002-023.indd 19 27/01/16 11:19 AM

20 Chapter 1 What Is Social Psychology?

students who were not made to experience or recall a peer rejection, those who did became materialistic, such as by answering the question, “What makes me happy?” by emphasizing material things such as new clothes and money. This is but one example of how subtle social psychological factors seemingly unrelated to an economic-related judgment can exert significant influence in ways that would not be predicted by economic analyses alone. The rise of behavioral economics has come with an increasing number of social psychologists hired by business schools and in the business world more generally.

Social psychologists are also increasingly involved in research intersecting other areas such as politics and moral philosophy. Jonathan Haidt (2012, 2017) is one prominent example of a social psychologist who recently has bridged both of these areas with social psychology, addressing questions such as whether we can better understand the political divide between liberals and conserva- tives by recognizing the different conceptions of morality that underlie political attitudes. The surge of interest in questions like this has brought together an exciting and productive mix of social psychology, political science, philosophy, and neuroscience.

Similarly, social psychologists are collaborating in greater numbers with researchers in environmental studies, public health, and related areas to address issues as varied as how to get people to conserve energy, adopt healthier hab- its, and avoid skin cancer, or to study how even simple alterations to the physi- cal surroundings around primary schools, such as providing some green space, can improve students’ education (Dadvand et al., 2015; Hamilton & Lobel, 2015; R. G. Jones, 2015; Wilson & Albarracín, 2015).

j The Social Brain and Body We are, of course, biological organisms, and it is clear that our brains and bodies influence, and are influenced by, our social experiences. This interaction between the physical and the social is the focus of more research than ever before in the field, and examples can be found throughout the textbook, such as in studies demonstrating how being the target of racial discrimination can affect individuals’ physical health, and research examining the role of hormones and neurotransmit- ters in human aggression (de Almeida et al., 2015; Dover et al., 2015; Gibbons et al., 2014.

A particularly exciting recent development is the emergence of the subfield of social neuroscience—the study of the relationship between neural and social processes. This intersection of social psychology and neuroscience is addressing a rapidly growing number of fascinating issues, such as how playing violent video games can affect brain activity and subsequent acts of aggression, or how dif- ferent patterns of activity in parts of the brain relate to how people are likely to perceive themselves or members of a different racial group (Amodio, 2014; Beer, 2015; Engelhardt et al., 2011).

Another interdisciplinary area of research attracting increasing interest among social psychologists today is known as embodied cognition, which focuses on the close links between our minds and the positioning, experiences, and actions of our bodies. According to this perspective, people’s perceptions and judgments reflect and can influence their bodily experiences. For example, recent studies have found that participants who were touching something rough (rough sand- paper) judged an interaction between two people as more rough and unfriendly than if they were touching something smooth, and that participants made to sit in an upright position experienced higher self-esteem and lower fear in response to a

“You carry [your friends and fam- ily] with you in your heart, your mind, your stomach, because you do not just live in a world but a world lives in you.”

— Frederick Buechner

social neuroscience The study of the relationship between neural and social processes.

embodied cognition An inter- disciplinary subfield that examines the close links between our minds and the positioning, experiences, and actions of our bodies.

80220_ch01_ptg01_hr_002-023.indd 20 27/01/16 11:19 AM

21 Social Psychology Today: What Is Trending Now?

stressful situation than did those sitting in a more slumped position (Meier et al., 2012; Nair et al., 2015; Schaefer et al., 2014).

We began this chapter discussing how the evolution of the human brain seems to have been linked closely to the social nature of our species. Research continues to find new evidence for how deeply rooted and basic this social nature is. Con- sider just a few examples from very recent studies:

d Having close friends and staying in contact with family members is associated with health benefits such as protecting against heart disease, infection, diabetes, and cancer, and with living longer and more actively (Cacioppo et al., 2015). One analy- sis reports that having a friend one sees on most days may affect one’s well-being as much as making an extra $100,000 a year does (Lieberman, 2013)!

d Experiencing a social rejection or loss is so painful that it produces activity in the same parts of the brain as when we feel physical pain. Being treated well and fairly by other people, on the other hand, activates parts of the brain associated with physical rewards such as desirable food and drink (Eisenberger, 2015).

d There is something medically very real about a “broken heart.” For example, a person is more than 20 times more likely than usual to suffer a heart attack within one day of the death of a loved one, and the effects of “broken heart syndrome” can endure for months (Mostofsky et al., 2012; Neil et al., 2015).

j New Technologies and the Online World The new wave of research on what we have called the social brain has been pos- sible only through advances in technology. Researchers today can see images of the brain at work through noninvasive procedures, enabling social psychologists to address questions they never could have attempted before. Social psychologists are now using techniques such as positron emission tomography (PET), event- related potential (ERP), transcranial magnetic stimulation (TMS), and functional magnetic resonance imaging (fMRI) to study the interplay of the brain and discrete

Advances in technology enable social psychologists to extend their research in exciting new directions, such as by using functional magnetic resonance imaging (fMRI) and Magnetoencephalograpy (MEG) to study activity in the brain in response to various thoughts or stimuli.

Sc ie

nc e

Ph ot

o Li

br ar

y/ Co

rb is

80220_ch01_ptg01_hr_002-023.indd 21 27/01/16 11:19 AM

22 Chapter 1 What Is Social Psychology?

thoughts, feelings, and behaviors. So- cial psychology research today benefits from other technological advances as well, such as new and better techniques to measure hormone levels and to code people’s everyday speech quickly into quantifiable units. Some researchers are using virtual reality technology to ex- amine a number of social psychological questions. The University of California at Santa Barbara’s Research Center for Vir- tual Environments and Stanford Univer- sity’s Virtual Human Interaction Lab are among the new labs conducting fascinat- ing high-tech research on issues such as conformity, group dynamics, aggression, altruism, social support, and eyewit- ness testimony (e.g., Bailey & Bailenson,

2015; Blascovich, 2014). Because participants in these experiments are immersed in a virtual reality that the experimenters create for them, the researchers can test questions that would be impractical, impossible, or unethical without this technology.

Awesome is an overused word, but it surely describes the revolution that is taking place in how we access information and communicate with each other. The waves of this revolution have carried social psychology research along with it. Social psychologists around the world can now not only communicate and col- laborate much more easily but can also gain access to research participants from populations that would otherwise never have been available. These developments have sparked the field’s internationalization, perhaps its most exciting course in its second century.

Online communication not only facilitates research but is also itself becom- ing a provocative topic of study. As more people interact with each other through social networking sites, online dating services, and communications services such as Skype or FaceTime, there is growing interest in studying how attraction, preju- dice, group dynamics, and a host of other social psychological phenomena unfold online versus offline. Other important, and potentially troubling, questions raised by our increasingly online lives are the subject of new research, such as: What factors contribute to or protect against cyberbullying? Can too much time on so- cial networking sites lead to depression or loneliness? Does the habit of frequent texting or checking to see who has commented on one’s most recent shared photo lead to attentional and social problems?

We would be presumptuous, and probably naive, to try to predict how new communication and new technologies will influence the ways that people will interact in the coming years, but it probably is safe to predict that their influence will be great. As more people fall in love online, or fall into social isolation, or react with violence to the loss of individual privacy, social psychology will explore these issues. We expect that some of the students reading this textbook today will be among those explorers in the years to come.

Together, and apart. Even when interacting in a group, many of us today are also pulled away by our individual phones, electronic games, laptops, tablets, and so on. How new technologies and living so much of our lives online affect human interaction is becoming increasingly important to social psychology.

Jo e

Ra ed

le /G

et ty

Im ag

es N

ew s/

G et

ty Im

ag es

80220_ch01_ptg01_hr_002-023.indd 22 27/01/16 11:19 AM

23 Review

1. Social psychology is the scientific study of how individu- als think, feel, and behave in a social context.

2. Women in one study achieved greater academic success at an engineering school if they received information early in their first year designed to reduce their concerns about fitting in and belonging in their male-dominated majors.

3. Early research by Triplett and Ringelmann established an enduring topic in social psychology: how the presence of others affects an individual’s performance.

4. Social psychology began to flourish as the world needed explanations for the tragic questions and challenges that World War II raised.

5. Stanley Milgram’s experiments in the 1960s demon- strated individuals’ vulnerability to the destructive com- mands of authority.

6. Individuals sometimes are faced with a conflict between two motivations that can affect cognitive processes: want- ing to be right and wanting to feel good about oneself.

7. A great deal of recent social psychological research has explored the automatic versus controllable nature of a number of processes, such as stereotyping.

8. Subfields that have been growing rapidly in recent years include: behavioral genetics, which examines the effects of genes on behavior; evolutionary psychology, which applies the principles of evolution to understand contemporary human behavior; behavioral economics, which studies how psychology relates to economic de- cision making; and social neuroscience, which involves the study of the relationship between neural and social processes.

9. Increasing numbers of social psychologists are evalu- ating the universal generality or cultural specificity of their theories and findings through cross-cultural research.

10. Online communication has fostered collaboration among researchers around the world, enabled research- ers to study participants from diverse populations, and inspired researchers to investigate whether various so- cial psychological phenomena are similar or different online versus offline.

Top 10 Key Points in Chapter 1

behavioral economics (19) behavioral genetics (17) cross-cultural research (18) culture (18)

embodied cognition (20) evolutionary psychology (17) interactionist perspective (14) multicultural research (18)

social cognition (16) social neuroscience (20) social psychology (6)

Key Terms

Review

80220_ch01_ptg01_hr_002-023.indd 23 27/01/16 11:19 AM

Doing Social Psychology Research

This chapter examines how social psychologists do their research. We begin by asking, “Why should you learn about research methods?” We answer this question by discussing how learning about research methods can benefit you both in this course and beyond. Then we consider how researchers come up with and develop ideas and begin the research process. Next, we provide an overview of the research designs that social psychologists use to test their ideas. Finally, we turn to important questions about ethics and values, including new controversies and practices, in social psychology.

80220_ch02_ptg01_hr_024-051.indd 24 30/01/16 10:43 AM

Doing Social Psychology Research

2 Why Should You Learn About Research Methods? | 27 Developing Ideas: Beginning the Research Process | 27

Getting Ideas and Finding Out What’s Been Done

Hypotheses and Theories

Basic and Applied Research

Refining Ideas: Defining and Measuring Social Psychological Variables | 29

From the Abstract to the Specific: Conceptual Variables and Operational Definitions

Measuring Variables: Using Self-Reports, Observations, and Technology

Testing Ideas: Research Designs | 33

Descriptive Research: Discovering Trends and Tendencies

Correlational Research: Looking for Associations

Experiments: Looking for Cause and Effect

Meta-Analysis: Combining Results Across Studies

Culture and Research Methods

Ethics and Values in Social Psychology | 46

Institutional Review Boards and Informed Consent

Debriefing

Ethics and Consent Online

Values and Science: Ongoing Debates and New Controversies

Review

25Co ur

te sy

o f A

lin a

P. F

ei n

80220_ch02_ptg01_hr_024-051.indd 25 30/01/16 10:43 AM

26 Chapter 2 Doing Social Psychology Research

You’re starting a new semester or quarter at school, and you’re just begin- ning to settle into a new schedule and routine. You’re looking forward to your new courses. In general, it’s an exciting time. But there’s one major catch: As you spend more and more time with your new classmates and new responsibilities, you’re leav- ing someone behind. It could be a boyfriend or girlfriend, a spouse, or a close friend— someone who is not involved in what you are doing now. You may now live far apart from each other or your new commitments in school may be keeping you apart from each other much more than you’d like. The romantic in you says, “Together forever.” Or at least, “No problem.” But the realist in you worries a bit. Will your love or friendship be the same? Can it survive the long distance, or the new demands on your time, or the new people in your respective environments? Your friends or family may have advice to offer in this situation. Some might smile and reassure you: “Don’t worry. Remember what they say, ‘Absence makes the heart grow fonder.’ This will only strengthen your relationship.” Others might call you aside and whisper, “Don’t listen to them. Everybody knows, ‘Out of sight, out of mind.’ You’d better be careful.”

Taking your mind off this problem, you begin to work on a class project. You have the option of working alone or as part of a group. Which should you do? You consult the wisdom of common sense. Maybe you should work in a group. After all, everyone knows that “two heads are better than one.” As some members of your group begin to miss meetings and shirk responsibilities, though, you remem- ber that “too many cooks spoil the broth.” Will you regret having been so quick to decide to join this group? After all, haven’t you been taught to “look before you leap”? Then again, if you had waited and missed the chance to join the group, you might have regretted your inaction, recalling that “he who hesitates is lost.”

Questions about the course of relationships, the efficiency of working in groups, and the regret from action versus inaction are social psychological questions. And because we all are interested in predicting and understanding people’s behaviors and their thoughts and feelings about each other, we all have our own opinions and intuitions about social psychological matters. If the discipline of social psychology were built on the personal experiences, observations, and intuitions of everyone who is interested in social psychological questions, it would be full of interesting theories and ideas, but it would also be a morass of contradictions, ambiguities, and relativism. Instead, social psychology is built on the scientific method.

Scientific? It’s easy to see how chemistry is scientific. When you mix two spe- cific compounds in the lab, you can predict exactly what will happen. The com- pounds will act the same way every time you mix them if the general conditions in the lab are the same. But what happens when you mix together two chemists, or any two people, in a social context? Sometimes you get great chemistry between them; other times you get apathy or even repulsion. How, then, can social behav- ior, which seems so variable, be studied scientifically?

To many of us in the field, that’s the great excitement and challenge of social psychology—the fact that it is so dynamic and diverse. Furthermore, in spite of these characteristics, social psychology can, and should, be studied according to scientific principles. Social psychologists develop specific, quantifiable hypotheses that can be tested empirically. If these hypotheses are wrong, they can be proven wrong. In addition, social scientists report the details of how they conduct their tests so that others can try to replicate their findings. They integrate evidence from across time and place. And slowly but steadily they build a consistent and ever more precise understanding of human nature. How social psychologists in- vestigate social psychological questions scientifically is the focus of this chapter. Before we explain the methodology they use, we first explain why it’s important and interesting for you to learn about these matters.

“The most exciting phrase to hear in science, the one that heralds new discoveries, is not ‘Eureka!’ (I found it!) but ‘That’s funny. . . .’”

—Isaac Asimov

80220_ch02_ptg01_hr_024-051.indd 26 30/01/16 10:43 AM

27 Developing Ideas: Beginning the Research Process

Why Should You Learn About Research Methods?

We are bombarded with information in our everyday lives, such as in the countless advertisements designed to persuade us to buy particular products or adopt particular opinions or attitudes. Learning the methods used in social psychology research can help students become more sophisticated consumers of this information.

Vi ct

or F

ra ile

/C or

bi s

N ew

s/ Co

rb is

One important benefit for learning about research methods is that it can make you a better, more sophisticated consumer of information. Training in research methods in psychology can improve your reasoning about real-life events and problems (Bensley et al., 2010; Daniel & Braasch, 2013; Lehman et al., 1988; VanderStoep & Shaughnessy, 1997). We are constantly bombarded with “facts” from the media, from sales pitches, and from other people. Much of this in- formation turns out to be wrong or, at best, oversimplified and misleading. We are told about the health benefits of eating certain kinds of food, the col- lege entrance exam score benefits of certain preparation courses, or the social status benefits of driving a certain kind of car or wearing a certain kind of shoe. To each of these pronouncements, we should say, “Prove it.” What is the evidence? What alternative explanations might there be?

For example, a commercial tells us that most doctors prefer a particular brand of aspirin. So should we buy this brand? Think about what it was compared with. Perhaps the doctors didn’t prefer that brand of aspirin more than other (and cheaper) brands of aspirin but rather were asked to compare that brand of as- pirin with several nonaspirin products for a particular problem. In that event, the doctors may have preferred any brand of aspirin more than nonaspirin products for that need. Thinking like a scientist while reading this text will foster a healthy sense of doubt about claims such as these. You will be in a better position to critically evaluate the information to which you’re exposed and separate fact from fiction.

More immediately, learning about re- search methods should help you better understand the research findings reported in the rest of this book, which will in turn help you on tests and in subsequent courses. If you simply read a list of social psychological findings without know- ing and understanding the evidence that researchers have produced to support the findings, you may discover later that the task of remembering which were the actual findings and which merely sound plausible can be difficult. Being able to understand, and therefore remember, the research evidence on which social psychological principles are based should provide you with a deeper com- prehension of the material.

Developing Ideas: Beginning the Research Process The research process involves coming up with ideas, refining them, testing them, and interpreting the meaning of the results obtained. This section describes the first stage of research—coming up with ideas. It also discusses the roles of hypotheses and theories and of basic and applied research.

80220_ch02_ptg01_hr_024-051.indd 27 30/01/16 10:43 AM

28 Chapter 2 Doing Social Psychology Research

j Getting Ideas and Finding Out What’s Been Done Every social psychology study begins with a question. And the questions come from everywhere. As discussed in Chapter 1, one of the first social psychology experiments published was triggered by the question “Why do bicyclists race faster in the presence of other bicyclists?” (Triplett, 1897–1898). Or consider a much more recent example—social psychologist Dylan Selterman has been con- ducting fascinating studies examining the connection between people’s dreams about romantic partners and their actual patterns of emotions and behaviors in personal relationships (Selterman et al., 2012, 2014). Where did this idea come from? It was inspired in part by the dreams some of his ex-girlfriends told him about!

Questions can come from a variety of sources, from something tragic, such as a controversial interracial shooting of an unarmed man; to something perplexing, such as why men are less likely to ask for help than are women; to something amusing, such as the lyrics of a country song suggesting that other patrons in a bar seem more attractive as closing time approaches (Correll et al., 2014; Pennebaker et al., 1979; Rosette et al., 2015).

Ideas also come from reading about research that has already been done. The most important research not only answers some pressing questions but raises new questions, inspiring additional research. The most reliable way to get ideas for new research, therefore, is to read about research already pub- lished. Even if you already have an idea, you’ll need to search the social psy- chological literature to find out what’s been published already. How do you find these published studies? Textbooks such as this one offer a good starting point. You also can find information about many research findings by searching the Internet, of course, but general searches online, such as through Google, can be wildly variable in the relevance, quality, and accuracy of the information pre- sented. Instead, scholars in the field rely on electronic databases of published research, typically available via college or university library systems. Some of these databases, such as PsycINFO, are specific to the psychology literature; others are more general. These databases allow one to instantly search massive numbers of published articles and books.

j Hypotheses and Theories An initial idea for research may be so vague that it amounts to little more than a hunch or an educated guess. Some ideas vanish with the break of day. But others can be shaped into a hypothesis—an explicit, testable prediction about the conditions under which an event will occur. Based on observation, exist- ing theory, or previous research findings, one might test a hypothesis such as “Teenage boys are more likely to be aggressive toward others if they have just played a violent video game for an hour than if they played a nonviolent video game for an hour.” This is a specific prediction, and it can be tested empirically. Formulating a hypothesis is a critical step toward planning and conducting research. It allows us to move from the realm of common sense to the rigors of the scientific method.

As hypotheses proliferate and data are collected to test the hypotheses, a more advanced step in the research process may take place: the proposal of a theory—an organized set of principles used to explain observed phenomena. Social psychologists aspire to do more than collect a list of findings. The goal

“Education is not the filling of a pail, but the lighting of a fire.”

—William Butler Yeats

“The currency of science is not truth, but doubt.”

—Dennis Overbye

hypothesis A testable prediction about the conditions under which an event will occur.

theory An organized set of principles used to explain observed phenomena.

80220_ch02_ptg01_hr_024-051.indd 28 30/01/16 10:43 AM

29 Refining Ideas: Defining and Measuring Social Psychological Variables

is to explain these findings, to articulate the connections between the variables that are studied, and to thereby predict and more completely understand our social worlds. All else being equal, the best theories are efficient and precise; encompass all of the relevant information; and lead to new hypotheses, further research, and better understanding. Good social psychological theories inspire subsequent research designed to test various aspects of the theories and the spe- cific hypotheses that are derived from them. Whether it truly is accurate or not, a theory has little worth if it cannot be tested.

A theory may make an important contribution to the field even if it turns out to be wrong. The research it inspires may prove more valuable than the theory itself, as the results shed light on new truths that might not have been discovered without the directions suggested by the theory. Indeed, the best theorists want their ideas to be debated and even doubted, to inspire others in the field to put their ideas to the test. The goal is for these theories to evolve, to become more and more accurate and complete.

j Basic and Applied Research Is testing a theory the purpose of research in social psychology? For some re- searchers, yes. Basic research seeks to increase our understanding of human be- havior and is often designed to test a specific hypothesis from a specific theory. Applied research focuses more specifically on making applications to the world and contributing to the solution of social problems.

Despite their differences, basic and applied research are closely connected in social psychology. Some researchers switch back and forth between the two— today basic, tomorrow applied. Some studies test a theory and examine a real- world phenomenon simultaneously. For instance, Carol Dweck and others have tested theories about the effects of people’s beliefs about human abilities while addressing important problems such as students dropping out of school or the underrepresentation of women in math and science (Good et al., 2012; Paunesku et al., 2015). As a pioneer in both basic and applied approaches, Kurt Lewin (1951) set the tone when he encouraged basic researchers to be concerned with complex social problems and urged applied researchers to recognize that “there is nothing so practical as a good theory.”

“Give people facts and you feed their minds for an hour. Awaken curiosity and they feed their own minds for a lifetime.”

—Ian Russell

“Close cooperation between theo- retical and applied psychology can be accomplished . . . if the theo- rist does not look toward applied problems with highbrow aversion or with a fear of social problems, and if the applied psychologist realizes that there is nothing so practical as a good theory.”

—Kurt Lewin

Refining Ideas: Defining and Measuring Social Psychological Variables To test their hypotheses, researchers always must decide how they will define and measure the variables in which they are interested. This is sometimes a straight- forward process. For example, if you are interested in comparing how quickly people run a 100-meter dash when alone and when racing against another per- son, you’re all set if you have a stopwatch and runners who can race alone or in pairs. Many other times, however, the process is less straightforward. For example, imagine you are interested in studying the effects of mood on altruistic (helpful) behavior. Sounds simple, right? But wait. You need to step back and ask yourself, “What do I mean by mood? How would I measure or manipulate it? What do I mean by altruistic behavior?” You will need to define these concepts, and there may be countless ways to do this. Which ones should you pick?

basic research Research whose goal is to increase the understanding of human behavior, often by testing hypotheses based on a theory.

applied research Research whose goal is to make applications to the world and contribute to the solution of social problems.

80220_ch02_ptg01_hr_024-051.indd 29 30/01/16 10:43 AM

30 Chapter 2 Doing Social Psychology Research

From this picture, we can guess that the boy sitting by himself on the playground is lonely, but how do researchers precisely define and measure conceptual variables such as loneliness? Researchers may use any of a number of approaches, such as asking people how they feel and observing their behavior.

m m

a Sh

er vi

ng to

n/ Fl

irt /C

or bi

operational definition The specific procedures for manipulating or measuring a conceptual variable.

an abstract conceptual variable such as love or group pressure and deciding how to operationally define it so as to manipulate or measure it.

Imagine, for example, wanting to conduct a study on the effects of alcohol intoxication on aggression. One of the conceptual variables might be whether or not participants are intoxicated. There are several ways of measuring this variable, most of which are relatively straightforward. For instance, one researcher might operationally define intoxication as when a participant has a blood alcohol level of .10 or more, whereas another might define it as when a participant says that he or she feels drunk. A second conceptual variable in this study would be aggres- sion. Measuring aggression in experiments is particularly difficult because of ethi- cal and practical issues—researchers can’t let participants in their studies attack each other. Researchers interested in measuring aggression are thus often forced to measure relatively unusual behaviors, such as administering shocks, blasts of noise, or even hot sauce to another person as part of a specific task.

Often there is no single best way to transform a variable from the abstract (conceptual) to the specific (operational). A great deal of trial and error may be involved. However, sometimes there are systematic, statistical ways of checking how valid various manipulations and measures are, and researchers spend a great deal of time fine-tuning their operational definitions to best capture the concep- tual variables they wish to study.

j From the Abstract to the Specific: Conceptual Variables and Operational Definitions

When a researcher first develops a hypothesis, the variables typically are in an abstract, general form. These are conceptual variables. Examples of conceptual variables include prejudice, con- formity, attraction, love, group pressure, and social anxiety. In order to test specific hypoth- eses, we must then transform these conceptual variables into variables that can be manipulated or measured in a study. The specific way in which a conceptual variable is manipulated or measured is called the operational definition of the variable. Part of the challenge and fun of de- signing research in social psychology is taking

This person would probably score low on the Rosenberg Self-Esteem Scale.

D ea

n Vi

et or

/T he

N ew

Y or

ke r C

ol le

ct io

n/ Th

e Ca

rt oo

n Ba

“Clemson here, how may I disappoint you?”

80220_ch02_ptg01_hr_024-051.indd 30 30/01/16 10:43 AM

31 Refining Ideas: Defining and Measuring Social Psychological Variables

Researchers evaluate the manipulation and measurement of variables in terms of their construct validity. Construct validity refers to the extent to which (1) the manipulations in an experiment really manipulate the conceptual variables they were designed to manipulate, and (2) the measures used in a study (experimental or otherwise) really measure the conceptual variables they were designed to measure.

j Measuring Variables: Using Self-Reports, Observations, and Technology

Social psychologists measure variables in many ways, but most can be placed into one of two categories: self-reports and observations. We discuss each of these methods in the next sec- tions, along with how advances in technology are enabling social psychologists to measure vari- ables in new ways.

Self-Reports Collecting self-reports—in which participants disclose their thoughts, feelings, desires, and actions—is a widely used measurement tech- nique in social psychology. Self-reports can consist of individual questions or sets of questions that together measure a single conceptual variable. For example, one popular self-report measure, the Rosenberg Self- Esteem Scale, consists of a set of questions that mea- sures individuals’ overall self-esteem. Respondents are asked the extent to which they agree with state- ments such as “I feel that I have a number of good qualities,” and “All in all, I am in- clined to feel that I’m a failure.” This scale, first developed by Morris Rosenberg in the 1960s, continues to be used today in a wide variety of settings in countries around the world because many researchers consider it to have good construct validity (Alessandri et al., 2015; Galanou et al., 2014; Westaway et al., 2015).

Self-reports give the researcher access to an individual’s beliefs and percep- tions. But self-reports are not always accurate and can be misleading. For exam- ple, the desire to look good to ourselves and others can influence how we respond. This is evident in the results of research using the bogus pipeline technique—a procedure in which participants are led to believe that their responses will be verified by an infallible lie detector. When participants believe their lies will be detected, they report facts about themselves more accurately and endorse socially unacceptable opinions more frequently. The bogus pipeline is, in fact, bogus; no such infallible device exists. But belief in its powers discourages people from lying (Brunell & Fisher, 2014; Grover & Miller, 2012; Jones & Sigall, 1971).

Self-reports are also affected by the way that questions are asked, such as how they are worded or in what order or context they are asked (Betts & Hartley, 2012; Schwarz & Oyserman, 2011). For instance, although “global warming” and “climate change” mean the same thing to most Americans, Jonathon Schuldt and others (2015) found that Republican (but not Democratic) respondents to a sur- vey in the United States were much less likely to indicate that they believe in the issue if it was referred to as “global warming” rather than “climate change.” In another study, a large majority (88%) of participants indicated that they thought condoms were effective in stopping AIDS when condoms were said to have a “95 percent success rate.” However, when condoms were said to have a “5 percent

construct validity The extent to which the measures used in a study measure the variables they were designed to measure and the manipulations in an experiment manipulate the variables they were designed to manipulate.

bogus pipeline technique A procedure in which research participants are (falsely) led to believe that their responses will be verified by an infallible lie detector.

The challenge of measuring variables. This may not be the most precise way to measure height, but it’s pretty charming.

A P

Im ag

es /K

irs ty

W ig

gl es

w or

80220_ch02_ptg01_hr_024-051.indd 31 30/01/16 10:43 AM

32 Chapter 2 Doing Social Psychology Research

failure rate” (which is merely another way of saying the same thing), less than half (42%) of the participants indicated that they thought condoms were effective (Linville et al., 1992). More recently, Erika Hall and others (2015) found in one study that that white Americans estimated a man’s salary as significantly lower, and in another study felt more negatively toward a criminal suspect, if the man was iden- tified as “Black” rather than “African American” (see d Figure 2.1). There are theoretical explanations that can account for the seem- ingly irrational results reported in these studies, but the point here is that subtle factors can have significant effects on the attitudes and opinions that people report.

Indeed, even the exact same question can elicit very different responses depending on the context in which the question occurs. For example, all individuals contacted in one survey were asked how important the issue of skin cancer was in their lives, but this ques- tion was asked either before or after a series of questions about other health concerns. Even though the wording of the question was iden- tical, respondents rated skin cancer as significantly more important if the question was asked first than if it came after the other health questions (Rimal & Real, 2005). In another example of the impact of previous questions, Kimberly Rios Morrison and Adrienne Chung (2011) found that white college students indicated significantly less support for multiculturalism if they had earlier marked their race/ ethnicity as “white” on a questionnaire than if the questionnaire used the term “European American” instead.

Another reason self-reports can be inaccurate is that they often ask participants to report on thoughts or behaviors from the past, and people’s memories are very prone to error, particularly if how they feel now about things is different from how they felt in the past. To minimize this problem, psychologists have developed ways to reduce the time that elapses between an actual experience and the person’s report of it. For example, some use interval-contingent self-reports, in which respondents report their experiences at regular intervals, usu- ally once a day. Researchers may also collect signal-contingent self- reports. Here, respondents report their experiences as soon as possible after being signaled to do so, usually by means of a text message or a special app. Finally, some researchers collect event-contingent self- reports, in which respondents report on a designated set of events as soon as possible after such events have occurred.

Observations Self-reports are but one tool social psychologists use to measure variables. Researchers also observe people’s actions. Sometimes these observations are very simple, as when a researcher notes which of two items a per- son selects. At other times, however, the observations are more elaborate, as when judging whether someone is acting warmly or coldly toward another person, and require that interrater reliability be established. Interrater reliability refers to the level of agreement among multiple observers of the same behavior. Only when different observers agree can the data be trusted.

The advantage of observational methods is that they avoid our sometimes- faulty recollections and distorted interpretations of our own behavior. Actions can speak louder than words. Of course, if individuals know they are being observed, their behaviors, like their self-reports, may be biased by the desire to present

d F I G U R e 2 . 1

“Black” vs. “African American”: What’s in a Name?

Although most Americans think of “Black” and “African American” as the same thing when identifying the race of someone, a 2015 experiment found that white American participants estimated someone’s salary as significantly lower (graph on top), and had more negative reactions to a suspected criminal (graph on bottom), if the individual was identified as Black rather than African American. Based on Hall et al., 2015.

Es ti

m at

ed A

n n

u al

S al

ar y

Black African American 10,000

20,000

30,000

40,000

Em o

ti o

n t

o w

ar d

s u

sp ec

t (h

ig h

er n

u m

b er

s in

d ic

at e

m o

re w

ar m

th )

Black African American

1.5

0.5

2.5

interrater reliability The degree to which different observers agree on their observations.

80220_ch02_ptg01_hr_024-051.indd 32 30/01/16 10:43 AM

33 Testing Ideas: Research Designs

themselves in a favorable light. Therefore, researchers sometimes make observa- tions much more subtly. For example, in experiments concerning interracial in- teractions, researchers may record participants’ eye contact and seating distance to demonstrate biases that would not be revealed using more overt measures (Goff, Steele, & Davies, 2008; Todd et al., 2011).

Technology Social psychologists use more than merely their eyes and ears to observe their subjects, of course. Various kinds of technology are used to measure cognitive and physi- ological responses such as reaction time or heart rate, levels of particular hormones, and sexual arousal. Eye-tracking technol- ogy is used to measure exactly where and for how long partici- pants look at particular parts of a stimulus, such as a face or a video of an interaction.

Most recently, social psychologists have begun opening a window into the live human brain—fortunately, without having to lift a scalpel. Brain-imaging technologies take and combine thousands of images of the brain in action. These images can show researchers what parts of the brain seem to “light up”—or show increased activity—in response to a particular stimulus or situation. For example, although par- ticipants in a study may show no signs of intergroup bias on their self-reports or through easily observable behavior in the lab, their patterns of brain activity may reveal very differ- ent emotional reactions to a picture of someone on the basis of their perceived group membership (Amodio, 2014; Cikara & Van Bavel, 2014).

Observational research can reveal some fascinating—and sometimes disturbing!—insights into social behavior.

“Just as we suspected—they’re beginning to form a boy band.”

Social psychologists use a variety of methods to test their research hypotheses and theories. Although methods vary, the field generally emphasizes objective, systematic, and quantifiable approaches. Social psychologists do not simply seek out evidence that supports their ideas; rather, they test their ideas in ways that could very clearly prove them wrong. We can divide these types of tests into three categories: descriptive, correlational, and experimental.

j Descriptive Research: Discovering Trends and Tendencies One obvious way of testing ideas about people is simply to record how frequently or how typically people think, feel, or behave in particular ways. The goal of descriptive research in social psychology is, as the term implies, to describe peo- ple and their thoughts, feelings, and behaviors. This method can test questions such as: Do most people support capital punishment? What percentage of people who encounter a person lying on the sidewalk would offer to help that person? What do men and women say are the things most likely to make them jealous of their partner? Particular methods of doing descriptive research include observing people, studying records of past events and behaviors, and surveying people. We discuss each of these methods in this section.

Testing Ideas: Research Designs

“You can observe a lot just by watching.”

—Yogi Berra

M at

th ew

D iff

ee /T

he N

ew Y

or ke

r C ol

le ct

io n/

Th e

Ca rt

oo n

Ba nk

80220_ch02_ptg01_hr_024-051.indd 33 30/01/16 10:43 AM

34 Chapter 2 Doing Social Psychology Research

Observational Studies We just discussed using observations as a way to measure and assess variables. Many social psychologists use these methods to test ideas involving descriptive research. For example, some researchers have gotten an unfiltered look at the frequency and severity of bullying among adolescents by spending time in playgrounds and school- yards carefully watching and taking notes on the children’s interactions, sometimes using hidden cameras and microphones (with the schools’ and parents’ consent) (Frey et al., 2009; Hawkins et al., 2001).

One particularly interesting study involved National Basketball Association (NBA) basketball games. Graeme Haynes and Thomas Gilovich (2010) wanted to see how often professional players missed free-throw attempts in cases when they were awarded the shots because of an obviously bad foul call against the other team. The researchers were testing the idea that when players benefit from an obviously bad call by the referee, they subtly, even unknowingly, may be troubled by the sense that they were awarded something they didn’t deserve, lead- ing them to be more likely to miss their free-throw attempt.

The researchers couldn’t run an experiment in which they induced refs into making bad calls during actual games, so they did the next best thing: they watched tapes of NBA games. A lot of them. One hundred and two games, to be precise. Obviously bad foul calls were identified, and whether the subsequent free-throw shooter made or missed his ensuing shots was noted.

Consistent with the adage used by some athletes, “The ball don’t lie,” the results indicated that justice was indeed served, specifically on the first of the two free throws players took after the bad call. As can be seen in d Figure 2.2, the percentage of shots made on the first shot after the bad call was significantly lower than normal. This effect disappeared, however, by the time of the second shot.

Archival Studies Archival research involves examining existing records of past events and behaviors, such as newspaper articles, medical records, diaries, sports statistics, personal ads, crime statistics, or hits on a website. A major advantage of ar- chival measures is that because the researchers are observing behavior secondhand, they can be sure that they did not influence the behavior by their presence. A limita- tion of this approach is that available records are not always complete or sufficiently detailed, and they may have been collected in a nonsystematic manner.

Archival measures are particularly valuable for examining cultural and his- torical trends. In Chapter 11 on Aggression, for example, we report a number of trends concerning how the rates of violent crimes have changed in recent years. These data come from archival records, such as the records of police stations, the Federal Bureau of Investigation (FBI), and the United Nations. Recent examples of archival research include studies that analyzed the content of TV commercials or the wording of job advertisements to examine if they were biased as a function of gender stereotypes (Gaucher et al., 2011; Kay & Furnham, 2013).

Surveys It seems that nobody in politics today sneezes without first conducting an opinion poll. Surveys have become increasingly popular in recent years, and they are conducted on everything from politics to attitudes about social issues to the percentages of people who think the best cure for hiccups is to drink a glass of water or hold your breath. (OK, we’ll tell you: According to a May 2012 poll on Yahoo.com with almost 60,000 responses, 52% say drink water, 37% say hold your breath—and 12% say have someone scare you.) Conducting surveys

Are professional basketball players more likely to miss a free throw if they were awarded the shot because of an obviously bad call by the referee? A 2010 study reported here attempted to answer that question.

A P

Im ag

es /C

hr is

to ph

er S

za go

80220_ch02_ptg01_hr_024-051.indd 34 30/01/16 10:43 AM

35 Testing Ideas: Research Designs

involves asking people questions about their at- titudes, beliefs, and behaviors. Surveys can be conducted in person, over the phone, by mail, or via the Internet. Many social psychological questions can be addressed only with surveys because they involve variables that are impos- sible or unethical to observe directly or manipu- late, such as people’s sexual behaviors or their optimism about the future.

Although anyone can conduct a survey (and sometimes it seems that everyone does), there is a science to designing, conducting, and in- terpreting the results of surveys properly and to avoid the kinds of problems we described earlier in this chapter about how wording and question order can bias self-reports.

One of the most important issues that sur- vey researchers face is how to select the people who will take part in the survey. The research- ers first must identify the population in which they are interested. Is this survey supposed to tell us about the attitudes of North Americans in general, shoppers at Wal-Mart, or students in an Introduction to Social Psychology course at Uni- versity X, for example? From this general popula- tion, the researchers select a subset, or sample, of individuals. For a survey to be accurate, the sample must be similar to, or representative of, the population on important characteristics such as age, sex, race, income, education, and cultural background. The best way to achieve this repre- sentativeness is to use random sampling, a method of selection in which every- one in a population has an equal chance of being selected for the sample. Survey researchers use randomizing procedures, such as randomly distributed numbers generated by computers, to decide how to select individuals for their samples.

To see the importance of random sampling, consider a pair of U.S. presidential elections (Rosnow & Rosenthal, 1993). Just before the 1936 election, a magazine called the Literary Digest predicted that Alfred Landon, the Republican governor of Kansas, would win by 14 percentage points over Franklin Roosevelt. The Digest based its prediction on a survey of more than 2 million Americans. In fact, though, Landon lost the election by 24 percentage points. The magazine, which had been in financial trouble before the election, declared bankruptcy soon after.

Twenty years later, the Gallup survey’s prediction of Dwight Eisenhower’s victory was almost perfect—it was off by less than 2%. The size of its sample? Only about 8,000. How could the 1936 survey, with its much larger sample of 2 million people, be so wrong and the 1956 survey be so right? The answer is that the 1936 sample was not randomly selected. The Digest contacted people through sources such as phone books and club membership lists. In 1936, many people could not afford to have phones or belong to clubs. The people in the sample, therefore, tended to be wealthier than much of the population, and wealthier people preferred Landon. In 1956, by contrast, Gallup pollsters randomly selected election districts throughout the country and then randomly selected households within those districts. Today,

d F I G U R e 2 . 2

Observing Basketball: The Ball Don’t Lie

This graph compares the free-throw success rates for (a) the average NBA player that season (left bar), the season average for the players included in the study (second bar), (c) the first shots after a bad call (third bar), and (d) the second shots after a bad call (right-most bar). As you can see, players shot much worse immediately after benefiting from a bad call. From Haynes & Gilovich, 2010.

NBA Mean %

Mean % for these Players

% on First Shot After Bad Call

% on Second Shot After Bad Call

random sampling A method of selecting participants for a study so that everyone in a population has an equal chance of being in the study.

80220_ch02_ptg01_hr_024-051.indd 35 30/01/16 10:43 AM

36 Chapter 2 Doing Social Psychology Research

because of improved sampling procedures, surveys conducted on little more than 1,000 Americans can be used to make accurate predictions about the entire U.S. population.

correlational research Research designed to measure the association between variables that are not manipulated by the researcher.

correlation coefficient A statistical measure of the strength and direction of the association between two variables.

In the 1948 U.S. presidential election, pollsters nationwide predicted that Thomas Dewey would defeat Harry Truman by a wide margin. As Truman basked in his victory, pollsters realized that their predictions were based on nonrandom samples of voters. Random sampling would have led to much more accurate predictions.

ve re

tt C

ol le

ct io

n H

is to

ric al

/A la

m y

St oc

k Ph

ot o

j Correlational Research: Looking for Associations

Although there is much to learn from descrip- tive research, social psychologists typically want to know more. Most research hypotheses in so- cial psychology concern the relationship between variables. For example, is there a relationship be- tween people’s gender and their willingness to ask for help from others or between how physically attractive people are and how much money they make?

One way to test such hypotheses is with cor- relational research. Like descriptive research, correlational research can be conducted using ob- servational, archival, or survey methods. Unlike

descriptive research, however, correlational approaches measure the relationship between different variables. The extent to which variables relate to each other, or correlate, can suggest how similar or distinct two different measures are (for example, how related are people’s self-esteem and popularity) and how well one variable can be used to predict another (for example, how well we can predict academic success in college from college entrance exam scores). It is important to note that researchers doing correlational research typically do not manipulate the variables they study; they simply measure them.

One interesting correlational study was conducted recently by a team of re- searchers who got access to 826 million tweets (!) made available to them by Twit- ter. The researchers looked at the relationship between the language that people in more than 1,300 counties in the United States used on Twitter and measures of health (from public records) in those counties. They found, for example, that communities in which people tended to tweet using angry language also tended to have greater rates of heart-disease mortality (Eichstaedt et al., 2015).

Correlation Coefficient When researchers examine the relationship between variables that vary in quantity (such as height or degree of self-esteem), they can measure the strength and direction of the relationship between the variables and calculate a statistic called a correlation coefficient. Correlation coefficients can range from 11.0 to 21.0. The absolute value of the number (the number itself, without the positive or negative sign) indicates how strongly the two variables are associated. The larger the absolute value of the number, the stronger the associa- tion between the two variables, and thus the better either of the variables is as a predictor of the other. Whether the coefficient is positive or negative indicates the direction of the relationship. A positive correlation coefficient indicates that as one variable increases, so does the other.

For example, college entrance exam scores correlate positively with grades. The positive direction of this relationship indicates higher entrance exam scores are associated with higher grades and lower entrance exam scores are associated with lower grades. This correlation is not perfect; some people with high entrance

80220_ch02_ptg01_hr_024-051.indd 36 30/01/16 10:43 AM

37 Testing Ideas: Research Designs

exam scores have poor grades and vice versa. Therefore, the correlation is less than 11.0, but it is greater than 0 because there is some association between the two. A negative coefficient indicates that the two variables go in opposite directions: As one goes up, the other tends to go down. For example, number of classes missed and GPA are likely to be negatively cor- related. And a correlation close to 0 indicates that there is no consistent rela- tionship at all. These three types of patterns are illustrated in d Figure 2.3. Because few variables are perfectly related to each other, most correlation coefficients do not approach 11.0 or 21.0 but have more moderate values, such as 1.39 or 2.57.

Some correlational studies involve a variable that does not vary in quan- tity, such as race, gender, political affiliation, or whether one’s favorite food is Italian, Mexican, or Thai. In this case, researchers cannot compute a typi- cal correlation coefficient but instead use different kinds of statistical analy- sis. The same point applies, though, as the researchers can determine if there is a relationship between the two variables.

Advantages and Disadvantages of Correlational Research Cor- relational research has many advantages. It can study the associations of naturally occurring variables that cannot be manipulated or induced—such as ethnicity, age, and income. It can examine phenomena that would be difficult or unethical to create for research purposes, such as love, hate, and abuse. And it offers researchers a great deal of freedom in where variables are measured. Participants can be brought into a laboratory specially constructed for research purposes or they can be approached in a real-world setting (often called “the field”) such as a shopping mall or airport.

Despite these advantages, however, correlational research has one very seri- ous disadvantage. And here it is in bold letters: Correlation is not causation.

Similarity is correlated with attraction—the more similar two people are (such as in their attitudes and personalities), the more attractive they are likely to find each other. But a correlation cannot identify the cause of this attraction. Chapter 9 discusses correlational and experimental research on the role of similarity in the attraction process.

M oo

db oa

rd /C

or bi

d F I G U R e 2 . 3

Correlations: Positive, Negative, and None

Correlations reveal a systematic association between two variables. Positive correlations indicate that variables are in sync: Increases in one variable are associated with increases in the other; decreases, with decreases. Negative correlations indicate that variables go in opposite directions: Increases in one variable are associated with decreases in the other. When two variables are not systematically associated, there is no correlation. © Cengage Learning®

Cold

Number of people who have the hiccups

Few Cold

Hot

Number of people who wear sweaters

Few

Hot

D ai

ly te

m pe

ra tu

D ai

ly te

m pe

ra tu

D ai

ly te

m pe

ra tu

Number of people who purchase cold drinks

Positive Correlation Negative Correlation No Correlation

Few Many Many Many Cold

Hot

80220_ch02_ptg01_hr_024-051.indd 37 30/01/16 10:43 AM

38 Chapter 2 Doing Social Psychology Research

In other words, a correlation cannot demonstrate a cause-and- effect relationship. Instead of revealing a specific causal pathway from one variable, A, to another variable, B, a correlation between variables A and B contains within it three possible causal effects: A could cause B; B could cause A; or a third variable, C, could cause both A and B. For example, imagine learning that the number of hours per night one sleeps is negatively correlated with the number of colds one gets. This means that as the amount of sleep increases, colds decrease in frequency; conversely as sleep decreases, colds become more frequent. One reasonable explanation for this relation- ship is that lack of sleep (variable A) causes people to become more vulnerable to colds (variable B). Another reasonable explanation, however, is that people who have colds can’t sleep well, and so colds (variable B) cause lack of sleep (variable A). A third reasonable ex- planation is that some other variable (C) causes both lack of sleep and greater frequency of colds. This third variable could be stress. Indeed, stress has many effects on people, as will be discussed in Chapter 14 on Health. d Figure 2.4 describes another correlation that can be explained in many ways—the correlation between play- ing violent video games and aggression.

We can guarantee you this: There will be many, many times in your life when you’ll encounter reports in the media that suggest cause-and-effect relationships based on correlational research. Even the most respectable news sources are guilty of this repeatedly. If you look for it, you can find numerous examples on a weekly basis. One of the great benefits of learning and gaining experience with the material in this chapter is that you can see the flaws in media reports such as these and become less likely to be taken in by them. Correlation is not causation.

Do we learn nothing, then, from correlations? To say that would be to take caution too far. Correlations tell researchers about the strength and direction of relationships between variables, thus help-

ing them understand these variables better and allowing them to use one variable to predict the other. Correlations can be extremely useful in developing new hy- potheses to guide future research. And by gathering large sets of correlations and using complicated statistical techniques to crunch the data, we can develop highly accurate predictions of future events.

j experiments: Looking for Cause and effect Social psychologists often do want to examine cause-and-effect relationships. Although it is informative to know, for example, that playing a lot of violent video games is correlated with violent behavior in real life, the inevitable next question is whether playing these video games causes an increase in violent behavior. If we want to examine cause-and-effect relationships, we need to con- duct an experiment. Experiments are the most popular method of testing ideas in social psychology, and they can range from the very simple to the incredibly elaborate. All of them, however, share two essential characteristics.

1. The researcher has control over the experimental procedures, manipulating the variables of interest while ensuring uniformity elsewhere. In other words, all par- ticipants in the research are treated in exactly the same manner—except for the specific differences the experimenter wants to create.

d F I G U R e 2 . 4

explaining Correlations: Three Possibilities

A positive correlation between how much children play violent video games and how aggressively they behave could potentially be explained in each of the following ways:

1. Playing violent video games causes aggressive behavior.

2. Children who behave aggressively like to play a lot of violent video games.

3. Children who have family troubles, such as parents who are not very involved in the children’s development, tend both to play a lot of violent video games and to behave aggressively.

C (family troubles)

A (video games)

B (aggression)

A (video games)

experiment A form of research that can demonstrate causal relationships because (1) the experimenter has control over the events that occur and (2) participants are randomly assigned to conditions.

80220_ch02_ptg01_hr_024-051.indd 38 30/01/16 10:43 AM

39 Testing Ideas: Research Designs

2. Participants in the study are randomly as- signed to the different manipulations (called “conditions”) included in the experiment. If there are two conditions, who goes where may be determined by simply flipping a coin. If there are many conditions, a computer pro- gram may be used. But however it’s done, random assignment means that participants are not assigned to a condition on the basis of their personal or behavioral characteris- tics. Through random assignment, the ex- perimenter attempts to ensure a level playing field. On average, the participants randomly assigned to one condition are no different from those assigned to another condition. Differ- ences that appear between conditions after an experimental manipulation can therefore be attributed to the impact of that manipula- tion and not to any preexisting differences be- tween participants.

Because of experimenter control and ran- dom assignment of participants, an experi- ment is a powerful technique for examining cause and effect. Both characteristics serve the same goal: to eliminate the in- fluence of any factors other than the experimental manipulation. By ruling out alternative explanations for research results, we become more confident that we understand just what has, in fact, caused a certain outcome to occur. Table 2.1 summarizes the distinctions between correlational and experimental research.

Random Sampling Versus Random Assignment You may recall that we mentioned random sampling earlier, in connection with surveys. It’s important to remember the differences between random sampling and random assignment. Ran- dom sampling concerns how individuals are selected to be in a study. It is important for generalizing the results obtained from a sample to a broader population, and it is therefore very important for survey research. Random assignment concerns not who is selected to be in the study but rather how participants in the study are assigned to different conditions, as explained above. Random assignment is essential to experi- ments because it is necessary for determining cause-and-effect relationships; without it, there is always the possibility that any differences found between the conditions in a study were caused by preexisting differences among participants. Random sam- pling, in contrast, is not necessary for establishing causality. For that reason, and be- cause random sampling is difficult and expensive, very few experiments use random sampling. We consider the implications of this fact later in the chapter.

Laboratory and Field experiments Most experiments in social psychology are conducted in a laboratory setting, usually located in a college or university, so that the environment can be controlled and the participants carefully studied. Social psychology labs do not necessarily look like stereotypical laboratories with liquid bubbling in beakers or expensive equipment everywhere (although many social psy- chology labs are indeed very high-tech). They can resemble ordinary classrooms or even game rooms. The key point here is that the laboratory setting enables research- ers to have control over the setting, measure participants’ behaviors precisely, and keep conditions identical for participants.

random assignment A method of assigning participants to the various conditions of an experiment so that each participant in the experiment has an equal chance of being in any of the conditions.

m T a b l e 2 . 1

Correlations Versus experiments

Correlational Research experimental Research

What does it involve?

Measuring variables and the degree of association between them

Random assignment to conditions and control over the events that occur; determining the effects of manipulations of the independent variable(s) on changes in the dependent variable(s)

What is the biggest advantage of using this method?

Enables researchers to study naturally occurring variables, including variables that would be too difficult or unethical to manipulate

Enables researchers to determine cause-and-effect relationships—that is, whether the independent variable can cause a change in the dependent variable

80220_ch02_ptg01_hr_024-051.indd 39 30/01/16 10:43 AM

40 Chapter 2 Doing Social Psychology Research

Field research is conducted in real-world set- tings outside the laboratory. Researchers inter- ested in studying helping behavior, for example, might conduct an experiment in a public park. The advantage of field experiments is that peo- ple are more likely to act naturally in a natural setting than in a laboratory in which they know they are being studied. The disadvantage of field settings is that the experimenter often has less control and cannot ensure that the participants in the various conditions of the experiment will be exposed to the same things.

Because of the important role that experi- ments play in social psychology, let’s take a closer look at the elements of experiments by focusing on one example.

A Social Psychology experiment on Presidential Debates While watch- ing a debate between candidates for the presidency of the United States, one of the authors of this textbook wondered if the applause, laughter, and jeering of people in the audience at the debate might affect the judgments of the millions of potential voters watching on television. This led to a series of experiments, including one in which college students watched a tape of a 1984 presidential debate between Ronald Reagan and Walter Mondale (Fein et al., 2007). During that debate Reagan fired off a pair of one-liners that elicited a great deal of laughter and applause from the audi- ence. Political analysts have wondered whether those one-liners may have won the debate, and possibly the election, for Reagan. The one-liners comprised only seconds of a long debate concerning the most important issues of the day. Could these few seconds have made such a difference?

To study this issue, we had students watch 40 minutes of the debate under one of three conditions. One-third of the students saw the debate as it was, with- out any editing. One-third of the students saw the debate with the two one-liners and the ensuing audience reaction edited out. By comparing these two conditions, we could see whether the presence versus absence of this pair of jokes could make a large difference in people’s impressions of Reagan from the debate. However, there was also a third condition. One-third of the students saw the debate with the one-liners intact but the audience reaction edited out. That is, Reagan told his jokes, but there appeared to be no audience response, and the debate continued uninterrupted.

After watching the debate, the students judged the performance of the can- didates on a scale ranging from 0 (terrible) to 100 (excellent). As you can see from the first two bars in d Figure 2.5, the students who saw the entire unedited tape did not rate Reagan much more positively than did the students who saw the debate without the one-liners. This suggests that Reagan’s jokes did not have much impact on these viewers’ perceptions of him. But look at the third bar in the figure. It illustrates that the students who saw the version of the debate with the one-liners kept in but the audience reaction edited out rated Reagan much less positively than did either of the other groups. What could explain their negativity toward Reagan’s debate performance? Perhaps when Reagan’s jokes appeared to elicit no reaction, the students unknowingly used the lack of reaction as an indica- tion that Reagan’s attempts at wit were inept, and this conclusion caused them to see Reagan in a much less positive light.

In field research, people are observed in real-world settings. Field researchers may observe children in a schoolyard, for example, to study any of a variety of social psychological issues, such as friendship patterns, group dynamics, conformity, helping, aggression, and cultural differences.

G id

eo n

M en

de l/

Co rb

is N

ew s/

Co rb

80220_ch02_ptg01_hr_024-051.indd 40 30/01/16 10:43 AM

41 Testing Ideas: Research Designs

What is interesting about these results from a social psychologi- cal standpoint is that the students’ judgments were influenced more by other people’s reactions to what Reagan said (that is, whether or not the audience appeared to laugh) than by the content of what he said (that is, whether or not the one-liners were edited out of the tape). And it is important to note that these “other people” were not in the room with the students; they were simply sounds on a video- tape recorded many years before. Findings such as this demonstrate that the “social context” can be very subtle and yet can have very powerful effects on our thoughts, feelings, and behaviors.

Independent and Dependent Variables Now that we’ve looked at an experiment, let’s focus on some of the specific elements. In an experiment, researchers manipulate one or more independent variables and examine the effect of these manipulations on one or more dependent variables. In the experiment we’ve just described, there was only one independent variable: the variations of the debate video, with each participant randomly assigned to one of the three versions. The dependent variable was the ratings of Reagan’s perfor- mance. It was the dependent variable because the researchers were in- terested in seeing if the ratings would depend on (that is, be influenced by) the manipulation of the independent variable (that is, by which version of the debate they watched).

Subject Variables Some experiments include variables that are neither dependent nor truly independent. The gender, ethnicity, and prior political leanings of the participants may vary, for example, and researchers may be interested in examining some of these differences. These variables cannot be manipulated and randomly assigned, so they are not true independent variables, and they are not influenced by the independent variables, so they are not dependent variables. Variables such as these are called subject variables because they characterize preexisting differences among the subjects, or partici- pants, in the experiment. If a study includes subject variables but no true, randomly assigned independent variable, it is not a true experiment. But experiments often include subject variables along with independent variables so that researchers can test whether the independent variables have the same or different effects on different kinds of participants. In our presidential debate experiment, for example, we could have examined if the results might have varied not only as a func- tion of our manipulated variable (the version of the debate) but also as a function of the participant’s gender. (In fact, we did look at this, and gender did not make a difference.)

Statistical Significance and Replications In the presidential debate experi- ment, the average rating of Reagan in the unedited condition was 66 on a 0–100 point scale, and it was 49 in the condition in which the audience reaction was removed. Is the difference between 66 and 49 large enough to be meaningful, or could this dif- ference simply be due to chance (that is, just random variation, like flipping a coin 10 times and getting heads 6 times and tails 4)? The answer is that you can’t tell just by looking at these numbers alone. Results obtained in an experiment are examined with statistical analyses that allow the researcher to determine how likely it is that the results could have occurred by random chance. The standard convention is that

d F I G U R e 2 . 5

Influence of Others’ Reactions

This graph shows the results of research in which participants saw different versions of a tape of a 1984 presidential debate between Ronald Reagan and Walter Mondale. During the debate, Reagan had delivered a pair of witty one-liners that elicited a positive audience reaction. Participants who saw an unedited version of the tape and participants who saw a version with the jokes and the audience reaction edited out judged Reagan’s performance similarly. Participants who saw a version with the jokes left in but the audience reaction edited out (suggesting that the audience didn’t find the jokes funny) rated Reagan much more negatively. Adapted from Fein, Goethals, & Kugler, 2007.

Unedited Jokes and reaction

edited out

Jokes left in, reaction

edited out

Ra tin

gs o

f Re

ag an

Version of Tape of 1984 Presidential Debate

independent variable In an experiment, a factor that experimenters manipulate to see if it affects the dependent variable.

dependent variable In an experiment, a factor that experimenters measure to see if it is affected by the independent variable.

subject variable A variable that characterizes preexisting differences among the participants in a study.

80220_ch02_ptg01_hr_024-051.indd 41 30/01/16 10:43 AM

42 Chapter 2 Doing Social Psychology Research

if the results could have occurred by chance 5 or fewer times in 100 possible out- comes, then the result is statistically significant and should be taken seriously.

The fact that results are statistically significant does not mean, however, that they are absolutely certain. In essence, statistical significance is an attractive bet- ting proposition. The odds are quite good (at least 95 out of 100) that the effects obtained in the study were due to the experimental manipulation of the indepen- dent variable. But there is still the possibility (as high as 5 out of 100) that the find- ings occurred by chance. This is one reason why it is important to try to replicate the results of an experiment—to repeat the experiment and see if similar results are found. If similar results are found, the probability that these results could have occurred by chance both times is less than 1 out of 400. Statistical significance is relevant not only for the results of experiments but also for many other kinds of data as well, such as correlations. A correlation between two variables may be statistically significant or not, depending on the strength of the correlation and the number of participants or observations in the data.

When the results of some research are reported in the media or an advertise- ment, it’s not always clear from the reporting whether the results are statistically significant, so it is important to be very cautious when learning about them. You can be sure, however, that whenever we report in this textbook that there is a dif- ference between conditions of an experiment or that two variables are correlated, these results are statistically significant.

Very recently there has been a growing emphasis in psychology (and some other disciplines, such as medicine) on the importance of both replicating re- search findings and using statistical techniques that serve as alternatives to the focus on statistical significance (Trafimow & Marks, 2015). In the coming years these alternatives will no doubt become more popular. We will return to this issue in the final section of this chapter.

Internal Validity: Did the Independent Variable Cause the effect? When an experiment is properly conducted, its results are said to have internal validity: There is reasonable certainty that the independent variable did, in fact, cause the effects obtained on the dependent variable (Cook & Campbell, 1979). As noted earlier, both experimenter control and random assignment seek to rule out alternative explanations of the research results, thereby strengthening the inter- nal validity of the research. If some other factor varies consistently along with the manipulation, this other factor is called a confound. A confound is a serious threat to internal validity and, therefore, makes the issue of cause and effect in the experi- ment uncertain. For example, if the students in the presidential debate study who watched the unedited version of the debate did so in one room, and the students who watched the edited versions did so in a different room, then this would be a confound. It would be impossible to know if it was the manipulation of the version of the debate or if it was some difference between the rooms (such as the tempera- ture, the art on the wall, etc.) that caused the effect on ratings of Reagan’s perfor- mance. Fortunately, the experimenters knew to avoid this problem.

Experiments often include control groups for purposes of internal validity. Typically, a control group consists of participants who experience all of the pro- cedures except the experimental treatment. In the debate study, for example, the participants in the condition in which the video of the debate was unedited was the control group, which provided a baseline against which to compare the judg- ments of participants who watched the other versions.

Outside the laboratory, creating control groups in natural settings that ex- amine real-life events raises many practical and ethical problems. For example,

internal validity The degree to which there can be reasonable certainty that the independent variables in an experiment caused the effects obtained on the dependent variables.

confound A factor other than the independent variable that varies between the conditions of an experiment, thereby calling into question what caused any effects on the dependent variable.

80220_ch02_ptg01_hr_024-051.indd 42 30/01/16 10:43 AM

43 Testing Ideas: Research Designs

researchers testing new medical treatments for deadly diseases face a terrible dilemma. Individuals randomly assigned to the control group are excluded for the duration of the study from what could turn out to be a life-saving new interven- tion. Yet without such a comparison, it is extremely difficult to determine which new treatments are effective and which are useless.

In assessing internal validity, researchers need to consider their own role as well. Unwittingly, they can sometimes sabotage their own research. For example, imagine you are a researcher and you know which participants are in which con- ditions of your experiment. You will no doubt have expectations (and possibly even strong hopes) about how your participants will respond differently between conditions. Because of these expectations, and without realizing it, you may treat the participants a little differently between conditions. It turns out that even very subtle differences in an experimenter’s behavior can influence participants’ be- havior (Rosenthal, 1976). Therefore, because of these experimenter expectancy effects, the results you find in your experiment may be produced by your own actions rather than by the independent variable.

The best way to protect an experiment from these effects is to keep experi- menters uninformed about assignments to conditions. This often is called being “blind to the conditions” in a study. If the experimenters do not know the con- dition to which a participant has been assigned, they cannot treat participants differently as a function of their condition. Of course, there may be times when keeping experimenters uninformed is impossible or impractical. In such cases, the opportunity for experimenter expectancy effects to occur can be reduced at least somewhat by minimizing the interaction between experimenters and participants. For example, rather than receiving instructions directly from an experimenter, participants can be asked to read the instructions on paper or a computer screen.

external Validity: Do the Results Generalize? In addition to guarding inter- nal validity, researchers are concerned about external validity, the extent to which the results obtained under one set of circumstances would also occur in a different set of circumstances (Berkowitz & Donnerstein, 1982). When an experiment has ex- ternal validity, its findings can be assumed to generalize to other people and to other situations. Both the participants in the experiment and the setting in which it takes place affect external validity.

To help increase external validity, social psychologists would love to conduct their experiments with huge samples of participants that are representative of the general population. Usually, however, they must rely on convenience samples drawn from populations that are readily available to them, which explains why so much of social psychological research is conducted on college students. There are very practical reasons for the use of convenience samples. Representative samples are fine for surveys that require short answers to a short list of questions. But what about complex, time-consuming experiments? The costs and logistical problems associated with this would be staggering. Advocates of convenience samples con- tend that the more basic the principle, the less it matters who participates in the research. For example, people from different cultures, regions, and ages might dif- fer in the form of aggression they typically exhibit when angry, but the situational factors that cause people to be more likely to aggress—in whatever way that ag- gression is expressed—may be similar for most individuals across time and place. Yet in spite of these arguments, having the most diverse, representative samples of research participants as possible is ideal. The growing interest in cross-cultural research in the field is certainly one step in the right direction.

experimenter expectancy effects The effects produced when an experimenter’s expectations about the results of an experiment affect his or her behavior toward a participant and thereby influence the participant’s responses.

external validity The degree to which there can be reasonable confidence that the results of a study would be obtained for other people and in other situations.

80220_ch02_ptg01_hr_024-051.indd 43 30/01/16 10:43 AM

44 Chapter 2 Doing Social Psychology Research

Another promising development is the rapidly increasing use of online data collection, which allows for far more diverse sets of participants. There are numerous challenges associated with this approach as well, however, such as having less control over what participants are seeing or doing as they participate in the study from afar. Fortu- nately, recent research testing the data collected via one of the most popular online services, called Mechanical Turk, suggests that the data are at least as reliable as data collected through tradi- tional methods and offer a much greater diversity of participants, although some reasons for caution have been noted as well (Bates & Lanza, 2013; Goodman et al., 2013; Gosling & Mason, 2015).

The external validity of an experiment may also depend in part on how realis- tic the study is for the participants. But what is meant by realistic is not as straight- forward as one might think. Two types of realism can be distinguished: mundane versus experimental (Aronson & Carlsmith, 1968). Mundane realism refers to the extent to which the research setting resembles the real-world setting of interest. In order to study interpersonal attraction, Theodore Newcomb (1961) set up an entire college dormitory—a striking example of mundane realism. Advocates of mundane realism contend that if research procedures are more realistic, research findings are more likely to reveal what really goes on. In contrast, experimental realism refers to the degree to which the experimental setting and procedures are real and involving to the participant, regardless of whether they resemble real life or not. According to those who favor experimental realism, if the experimental situation is compelling and real to the participants while they are participating in the study, their behavior in the lab—even if the lab is in the basement of the psy- chology building—will be as natural and spontaneous as their behavior in the real world. The majority of social psychologists who conduct experiments emphasize experimental realism.

Deception in experiments Researchers who strive to create a highly involving experience for participants often rely on deception, providing participants with false information about experimental procedures. Toward this end, social psychologists sometimes employ confederates, people who act as though they are participants in the experiment but are really working for the experimenter. For example, in Solomon Asch’s (1956) classic research on conformity, research participants made judgments about the lengths of lines while in the midst of a number of confederates—who were pretending to be ordinary participants—who at various times all gave wrong answers. The researchers wanted to see if the real participants would conform to the confederates and give the obviously wrong answer that the confederates had given. Although it was a very odd setting, the situation was a very real one to the partici- pants (and therefore was high in experimental realism), and many of the participants clearly struggled with the decision about whether or not to conform.

Deception not only strengthens experimental realism but also provides other benefits: It allows the experimenter to create situations in the laboratory that

mundane realism The degree to which the experimental situation resembles places and events in the real world.

experimental realism The degree to which experimental procedures are involving to participants and lead them to behave naturally and spontaneously.

deception In the context of research, a method that provides false information to participants.

confederate Accomplice of an experimenter who, in dealing with the real participants in an experiment, acts as if he or she is also a participant.

Many individuals earn money at home by participating in online research projects through a service provided by Amazon called Mechanical Turk. Online services like this now allow social psychologists to reach out to vastly more diverse samples of people from around the world to participate in their studies.

Bl ue

J ea

n Im

ag es

/C ol

la ge

/C or

bi s

80220_ch02_ptg01_hr_024-051.indd 44 30/01/16 10:43 AM

45 Testing Ideas: Research Designs

would be difficult to find in a natural setting, such as a regulated, safe environ- ment in which to study a potentially harmful behavior such as aggression or discrimination. Some research has shown that participants are rarely bothered by deception and often particularly enjoy studies that use it (Smith & Richardson, 1983). Nevertheless, the use of deception creates some serious ethical concerns, leading to debate about whether and how it should be used (Hegtvedt, 2014; Kimmel, 2012). Fortunately, as we will see a bit later in the chapter, procedures have been put in place to try to ensure the ethical integrity of research today.

j Meta-Analysis: Combining Results Across Studies We have seen that social psychologists conduct descriptive, correlational, and experimental studies to test their hypotheses. Another way to test hypotheses in social psychology is to use a set of statistical procedures to examine, in a new way, relevant research that has al- ready been conducted. This tech- nique is called meta-analysis. By “meta-analyzing” the results of a number of studies that have been conducted in different places and by different researchers, research- ers can measure precisely how strong and reliable particular ef- fects are. For example, studies published concerning the effects of alcohol on aggression may sometimes contradict each other. Sometimes alcohol increases ag- gression; sometimes it doesn’t. By combining the data from all the studies that are relevant to this hypothesis and conduct- ing a meta-analysis, a researcher can determine what effect al- cohol typically has, how strong that effect typically is, and perhaps under what specific conditions that effect is most likely to occur. This technique is being used with increasing frequency in social psychology today, and we report the results of many meta-analyses in this textbook.

j Culture and Research Methods Throughout this book you will see examples of social psychological research that examines differences and similarities across cultures. One of the advantages of this approach is that it provides better tests of the external validity of research that has been conducted in any one setting. By examining whether the results of an experiment generalize to a very different culture, social psychologists can begin to answer questions about the universality or cultural specificity of their research. It is important to keep in mind that when a finding in one culture does not general- ize well to another culture, this should not be seen simply as a failure to replicate but instead may be an opportunity to learn about potentially interesting and im- portant cultural differences.

meta-analysis A set of statistical procedures used to review a body of evidence by combining the results of individual studies to measure the overall reliability and strength of particular effects.

The setting in which children attend school can vary dramatically across cultures. Here students sit outside in a class in Malawi, Africa. Recognizing cultural variation has become increasingly important in social psychology today, and social psychologists are conducting their research across a wider range of cultures and contexts than ever before.

Ro be

rt H

ar di

ng /C

or bi

80220_ch02_ptg01_hr_024-051.indd 45 30/01/16 10:43 AM

46 Chapter 2 Doing Social Psychology Research

Cultural investigations present special challenges to researchers. For example, cultural differences have been found in how affected people are by the con- text of questions as they complete a survey, or about the assumptions respondents make about what the researchers have in mind for a given question (Schwarz et al., 2010; Uskul et al., 2013). It also can be difficult for researchers to translate materials from one language into another. Although it is relatively easy to create literal translations, it can be surpris- ingly challenging to create translations that have the same meaning to people from various cultures. Table 2.2 presents examples—from signs displayed around the world—of what can go wrong when simple sentences are poorly translated.

An even more subtle point about language is that multilingual people may think or act differently as a function of what language is being used in a particu- lar setting. A study by Nairán Ramírez-Esparza and others (2008) illustrates this point. They found that how agreeable a sample of bilingual Mexican American participants appeared to be—either on a self-report questionnaire or in their be- havior in an interview—varied significantly as a function of whether the study was conducted in Spanish or in English.

Source: Triandis (1994).

m T a b l e 2 . 2

lost in Translation

d “Drop your trousers here for best results.” (a dry cleaner in Thailand)

d “You are invited to take advantage of the chambermaid.” (a hotel in Japan)

d “Ladies are requested not to have children in the bar.” (a cocktail lounge in Mexico)

d “Take one of our horse-driven city tours—we guarantee no miscarriages.” (a tourist agency in the former Czechoslovakia)

d “We take your bags and send them in all directions.” (an airline in Denmark)

Ethics and Values in Social Psychology Regardless of where research is conducted and what method is used, ethical issues must always be considered. Researchers in all fields have a moral and legal responsibility to abide by ethical principles. In social psychology the use of deception has caused some concern about ethics, as we indicated earlier. In ad- dition, several studies have provoked fierce debate about whether the procedures used in the studies went beyond the bounds of ethical acceptability. The most famous of these controversial studies was designed by Stanley Milgram in the early 1960s. Milgram (1963) designed a series of studies to address the question “Would people obey orders to harm an innocent person?” To test this question, he put volunteers into a situation in which an experimenter commanded them to administer painful electric shocks to someone they thought was another volunteer participant. (In fact, the other person was a confederate who was not actually re- ceiving any shocks.) The experiment had extremely high experimental realism— many of the participants experienced a great deal of anxiety and stress as they debated whether they should disobey the experimenter or continue to inflict pain on another person. The details and results of this experiment will be discussed in Chapter 7 on Conformity, but suffice it to say that the results of the study made people realize how prevalent and powerful obedience can be.

Milgram’s research was inspired by the obedience displayed by Nazi officers in World War II. No one disputes the importance of his research question. What has been debated, however, is whether the significance of the research topic justified exposing participants to possibly harmful psychological consequences. Even though no one in Milgram’s studies actually received the electric shocks, the participants were quite stressed during the study because they thought they

80220_ch02_ptg01_hr_024-051.indd 46 30/01/16 10:43 AM

47 Ethics and Values in Social Psychology

were harming another person, until the experimenter finally told them the truth at the conclusion of the experiment. Under today’s provisions for the protection of human participants, Milgram’s classic experiments probably could not be con- ducted in their original form. (In an interesting twist, even though conducting an experiment like Milgram’s might be impossible now, in popular culture today individuals endure far greater stress and even humiliation in numerous unscripted TV shows for the entertainment of viewers at home.)

Several studies in the history of social psychology have sparked ethical debate or controversy, including a famous study in which Philip Zimbardo and others sim- ulated a prison environment in the basement of Stanford University’s psychology department building to study how ordinary people can be affected in extraordinary ways by the roles they are assigned in a prison environment (Haney et al., 1973). This study is discussed in detail in Chapter 12 on Law. Although the controversial studies such as Milgram’s and Zimbardo’s have received the most public attention, today virtually every social psychology study is evaluated for its ethics by other people before the study can be conducted. In the following sections, we describe current policies and procedures as well as continuing and new concerns about ethics and values in social psychological research.

j Institutional Review Boards and Informed Consent

In 1974, the agency then called the U.S. Department of Health, Education, and Welfare established regulations de- signed to protect human participants in research. These regulations created institutional review boards (IRBs) at all institutions seeking federal funding for research involv- ing human participants. IRBs became a key safeguard for research, taking on the responsibility of reviewing research proposals to ensure that the welfare of participants is adequately protected.

Besides submitting their research to IRBs, researchers must also abide by their profession’s code of ethics. The American Psychological Association (APA), for example, re- quires psychologists to follow its Ethical Principles of Psychologists and Code of Conduct (2002, 2010), which considers a wide range of ethical issues, including those related to research procedures and practices. The APA code stipulates that researchers are obligated to guard the rights and welfare of all those who partici- pate in their studies.

One extremely important practice is to obtain informed consent. Through informed consent individuals are asked whether they wish to participate in the research project and must be given enough information to make an informed decision. Participants must also know that they are free to withdraw from partici- pation in the research at any point. The APA code also recognizes that research “involving only anonymous questionnaires, naturalistic observations, or certain kinds of archival research” may not require informed consent.

j Debriefing Just as participants should receive informed consent before they begin their par- ticipation in a study, they should receive a debriefing at the end of it, especially if deception was used. Debriefing is a process of disclosure in which researchers

informed consent An individual’s deliberate, voluntary decision to participate in research, based on the researcher’s description of what will be required during such participation.

debriefing A disclosure, made to participants after research procedures are completed, in which the researcher explains the purpose of the research, attempts to resolve any negative feelings, and emphasizes the scientific contribution made by the participants’ involvement.

One reason for the use of deception in an experiment is so that the participants will act more naturally when they are not aware of what is being studied. In these cases, it is especially important for the researchers to provide a full and thorough debriefing.

“What if these guys in white coats who bring us food are, like, studying us and we’re part of some kind of

big experiment?”

M ik

e Tw

oh y/

Th e

N ew

Y or

ke r C

ol le

ct io

n/ Th

e Ca

rt oo

n Ba

80220_ch02_ptg01_hr_024-051.indd 47 30/01/16 10:43 AM

48 Chapter 2 Doing Social Psychology Research

fully inform their participants about the nature of the research in which they have participated. During a debriefing, the researcher goes over all procedures, explaining exactly what happened and why. The researcher discusses the pur- pose of the research, reveals any deceptions, and makes every effort to help the participant feel good about having participated. A skillful debriefing takes time and requires close attention to the individual participant. Indeed, we have known students who became so fascinated by what they learned during a debriefing that it sparked their interest in social psychology, and eventually they became social psychologists themselves!

j ethics and Consent Online Along with the tremendous benefits of our ever-expanding online worlds has come troubling losses of privacy. This loss of privacy has opened the window for corporations, marketers, and researchers to peek in to record your actions in ways that raise new questions about ethics. A firestorm of controversy erupted in the summer of 2014, for instance, when scientists working with Facebook published a paper in which they revealed having conducted an experiment on nearly 700,000 Facebook users—without their knowledge—by manipulating their news feeds and recording how the manipulation affected users’ subsequent status updates. As the opening line of a New York Times article about the controversy put it, “To Face- book, we are all lab rats” (Goel, 2014).

Technically, every Facebook user gives consent to being “lab rats” to Facebook scientists when he or she agrees to its terms of service, but in reality we all know that very few people pay attention to all that fine print when they begin to use a new app or service. Other Internet giants like Google routinely manipulate and measure what their users are doing, although typically the (stated) purpose is to improve the user experience or make a site or app more popular. Questions about ethics and informed consent in the online world will no doubt continue to be asked and wrestled with in the coming years.

j Values and Science: Ongoing Debates and New Controversies

Ethical principles are based on moral values. These values set standards for and impose limits on the conduct of research, but do values affect science in other ways as well? Although many people hold science to a standard of complete objec- tivity, science can probably never be completely unbiased and objective because it is a human enterprise. Scientists choose what to study and how to study it; their choices are affected by personal values as well as by professional rewards. Indeed, some think that values should fuel scientific research and that scientists would be not only naive but also irresponsible to try to keep values out of the picture. Most social psychologists, however, strive to use the scientific methods described in this chapter to free themselves of their preconceptions and, thereby, to see reality more clearly and objectively, even if never perfectly.

One value on which the entire field agrees is that researchers must conduct and report their research with complete honesty. It is therefore both shocking and deeply disturbing when a case of academic fraud is revealed. One such example rocked the field toward the end of 2011, when a Dutch social psychologist was caught—and soon confessed to—having committed a massive amount of dishon- esty, involving the fabrication of data published in dozens of studies for about a

80220_ch02_ptg01_hr_024-051.indd 48 30/01/16 10:43 AM

49 Ethics and Values in Social Psychology

decade (Bartlett, 2011). The news of this scandal, together with the announce- ments of several other cases of fraud or suspected fraud in the next few years, caused some social psychologists to question the field’s practices. Also around this time, social psychologist Jonathan Haidt made news criticizing the lack of diversity of political ideologies represented by social psychologists today and sug- gesting that this could bias research (Tierney, 2011). This too triggered a good deal of criticism and controversy.

With all this in the air, a wave of suggestions for how the field should better protect itself against intentional or unintentional bias or dishonesty has emerged. These suggestions include using more advanced and precise statistical methods to better and more fairly test researchers’ ideas, demanding researchers to be more open for public scrutiny of their materials and data, instituting much more emphasis on replicating each other’s research, making public commitments to articulating predictions and planning data analyses before the data are collected— and for journals to agree to publish the results no matter how the data turn out, and so on (Cumming, 2014; Finkel, Eastwick, & Reis, 2015; Klein, Ratliff, et al., 2014; Nosek, Alter, et al., 2015).

It is important to note that these concerns are by no means specific to social psychology. High-profile cases of fraud, for example, span across all fields, includ- ing physics, medicine, history, literature, and journalism (Coscarelli, 2012; Deer, 2011; Rayner, 2010; Sovacool, 2008). For example, a political science article that reported a remarkably effective way of changing people’s attitudes about equal rights for gays received a great deal of national attention in late 2014 and early 2015. In May 2015 evidence came to light that the results apparently were totally fabricated (Bartlett, 2015). The fraud was discovered in large part because other researchers were attempting to replicate this research—providing an important lesson about the value of replication for advancing science and correcting errone- ous or fraudulent findings.

It is also important to note that the strong reaction by the social psychology community to these issues is a testament to how much it cares about its integrity and will work diligently to reassert and protect it in the years to come. The new practices and directions that researchers are taking to ensure integrity and ac- curacy are exciting and energizing, particularly for the next generation of young social psychologists who are beginning to take the reins of the field as it continues to grow rapidly in its findings, scope, and importance.

Your introduction to the field of social psychology is now complete. In these first two chapters, you have gone step by step through a definition of social psy- chology, a review of its history and discussion of its future, an overview of its research methods, and a consideration of ethics and values. As you study the material presented in the coming chapters, the three of us who wrote this book invite you to share our enthusiasm. You can look forward to information that overturns commonsense assumptions, to lively debate and heated controversy, and to a better understanding of yourself and other people. Welcome to the world according to social psychology. We hope you enjoy it!

“[Objectivity in science] is will- ingness (even the eagerness in truly honorable practitioners) to abandon a favored notion when testable evidence disconfirms key expectations.”

—Stephen Jay Gould

80220_ch02_ptg01_hr_024-051.indd 49 30/01/16 10:43 AM

50 Chapter 2 Doing Social Psychology Research

1. Theories in social psychology attempt to explain and predict social psychological phenomena. The best theo- ries are precise, explain all the relevant information, and generate research that can support or disconfirm them. They should be revised and improved as a result of the research they inspire.

2. The goal of basic research is to increase understanding of human behavior; the goal of applied research is to make applications to the world and contribute to the solution of social problems.

3. Researchers often must transform abstract, conceptual variables into specific operational definitions that indi- cate exactly how the variables are to be manipulated or measured. The construct validity of a study is the extent to which the variables were operationalized well.

4. Researchers use self-reports, observations, and technol- ogy to measure variables.

5. Correlational research examines the association between variables. Correlation does not indicate causation; the fact that two variables are correlated does not necessarily mean that one causes the other.

6. Experiments require (1) control by the experimenter over events in the study and (2) random assignment of participants to conditions. Experiments examine the effects of one or more independent variables on one or more dependent variables.

7. Experimental findings have internal validity to the ex- tent that changes in the dependent variable can be at- tributed to the independent variables. Research results have external validity to the extent that they can be generalized to other people and other situations.

8. Meta-analysis uses statistical techniques to integrate the quantitative results of different studies.

9. Established by the federal government, Institutional Review Boards are responsible for reviewing research proposals to ensure that the welfare of participants is adequately protected.

10. Recent controversies in social psychology have led to a variety of suggestions for how the field should better protect itself against intentional or unintentional bias or dishonesty, including more openness to scrutiny, use of different statistical analyses, and greater emphasis on replication.

Review Top 10 Key Points in Chapter 2

80220_ch02_ptg01_hr_024-051.indd 50 30/01/16 10:43 AM

51 Review

applied research (29) basic research (29) bogus pipeline technique (31) confederate (44) confound (42) construct validity (31) correlation coefficient (36) correlational research (36) debriefing (47) deception (44)

dependent variable (41) experiment (38) experimental realism (44) experimenter expectancy effects (43)

external validity (43) hypothesis (28) independent variable (41) informed consent (47) internal validity (42)

interrater reliability (32) meta-analysis (45) mundane realism (44) operational definition (30) random assignment (39) random sampling (35) subject variable (41) theory (28)

Key Terms

80220_ch02_ptg01_hr_024-051.indd 51 30/01/16 10:43 AM

The Social Self

This chapter examines three interrelated aspects of the “social self.” First, it considers the self-concept and the question of how people come to understand who they are, their own actions, emotions, and motivations. Second, it considers self-esteem, the affective component, and the question of how people evaluate themselves and defend against threats to their self- esteem. Third, it considers self-presentation, a behavioral manifestation of the self, and the question of how people present themselves to others. As we will see, the self is complex and multifaceted.

80220_ch03_ptg01_hr_052-103.indd 52 27/01/16 11:21 AM

The Social Self 3 The Self-Concept | 55

Rudiments of the Self-Concept

Introspection

Self-Perception

Influences of Other People

Autobiographical Memories

Culture and the Self-Concept

Self-Esteem | 76 The Need for Self-Esteem

Are There Gender and Race Differences?

Self-Discrepancy Theory

The Self-Awareness “Trap”

Self-Regulation and Its Limits

Ironic Mental Processes

Mechanisms of Self-Enhancement

Are Positive Illusions Adaptive?

Culture and Self-Esteem

Self-Presentation | 96 Strategic Self-Presentation

Self-Verification

Self-Monitoring

Reflections: The Multifaceted Self | 101 Review

53AP Im

ag es

/H as

sa n

A m

m ar

80220_ch03_ptg01_hr_052-103.indd 53 27/01/16 11:21 AM

54 Chapter 3 The Social Self

Can you imagine living a meaningful or coherent life without a clear sense of who you are? In The Man Who Mistook His Wife for a Hat, neurologist Oliver Sacks (1985) described such a person—a patient named William Thompson. According to Sacks, Thompson suffered from an organic brain disorder that im- pairs a person’s memory of recent events. Unable to recall anything for more than a few seconds, Thompson was always disoriented and lacked a sense of inner continuity. The effect on his behavior was startling. Trying to grasp a constantly vanishing identity, Thompson would construct one tale after another to account for who he was, where he was, and what he was doing. From one moment to the next, he would improvise new identities—a grocery store clerk, a minister, or a medical patient, to name just a few. In social settings, Thompson’s behavior was especially intriguing. As Sacks (1985) observed,

The presence of others, other people, excite and rattle him, force him into an end- less, frenzied, social chatter, a veritable delirium of identity-making and -seeking; the presence of plants, a quiet garden, the nonhuman order, making no social demands upon him, allow this identity-delirium to relax, to subside. (p. 110)

Thompson’s plight is unusual, but it highlights two important points— one about the private “inner” self, the other about the “outer” self we show to others. First, the capacity for self-reflection is necessary for people to feel as if they understand their own motives and emotions and the causes of their behavior. Unable to ponder his own actions, Thompson appeared vacant and without feeling—”desouled,” as Sacks put it. Second, the self is heavily influenced by

social factors. Thompson himself seemed compelled to put on a face for others and to improvise characters for the company he kept. We all do, to some extent. We may not create a kaleidoscope of multiple identities as Thompson did, but the way we manage ourselves is influenced by the people around us.

This chapter examines the ABCs of the self: A for affect, B for behavior, and C for cognition. First, we ask a cognitive question: How do people come to know themselves, develop a self-concept, and maintain a stable sense of identity? Second, we explore an affec- tive, or emotional, question: How do people evaluate themselves, enhance their self-images, and defend against threats to their self-esteem? Third, we confront a behavioral question: How do people regulate their own actions and present themselves to others accord- ing to interpersonal demands? As we’ll see, the self is a topic that has attracted unprecedented interest among social psychologists (Leary & Tangney, 2003; Sedikides & Spencer, 2007; Swann & Bosson, 2010; Vohs & Finkel, 2006).

Putting Common SenSe to the Test

T F Humans are the only animals who recognize themselves in the mirror.

T F Smiling can make you feel happier.

T F Sometimes the harder you try to control a thought, feeling, or behavior, the less likely you are to succeed.

T F People often sabotage their own performance in order to protect their self-esteem.

T F It’s more adaptive to alter one’s behavior than to stay consistent from one social situation to the next.

Circle Your Answer

80220_ch03_ptg01_hr_052-103.indd 54 27/01/16 11:21 AM

55 The Self-Concept

self-concept The sum total of an individual’s beliefs about his or her own personal attributes.

self-schema A belief people hold about themselves that guides the processing of self-relevant information.

The Self-Concept Have you ever been at a noisy gathering—holding a drink in one hand and a spring roll in the other, struggling to have a conversation over music, vibrating phones, and the chatter of voices—and yet managed to hear someone at the other end of the room say your name? If so, then you have experienced the “cocktail party effect”—the tendency of people to pick a personally relevant stimulus, like a name, out of a complex and noisy environment (Cherry, 1953; Conway, Cowan, & Bunting, 2001). Even infants who are too young to walk or talk exhibit the ten- dency to respond to their own name (Newman, 2005). To the cognitive psycholo- gist, this phenomenon shows that human beings are selective in their attention. To the social psychologist, it also shows that the self is a brightly lit object of our own attention.

The term self-concept refers to the sum total of beliefs that people have about themselves. But what specifically does the self-concept consist of? Ac- cording to Hazel Markus (1977), the self-concept is made up of cognitive mole- cules she called self-schemas: beliefs about oneself that guide the processing of self-relevant information. Self-schemas are to an individual’s total self-concept what hypotheses are to a theory or what books are to a library. You can think of yourself as masculine or feminine, as independent or dependent, as liberal or conservative, as introverted or extroverted. Indeed, any specific attribute may have relevance to the self-concept for some people but not for others. The self- schema for body weight is a good example. Men and women who regard them- selves as overweight or underweight, or for whom body image is a conspicuous aspect of the self-concept, are considered schematic with respect to weight. For body-weight schematics, a wide range of otherwise mundane events—maybe a trip to the supermarket, the sight of a fashion model, dinner at a restaurant, a day at the beach, or watching a friend diet—may trigger thoughts about the self. In contrast, those who do not regard their own weight as extreme or as an important part of their lives are aschematic on that attribute (Markus et al., 1987).

It is important to realize that people are multifaceted and that our self-concept may consist of a multitude of self-schemas. As we will see shortly, people who identify with two cultures may have a “double consciousness” about who they are and hold different self-schemas that fit within each culture. African Americans, for example, have one self-schema that fits generally within mainstream American culture and another tied more specifically to African American culture (Brannon, Markus, & Taylor, 2015).

j Rudiments of the Self-Concept Clearly the self is a special object of our attention. Whether you are mentally focused on a memory, a tweet, a foul odor, the song in your head, your growl- ing stomach, or this sentence, consciousness is like a “spotlight.” It can shine on only one object at a point in time, but it can also shift rapidly from one object to another and process information outside of awareness. In this spotlight, the self is front and center. But is the self so special that it is uniquely represented in the neural circuitry of the brain? And is the self a uniquely human concept, or do other animals also distinguish the self from everything else?

Is the Self Specially Represented in the Brain? As illustrated by the story of William Thompson that opened this chapter, our sense of identity is biologically

80220_ch03_ptg01_hr_052-103.indd 55 27/01/16 11:21 AM

56 Chapter 3 The Social Self

rooted. In The Synaptic Self: How Our Brains Become Who We Are, neuroscientist Joseph LeDoux (2002) argues that the synaptic connections in the brain provide the biological base for memory, which makes possible the sense of continuity that is needed for a normal identity. In The Self Illusion: How the Social Brain Creates Identity, developmental psychologist Bruce Hood (2012) notes that our sense of self emerges in childhood through our social interactions—and that it is a mere illusion, “a powerful deception generated by our brains for our own benefit.” In The Lost Self: Pathologies of the Brain and Identity, Todd Feinberg and Julian Keenan (2005) describe how the self can be transformed or completely destroyed by severe head injuries, brain tumors, diseases, and exposure to toxic substances that damage the brain and nervous system.

Social neuroscientists have started to explore these possibilities. Using PET scans, fMRI, and other imaging techniques that capture the brain in ac- tion, these researchers are finding that certain areas become more active when laboratory participants see a picture of themselves rather than a picture of another person (Platek et al., 2008), when they viewed self-relevant words such as their own name or street address rather than other-relevant words (Moran et al., 2009), and when they take a first-person perspective while play- ing a video game as opposed to a third-person perspective (David et al., 2006). As we will see throughout this chapter, the self is a frame of reference that powerfully influences our thoughts, feelings, and behaviors. Not all aspects of the self are housed in a single structure of the brain. However, the bulk of research does seem to suggest that various self-based processes can be traced to activities occurring in certain areas (Qin, Duncan, & Northoff, 2013; Heatherton, 2011).

Do Nonhuman Animals Show Self-Recognition? When you stand in front of a mirror, you recognize the image as a reflection of yourself. But what about dogs, cats, and other animals—how can we possibly know what nonhumans think about mirrors? In a series of studies, Gordon Gallup (1977) placed different species of animals in a room with a large mirror. At first, they greeted their own images by vocalizing, gesturing, and making other social responses. After several days, only great apes (chimpanzees, gorillas, and orangutans)—but not other animals—seemed capable of self-recognition, using the mirror to pick food out of their teeth, groom themselves, blow bubbles, and make faces for their own entertainment. From all appearances, the apes recognized themselves.

In other studies, Gallup anesthetized the animals, then painted an odor- less red dye on their brows and returned them to the mirror. Upon seeing the red spot, only the apes spontaneously reached for their own brows—proof that they perceived the image as their own (Povinelli et al., 1997; Keenan et al., 2003). Among the apes, this form of self-recognition emerges in young ad- olescence and is stable across the life span, at least until old age (de Veer et al., 2003). By using a similar red dye test (without anesthetizing the infants), developmental psychologists have found that most humans begin to recognize themselves in the mirror between the ages of 18 and 24 months (Lewis & Brooks-Gunn, 1979).

Many researchers believe that self-recognition among great apes and human infants is the first clear expression of the concept “me.” Recent research suggests that certain intelligent non-primates can also recognize themselves. In one study, researchers at a New York aquarium found that two bottlenose dolphins marked with black ink often stopped to examine themselves in a mirror (Reiss & Marino, 2001). In a second study, researchers found that three Asian elephants placed in

80220_ch03_ptg01_hr_052-103.indd 56 27/01/16 11:21 AM

57 The Self-Concept

front of a jumbo-sized mirror used the mirror to inspect themselves—as when they moved their trunks to see the insides of their mouths, a part of the body they usually cannot see (Plotnik et al., 2006). In contrast, testing of thirty-four giant pandas of varying ages showed that they did not recognize their own images in the mirror (Ma et al., 2015).

It’s important not to assume from this research that the mirror test is a pure measure of self-recognition or that it emerges at the same age throughout the world. Tanya Broesch and others (2011) tested children between the ages of 33 and 72 months in a number of countries. In line with past research, 88% of American children and 77% of Canadian children “passed” the test. Yet elsewhere it was only 58% in Saint Lucia, 52% in Peru, and 51% in Grenada; only two children passed in Kenya and none did so in Fiji. Based on their observations, the re- searchers speculated that the children in these non-Western countries did not lack self-recognition. They knew it was their image in the mirror but—having been raised for compliance and trained not to ask questions—they did not dare touch or remove the mark. Whatever the interpretation, this cross-cultural research raises questions as to whether the mirror test can be used to measure the self-concept (Broesch et al., 2011).

What Makes the Self a Social Concept? The ability to see yourself as a dis- tinct entity in the world may be a necessary first step in the evolution and develop- ment of a self-concept. The second step involves social factors. Sociologist Charles Horton Cooley (1902) introduced the term looking-glass self to suggest that other people serve as a mirror in which we see ourselves. Expanding on this idea, George Herbert Mead (1934) added that we often come to know ourselves by imagining what significant others think of us and then incorporating these perceptions into our self-concepts.

Picking up where the classic sociologists left off, Susan Andersen and Serena Chen (2002) theorized that the self is “relational”—that we draw our sense of who

Humans are the only animals who recognize themselves in the mirror.

FAlSe

Researchers found that elephants placed in front of the jumbo-size mirror pictured here used the mirror to inspect themselves. This observation suggests that elephants join human beings, great apes, and bottlenose dolphins as animal species that exhibit self-recognition (Plotnik et al., 2006).

Jo sh

ua P

lo tn

ik /P

A P

ho to

s/ La

nd ov

80220_ch03_ptg01_hr_052-103.indd 57 27/01/16 11:21 AM

58 Chapter 3 The Social Self

we are from our past and current relationships with the significant others in our lives. It is interesting that when Gallup tested his apes, those that had been raised in isolation—without exposure to peers—did not recognize themselves in the mirror. Only after such exposure did they begin to show signs of self-recognition. Among human beings, our self-concepts match our perceptions of what others think of us. Illustrating our capacity for “meta-insight,” research also shows that people can distinguish between how they perceive themselves—for example, how smart, funny, or outgoing—and how others see them (Carlson et al., 2011). In fact, it seems that we can tell when our perceptions of what others think of us are more or less correct (Carlson & Furr, 2013).

In recent years, social psychologists have broken new ground in the effort to understand the social self. People are not born thinking of themselves as smart, lazy, reckless, likable, shy, or outgoing. So where do their self-concepts come from? In the coming pages, the following sources are considered: introspection, perceptions of our own behavior, other people, autobiographical memories, and the cultures in which we live.

j Introspection Let’s start at the beginning: How do people achieve insight into their own beliefs, attitudes, emotions, desires, personalities, and motivations? Although common sense makes this question seem ludicrous, many social psychologists have sought to answer the question of how, and how well, people gain self-knowledge (Vazire & Wilson, 2012).

Think about this: Don’t you know what you think because you think it? And don’t you know how you feel because you feel it? Look through popular books on how to achieve self-insight, and you’ll find the unambiguous answers to these questions to be yes. Whether the prescribed technique is yoga, meditation, psy- chotherapy, religion, dream analysis, or hypnosis, the advice is basically the same: Self-knowledge is derived from introspection, a looking inward at one’s own thoughts and feelings.

If the how-to books are correct, it stands to reason that no one can know you as well as you know yourself. Thus, people tend to assume that for others to know you at all, they would need information about your private thoughts, feelings, and other inner states—not just your behavior. But is this really the case? Most social psychologists are not sure that this faith in introspection is justified. Some forty years ago, Richard Nisbett and Timothy Wilson (1977) conducted a series of experiments in which they found that research participants often could not ac- curately explain the causes or correlates of their own behavior. This observation forced researchers to confront a thorny question: Does introspection provide a direct pipeline to self-knowledge?

In Strangers to Ourselves, Wilson (2002) argued that it does not. In fact, he finds that introspection can sometimes lead us astray on the road to self-knowledge. In a series of studies, he found that the attitudes people reported having about different objects corresponded closely to their behavior toward those objects. The more participants said they enjoyed a task, the more time they spent on it; the more attractive they found a scenic landscape, the more pleasure was revealed in their facial expressions; the happier they said they were with a current dat- ing partner, the longer the relationship with that partner ultimately lasted. Yet after participants were told to analyze the reasons for how they felt, the attitudes that they reported no longer corresponded to their behavior. There are two prob- lems. The first, as described by Wilson, is that human beings are mentally busy

80220_ch03_ptg01_hr_052-103.indd 58 27/01/16 11:21 AM

59 The Self-Concept

processing information, which is why we so often fail to understand our own thoughts, feelings, and behaviors. Apparently, it is possible to think too much and be too analytical, only to get confused.

In Self-Insight: Roadblocks and Detours on the Path to Knowing Thyself, David Dunning (2005) points to a second type of problem in self- assessment: that people overestimate the posi- tives. Most people, most of the time, think they are better than average, even though it is statisti- cally impossible for this to happen. As we will see in our later discussion of self-enhancement, people from all walks of life tend to overrate their own skills, their prospects for success, the accuracy of their opinions, and the impressions they form of others—possibly with dire consequences for their health and well-being. We will also see, however, that many people have insight into their own positive—and sometimes negative—biases. In a study that demonstrates the point, Kathryn Bollich and others (2015) found that most people who harbor biased self-perceptions (for example, about how warm, dependable, stable, and funny they are relative to how they are rated by their own peers) accurately describe themselves as biased when prompted.

When it comes to self-insight, people do have difficulty projecting forward and predicting how they would feel in response to future emotional events—a pro- cess known as affective forecasting. Imagine that you have a favorite candidate in an upcoming political campaign. Can you anticipate how happy you would be one month after the election if this candidate were to win? How unhappy would you be if he or she were to lose? Closer to home, how happy would you be six months after winning a million-dollar lottery? Or how unhappy would you be if you were injured in an automobile accident?

In a series of studies, Timothy Wilson and Daniel Gilbert (2003) asked re- search participants to predict how they would feel after various positive and nega- tive life events and compared their predictions to how others experiencing those events said they actually felt. Consistently, they and others have found that people overestimate the strength and duration of their emotional reactions, a phenom- enon they call the impact bias (Wilson & Gilbert, 2013). In one study, junior pro- fessors predicted that receiving tenure would increase their happiness levels for several years, yet professors who actually received tenure were no happier several years later than those not granted tenure. In a second study, voters predicted they would be happier a month after an election if their candidate won than if he or she lost. In actuality, supporters of the winning and losing candidates did not differ in their happiness levels one month after the election.

There are two possible reasons for the impact bias in affective forecasting. First, when it comes to negative life events—such as an injury, illness, or big fi- nancial loss—people do not fully appreciate the extent to which our psychological coping mechanisms help us to cushion the blow. In the face of adversity, human beings can be remarkably resilient—and not as devastated as we fear we will be (Gilbert et al., 1998). In fact, people are even more likely to overlook the coping mechanisms that others use. The result is a self–other difference by which we tend to predict that others will suffer even longer than we will (Igou, 2008). A second reason for overestimates is that when we introspect about the emotional impact on us of a future event—say, the breakup of a close relationship—we become so focused on that single event that we neglect to take into account the effects of

In some ways, our sense of self is malleable and subject to change.

Ro be

rt M

an ko

ff /T

he N

ew Y

or ke

r C ol

le ct

io n/

Th e

Ca rt

oo n

Ba nk

“Look, babe. At this point, you’ve reinvented yourself so many times you’re back to who you were at the start.”

affective forecasting The process of predicting how one would feel in response to future emotional events.

80220_ch03_ptg01_hr_052-103.indd 59 27/01/16 11:21 AM

60 Chapter 3 The Social Self

other life experiences. To become more accurate in our predictions, then, we need to force ourselves to think more broadly, about all the events that impact us. In one study, college students were asked to predict their emotional reactions to their school football team’s winning or losing an important game. As usual, they tended to overestimate how long it would take them to recover from the victory or defeat. This bias disappeared, however, when the students first completed a “prospec- tive diary” in which they estimated the amount of future time they will spend on everyday activities like going to class, talking to friends, studying, and eating meals (Wilson & Ross, 2000).

j Self-Perception Regardless of what we can learn from introspection, Daryl Bem (1972) proposed that people can learn about themselves the same way outside observers do—by watching their own behavior. Bem’s self-perception theory is simple yet pro- found. To the extent that internal states are weak or difficult to interpret, people infer what they think or how they feel by observing their own behavior and the situation in which that behavior takes place. Have you ever listened to yourself arguing with someone in an e-mail exchange, only to realize with amazement how angry you were? Have you ever devoured a sandwich in record time, only then to conclude that you must have been incredibly hungry? In each case, you made an inference about yourself by watching your own actions.

There are limits to self-perception, of course. According to Bem, people do not infer their own internal states from behavior that occurred in the presence of com- pelling situational pressures such as reward or punishment. If you argued vehe- mently or wolfed down a sandwich because you were paid to do so, you probably would not assume that you were angry or hungry. In other words, people learn about themselves through self-perception only when the situation alone seems insufficient to have caused their behavior.

Over the years, a good deal of research has supported self-perception theory. When people are gently coaxed into saying or doing something and when they are not otherwise certain about how they feel, they often come to view themselves in ways that are consistent with their public statements and behaviors (Chaiken & Baldwin, 1981; Kelly & Rodriguez, 2006; Schlenker & Trudeau, 1990). In one study, participants induced to describe themselves in flattering terms scored higher on a later test of self-esteem than did those who were led to describe themselves more modestly (Jones et al., 1981; Rhodewalt & Agustsdottir, 1986). In another study, people who were maneuvered by leading questions into describing themselves as introverted or extroverted—whether or not they really were—often came to define themselves as such later on (Fazio & Zanna, 1981; Swann & Ely, 1984). British author E. M. Forster long ago anticipated the theory when he asked, “How can I tell what I think ‘til I see what I say?”

Self-perception theory may have even more reach than Bem had anticipated. Bem argued that people sometimes learn about themselves by observing their own freely chosen behavior. But might you also infer something about yourself by observing the behavior of someone else with whom you completely identify? In a series of studies, Noah Goldstein and Robert Cialdini (2007) demonstrated this phenomenon, which they call vicarious self-perception. In one experiment, for example, they asked college students to listen to an interview with a fellow stu- dent who had agreed afterward to spend a few extra minutes helping out on a project on homelessness. Before listening to the interview, all participants were fitted with an EEG recording device on their foreheads that allegedly measured

self-perception theory The theory that when internal cues are difficult to interpret, people gain self-insight by observing their own behavior.

80220_ch03_ptg01_hr_052-103.indd 60 27/01/16 11:21 AM

61 The Self-Concept

brain activity as they viewed a series of images and words. By random assign- ment, some participants but not others were then told that their brain-wave pat- terns closely resembled that of the person whose interview they would soon hear—a level of resemblance, they were told, that signaled genetic similarity and relationship closeness. Would participants in this similarity feedback condition draw inferences about themselves by observing the behavior of a fellow student? Yes. In a post-interview questionnaire, these participants (compared to those in the no-feedback control group) rated themselves as more sensitive and as more self-sacrificing if the student whose helpfulness they observed was said to be similar, biologically. What’s more, when the session was over, 93% of those in the similarity condition agreed to spend some extra time themselves helping the experimenter— compared to only 61% in the no-feedback control group.

Introspection and self-perception theory make different predictions about the extent to which people can know themselves. If self-knowledge derives from pri- vate introspection, then clearly you know yourself better than anyone else can. If self-knowledge derives solely from observations of behavior, then it should be possible for others to know us as well as we know ourselves. Assuming that self-knowledge is gained from both sources, then the truth lies somewhere in the middle. But wait. Is it ever possible for others to know us better than we know ourselves?

Simine Vazire (2010) asked this very question and came up with a surprising answer. Vazire pro- posed a Self–Other Knowledge Asymmetry (SOKA) model in which she predicts that we know ourselves better than others do when it comes to traits that are “internal” and hard to observe (such as how opti- mistic, anxious, or easily upset a person is) and that there is no self-other difference when it comes to traits that are “external” and easy to observe (such as how quiet, sociable, or messy a person is). She also predicts that others may actually know us better than we know ourselves when it comes to observ- able traits that can be so touchy for self-esteem pur- poses that we have motivated “blind spots” (such as how smart, creative, or rude a person is). In these latter instances, Vazire predicts, others can be more objective than we are about ourselves.

To test these predictions, Vazire asked college students to rate themselves—and then had their friends rate the participants—on a number of per- sonality traits. Three types of traits were studied: (1) high in observability (talkativeness, dominance, and leadership), (2) low in observability and not evaluative (self-esteem and anxiety), and (3) low in observability and highly evaluative (intelligent and creative). To determine accuracy, Vazire then mea- sured how participants fared on objective measures of these traits using various laboratory exercises and paper-and-pencil tests. The results provided strong support for the SOKA model. d Figure 3.1 shows that self- and friend-ratings were equally ac- curate for highly observable traits, that self-ratings

d F I g u R e 3 . 1

The Self–Other Knowledge Asymmetry (SOKA) Model

Participants rated themselves and were rated by friends on personality traits for which they took various objective tests. As shown, self- ratings were more accurate for internal/nonevaluative traits (left) and self- and friend-ratings were equally accurate for observable/ nonevaluative traits (center). Interestingly, however, friend-ratings were more accurate for internal/evaluative traits. Supporting SOKA, this pattern shows that “know thyself” requires a combination of information and objectivity. Based on Vazire, S. (2010). Who knows what about a person? The self–other knowledge asymmetry (SOKA) model. Journal of Personality and Social Psychology, 98, 281–300.

Internal/ non-evaluative

Self Friends

Type of Trait

Observable/ non-evaluative

0.1

0.2

0.3

0.4

A cc

ur ac

y of

R at

in gs

Internal/ evaluative

80220_ch03_ptg01_hr_052-103.indd 61 27/01/16 11:21 AM

62 Chapter 3 The Social Self

were more accurate for internal non-evaluative traits, but that friend-ratings were more accurate for internal evaluative traits. Clearly, to know thyself requires a combination of information and objectivity (Vazire & Carlson, 2011).

Self-Perceptions of emotion Draw the corners of your mouth back and up and tense your eye muscles. Okay, relax. Now raise your eyebrows, open your eyes wide, and let your mouth drop open slightly. Relax. Now pull your brows down and to- gether and clench your teeth. Relax. If you followed these direc- tions, you would have appeared to others to be feeling first happy, then fearful, and finally angry. The question is, how would you have appeared to yourself?

Social psychologists who study emotion have asked precisely that question. Viewed within the framework of self-perception theory, the facial feedback hypothesis states that changes in facial expression can trigger corresponding changes in the sub- jective experience of emotion. In the first test of this hypoth- esis, James Laird (1974) told participants that they were taking part in an experiment on activity of the facial muscles. After attaching electrodes to their faces, he showed them a series of cartoons. Before each one, the participants were instructed to contract certain facial muscles in ways that created either a

smile or a frown. As Laird predicted, participants rated what they saw as funnier when they were smiling than when they were frowning.

Suggesting that this effect is universal, researchers recently replicated these find- ings in Ghana, West Africa (Dzokoto et al., 2014).

In one particularly interesting field study that illustrates how our emotional state can be influenced by naturally occurring changes in facial expression, re- searchers from Italy stopped a random sample of men and women on a beach and asked them in a questionnaire to report on how angry and aggressive they were feeling. Some of the beach goers faced the bright sun while answering the ques- tion, causing “sun-induced frowning” on their faces. Others were questioned with their back to the sun or while wearing sunglasses. As predicted by the facial feed- back hypothesis, and even though participants themselves said that their mood was not affected by the sunlight, those who frowned at the sun reported higher levels of anger than all others (Marzoli et al., 2013).

It is clear that facial feedback can evoke and magnify certain emotional states. It’s important to note, however, that the face is not necessary to the subjective experience of emotion. When neuropsychologists recently tested a young woman who suffered from bilateral facial paralysis, they found that despite her inability to outwardly show emotion, she reported feeling various emotions in response to positive and negative visual images (Keillor et al., 2003).

How does facial feedback work? With 80 muscles in the human face that can create over 7,000 expressions, can we actually vary our own emotions by contract- ing certain muscles and wearing different expressions? Research suggests that we can, though it is not clear what the results mean. Laird argues that facial expres- sions affect emotion through a process of self-perception: “If I’m smiling, I must be happy.” Consistent with this hypothesis, Chris Kleinke and others (1998) asked people to emulate either happy or angry facial expressions depicted in a series of photographs. Half the participants saw themselves in a mirror during the task; the others did not. Did these manipulations affect mood states? Yes. Compared

As suggested by self-perception theory, we sometimes infer how we feel by observing our own behavior.

facial feedback hypothesis The hypothesis that changes in facial expression can lead to corresponding changes in emotion.

“I don’t sing because I am happy. I am happy because I sing.”

Ed w

ar d

Fr as

ci no

/T he

N ew

Y or

ke r C

ol le

ct io

n/ Ca

rt oo

n Ba

nk .c

80220_ch03_ptg01_hr_052-103.indd 62 27/01/16 11:21 AM

63 The Self-Concept

to participants in a no-expression control group, those who put on happy faces felt better, and those who put on angry faces felt worse. As predicted by self- perception theory, the differences were particularly pronounced among partici- pants who saw themselves in a mirror.

Other researchers believe that facial movements spark emotion by produc- ing physiological changes in the brain. For example, Robert Zajonc (1993) argues that smiling causes facial muscles to increase the flow of air-cooled blood to the brain, a process that produces a pleasant state by lowering brain temperature. Conversely, frowning decreases blood flow, producing an unpleasant state by rais- ing temperature. To demonstrate, Zajonc and his colleagues (1989) conducted a study in which they asked participants to repeat certain vowels 20 times each, including the sounds ah, e, u, and the German vowel ü. In the meantime, tem- perature changes in the forehead were measured and participants reported on how they felt. As it turned out, ah and e (sounds that cause people to mimic smil- ing) lowered forehead temperature and elevated mood, whereas u and ü (sounds that cause us to mimic frowning) increased temperature and dampened mood. In short, people need not infer how they feel. Rather, facial expressions evoke physi- ological changes that produce an emotional experience.

Other expressive behaviors, such as body posture, can also provide us with sensory feedback and influence the way we feel. When people feel proud, they stand erect with their shoulders raised, chest expanded, and head held high (expansion). When dejected, however, we slump over with shoulders drooping and head bowed (contraction). Clearly, your emotional state is revealed in the way you carry yourself. But is it also possible that the way you carry yourself affects your emotional state? Can people lift their spirits by expansion or lower their spir- its by contraction? Yes. Sabine Stepper and Fritz Strack (1993) arranged for people to sit in either a slumped or an upright position by varying the height of the table they had to write on. Those forced to sit upright reported feeling more pride after succeeding at a task than did those placed in a slumped position.

Self-Perceptions of Motivation Without quite realizing it, the American author Mark Twain was a self-perception theorist. In The Adventures of Tom Sawyer, written in the late 1800s, he quipped, “There are wealthy gentlemen in England who drive four-horse passenger coaches 20 or 30 miles on a daily line, in the summer, because the privilege costs them considerable money; but if they were offered wages for the service that would turn it into work then they would resign.” Twain’s hypothesis—that reward for an enjoyable activity can under- mine interest in that activity—seems to contradict both our intuition and the results of psychological research. After all, aren’t we all motivated by reward, as B. F. Skinner and other behaviorists have declared? The answer depends on how motivation is defined.

A keen observer of human behavior, Twain anticipated a key distinction be- tween intrinsic and extrinsic motivation. Intrinsic motivation originates in factors within a person. People are said to be intrinsically motivated when they engage in an activity for the sake of their own interest, the challenge, or sheer enjoy- ment. Eating a good meal; listening to music; spending time with friends, or on Facebook; or getting engrossed in a book, a Netflix movie, a sports event, or a video game—these are the kinds of activities that you might find intrinsically mo- tivating. In contrast, extrinsic motivation originates in factors outside the person. People are said to be extrinsically motivated when they engage in an activity as a means to an end, for tangible benefit. It might be to acquire money, grades, or some other kind of recognition; to fulfill an obligation; or to avoid a penalty or

Smiling can make you feel happier.

TRue

Ti m

e Li

fe P

ic tu

re s/

G et

ty Im

ag es

Nineteenth century American author Mark Twain understood the effects of reward on intrinsic motivation.

80220_ch03_ptg01_hr_052-103.indd 63 27/01/16 11:21 AM

64 Chapter 3 The Social Self

punishment. Clearly, people strive for reward. But what happens to the intrinsic motivation once that reward is no longer available?

From the standpoint of self-perception theory, Twain’s hypothesis makes sense. When someone is offered a reward for something they already like to do—whether it’s listening to music, playing a game, or eating a tasty food—that behavior becomes overjustified, or overrewarded, which means that it can be attributed to extrinsic as well as intrinsic motives. By creating ambiguity about a person’s motivation, can the overjustification effect have unintended conse- quences? When athletes are paid millions of dollars to play their sport, does the money overwhelm their love of the game, making play feel like work? Once paid, do people begin to wonder if the activity was ever worth pursuing in its own right?

Research has shown that when people start getting rewarded for a task they already enjoy, they sometimes lose interest in it over time. In a classic demon-

stration of this phenomenon, Mark Lepper and his colleagues (1973) gave preschool children an opportunity to play with colorful felt-tipped markers—an opportunity most could not resist. By observing how much time the children spent on the activity, the researchers were able to measure their intrinsic motivation. Two weeks later, the children were divided into three groups, all about equal in terms of initial levels of in- trinsic motivation. In one, the children were simply asked to draw some pictures with the markers. In the second, they were told that if they used the markers, they would receive a “Good Player Award,” a certificate with a gold star and a red ribbon. In a third group, the children were not offered a reward for drawing pictures, but—like those in the second group—they received a reward when they were done.

About a week later, the teachers placed the markers and paper on a table in the classroom while the experimenters observed through a one-way mirror. Since no rewards were offered on this occasion, the amount of free time the children spent playing with the markers reflected their intrinsic moti- vation. As predicted, those who had expected and received a reward for their efforts were no longer as interested in the markers as they had been. Children who had not received a reward were not adversely affected, nor were those who had received the unexpected reward. Having played with the markers without the promise of tangible benefit, these chil- dren remained intrinsically motivated (see d Figure 3.2).

The paradox that reward can undermine rather than enhance intrinsic motivation has been observed in many settings and with both children and adults (Deci & Ryan, 1985; Tang & Hall, 1995). Accept money for a leisure activity, and before you know it, what used to be “play” comes to feel more like “work.” In the long run, this can have unintended negative effects on the quality of your performance.

In a series of studies, Teresa Amabile (1996) and others had participants write poems, draw or paint pictures, make paper collages, and generate creative solu- tions to business dilemmas. Consistently, they found that people are more creative when they feel interested and challenged by the work itself than when they feel pressured to make money, fulfill obligations, meet deadlines, win competitions, or impress others. When Amabile had art experts rate the works of professional

overjustification effect The tendency for intrinsic motivation to diminish for activities that have become associated with reward or other extrinsic factors.

d F I g u R e 3 . 2

Paradoxical effects of Reward on Intrinsic Motivation

In this study, an expected reward undermined children’s intrinsic motivation to play with felt-tipped markers. Children who received an unexpected reward or no reward did not lose interest. Based on Lepper, M. R., Greene, D., and Nisbett, R. E. (1973). Undermining children’s intrinsic interest with extrinsic reward: A test of the ‘overjustification’ hypothesis. Journal of Personality and Social Psychology, 28, 129–137.

No reward

Unexpected reward

Expected reward

Pe rc

en ta

ge o

f t im

e sp

en t p

la yi

ng w

ith m

ar ke

80220_ch03_ptg01_hr_052-103.indd 64 27/01/16 11:21 AM

65 The Self-Concept

artists, she found that their commissioned work (art they were contracted for) was judged as lower in quality than their noncommissioned work. People are likely to be more creative when they are intrinsically motivated in relation to the task, not compelled by outside forces.

Looking back over this research, Beth Hennessey (2015) notes that “[i]f I were secretary of education,” she would focus on ways to enhance intrinsic motivation in the classroom—as opposed to the use of grades, testing standards, competition, and other extrinsic means of motivating students.

But wait. If extrinsic benefits serve to undermine intrinsic motivation, should teachers and parents not offer rewards to their children? Are the employee incen- tive programs that are so often used to motivate workers in the business world doomed to fail, as some (Kohn, 1993) have suggested? This turns out to be a complex question that depends on how the reward is perceived and by whom. If a reward is presented in the form of verbal praise that is perceived to be sincere or as a special “bonus” for superior performance, then it can enhance intrinsic motivation by providing positive feedback about competence—as when people win competitions, scholarships, or a pat on the back from people they respect (Cameron & Pierce, 1994; Cameron et al., 2005; Eisenberger & Cameron, 1996; Henderlong & Lepper, 2002).

The notion that intrinsic motivation is undermined by some types of reward but not others was observed even among 20-month-old babies. In a clever study, Felix Warneken and Michael Tomasello (2006) brought babies into a lab, where the experimenter accidentally dropped a pen or crumpled paper onto the floor and appeared unable to reach it. The child could help by picking up the object and handing it to the experimenter. Most of the babies helped in this situation. In a treatment phase, the researchers responded to the assistance by giving the child a toy cube (“For this you get a cube”), verbal praise (“Thank you, that’s really nice!”), or nothing at all. Would these same children continue to help? Re- sults showed that in a later test phase, when presented with a number of helping opportunities, those in the no-response condition continued to help 89% of the time and that this tendency remained high at 81% in the verbal praise condition. Yet among children who had earlier received a reward, helping in the test phase dropped to 53% when that reward was no longer available (see also Warneken & Tomasello, 2014).

Individual differences in people’s motivational orientation toward work must also be considered. For intrinsically oriented people who say, “What matters most to me is enjoying what I do” and “I seldom think about salary and promo- tions,” reward may be unnecessary and may even be detrimental (Amabile et al., 1994). Yet for people who are laser-focused on the achievement of certain goals— whether at school, at work, or in sports—inducements such as grades, scores, bonuses, awards, trophies, and the sheer thrill of competition, as in team sports, tend to boost intrinsic motivation (Durik & Harackiewicz, 2007; Harackiewicz & Elliot, 1993).

j Influences of Other People As noted earlier, Cooley’s (1902) theory of the looking-glass self emphasized that other people help us define ourselves. In this section, we will see the importance of this proposition to our self-concepts.

Social Comparison Theory Suppose a stranger were to ask, “Who are you?” If you had only a minute or two to answer, would you mention your religion or your

80220_ch03_ptg01_hr_052-103.indd 65 27/01/16 11:21 AM

66 Chapter 3 The Social Self

ethnic background? What about your hometown? Would you describe your talents and your interests or your likes and dislikes? When asked this question, people tend to describe themselves in ways that set them apart from others in their immediate vicinity (McGuire & McGuire, 1988). Among children, boys are more likely to cite their gender when they grow up in families that are predominantly female; girls do the same when living in families that are predominantly male (McGuire et al., 1979). Similarly, on the college campus, “nontraditional” older students are more likely to cite their age than are traditional younger students (Kite, 1992). Regardless of whether the unique attribute is gender, age, height, or eye color, this pattern is basi- cally the same. The implication is intriguing: Change someone’s social surroundings, and you can change that person’s spontaneous self-description.

This reliance on distinguishing features in self-description indicates that the self is “relative,” a social construct, and that each of us defines ourselves in part by using family members, friends, acquaintances, and others as a benchmark (Mussweiler & Rüter, 2003; Mussweiler & Strack, 2000). Importantly, the self is also “malleable” according to our need to fit in with those around us. In an article entitled “Reaching Out by Changing What’s Within,” Stephanie Richman and her colleagues (2015) reported on a series of studies showing that when college stu- dents are induced to suffer through a social exclusion experience—being left out of an online three-person game—they go on to modify their self-concept descrip- tions (for example, on such traits as warm, adventurous, creative, enthusiastic, thoughtful, and funny) in ways that make them more similar to a fellow student who looms as a potential friend.

Enter Leon Festinger’s (1954) social comparison theory. Festinger argued that when people are uncertain of their abilities or opinions—that is, when objective information is not readily available—they evaluate themselves through compari- sons with similar others. The theory seems reasonable, but is it valid? Over the years, social psychologists have put social comparison theory to the test, focusing on two key questions: (1) When do people turn to others for comparative infor- mation? (2) Of all the people who inhabit the Earth, with whom do we choose to compare ourselves? (Suls & Wheeler, 2000).

As Festinger proposed, the answer to the “when” question appears to be that people engage in social comparison in states of uncertainty, when more objective means of self-evaluation are not available. It’s not clear whether Festinger under- stated the importance of social comparison processes. Some research suggests that people judge themselves in relation to others even when more objective standards really are available (Klein, 1997). Yet other research supports Festinger’s theory that people are less influenced by social comparisons when objective information is available—for example, through our personal histories of success and failure (Steyn & Mynhardt, 2008).

The “with whom” question has also been the subject of many studies. The answer seems to be that when we evaluate our own taste in music, value on the job market, or athletic ability, we look to others who are similar to us in relevant ways (Goethals & Darley, 1977; Wheeler et al., 1982). If you are curious about your flair for writing, for example, you’re more likely to compare yourself with other college students than with high schoolers or best-selling authors. There are exceptions to this rule, of course. Later in this chapter, we will see that people often cope with personal inadequacies by focusing on others who are less able or less fortunate than themselves.

Facebook as a Venue for Social Comparison Currently, Facebook—the most heavily populated social networking site—has 1.5 billion active users worldwide

social comparison theory The theory that people evaluate their own abilities and opinions by comparing themselves to others.

80220_ch03_ptg01_hr_052-103.indd 66 27/01/16 11:21 AM

67 The Self-Concept

(Facebook, 2015). On computers, tablets, and mobile phones, more than a billion people a day log into their Facebook accounts, Twitter, LinkedIn, Pinterest, Tumblr, and other sites. Two types of usage can be distinguished: active usage, where peo- ple post information about themselves and communicate with others; and passive usage, in which people consume information from other people’s Facebook pages without making direct contact (Deters & Mehl, 2013; Verduyn et al., 2015).

Now that social networking sites enable us to access countless numbers of people, what effect do all the social comparison opportunities available to us have on our self-concepts, our self-evaluations, and our overall well-being? Does looking at other people’s Facebook pages make you feel better about yourself, or worse, or does it depend on whose pages you visit and how they present them- selves? At first, research was reported in the news suggesting a phenomenon that was being called “Facebook Depression”—the more time people spent on Face- book, the more unhappy they were (Feinstein et al., 2013; Kalpidou, Costin, & Morris, 2011; O’Keeffe & Clarke-Pearson, 2011). Immediately, social psychologists were quick to warn that this correlation should not be interpreted to mean that Facebook usage causes depression.

There are two reasons why Facebook usage may undermine a person’s well- being. First, as predicted by Festinger’s social comparison theory, recent studies have shown that the link between Facebook usage and self-evaluation depends on whom we compare ourselves to. After college-age adults were randomly assigned to engage in upward—as opposed to downward—social comparisons, with others who are highly active and successful, they came to rate themselves less favorably (Vogel et al., 2014). A second possible reason for this negative effect is that people on Facebook, as in life more generally, tend to portray themselves in overly flat- tering ways—which increases the likelihood that the social comparisons we make are not personally favorable. For that reason, research shows that the more Face- book time that people passively scroll through other people’s pages—rather than directly interacting with others—the worse they felt about themselves (Verduyn et al., 2015).

Two-Factor Theory of emotion People seek social comparison information to evaluate their abilities and opinions. Do they also turn to others to determine something as subjective as their own emotions? In classic experiments on affiliation, Stanley Schachter (1959) found that when people were frightened into thinking they would receive painful electric shocks, most sought the company of others who were in the same predicament. Nervous and uncertain about how they should be feeling, participants wanted to affiliate with similar others, presumably for the purpose of comparison. Yet when they were not fearful and expected only mild shocks or when the “others” were not taking part in the same experiment, participants preferred to be alone. As Schachter put it, “Misery doesn’t just love any kind of company; it loves only miserable company” (p. 24).

Intrigued by the possibilities, Schachter and his research team took the next step. Could it be, they wondered, that when people are uncertain about how they feel, their emotional state is actually determined by the reactions of oth- ers around them? In answer to this question, the researchers proposed that two factors are necessary to feel a specific emotion. First, the person must experi- ence the symptoms of physiological arousal—such as a racing heart, perspira- tion, rapid breathing, and tightening of the stomach. Second, the person must make a cognitive interpretation that explains the source of the arousal. And that is where the people around us come in: Their reactions help us interpret our own arousal.

80220_ch03_ptg01_hr_052-103.indd 67 27/01/16 11:21 AM

68 Chapter 3 The Social Self

To test this provocative two-factor theory of emotion, Schachter and Singer (1962) injected male volunteers with epinephrine, a drug that heightens physi- ological arousal. Although one group was forewarned about the drug’s effects, a second group was not. Members of a third group were injected with a harmless placebo. Before the drug (which was described as a vitamin supplement) actually took effect, participants were left alone with a male confederate introduced as another participant who had received the same injection. In some sessions, the confederate behaved in a euphoric manner. For 20 minutes, he bounced around happily, doodling on scratch paper, sinking jump shots into the wastebasket, fly- ing paper airplanes across the room, and playing with a Hula-Hoop. In other ses- sions, the confederate displayed anger, making fun of a questionnaire they were filling out and, in a fit of rage, ripping it up and hurling it into the wastebasket.

Think for a moment about these various combinations of situations. As the drug takes effect, participants in the drug-informed group will begin to feel their hearts pound, their hands shake, and their faces flush. Having been told to expect these symptoms, however, they need not search for an explanation. Participants in the placebo group will not become aroused in the first place, so they will have no symptoms to explain. But now consider the plight of those in the drug-uninformed group, who suddenly become aroused without knowing why. Trying to identify the sensations, these participants, according to the theory, should take their cues from someone else in the same predicament—namely, the confederate.

In general, the experimental results supported Schachter and Singer’s line of reasoning. Drug-uninformed participants reported feeling relatively happy or angry depending on the confederate’s performance. In many instances, they even exhibited similar kinds of behavior. One participant, for example, “threw open the window and, laughing, hurled paper basketballs at passersby.” In the drug- informed and placebo groups, however, participants were, as expected, less influ- enced by these social cues.

Schachter and Singer’s two-factor theory attracted controversy when some studies corroborated their findings but others did not. In one experiment, for example, participants who were injected with epinephrine and not forewarned about the symptoms later exhibited more fear in response to a scary film, but they were not more angry or amused while seeing films that tend to elicit these other emotions (Mezzacappa et al., 1999). It now appears that one limited but important conclusion can safely be drawn: When people are unclear about their own emotional states, they sometimes interpret how they feel by watching oth- ers (Reisenzein, 1983). The “sometimes” part of the conclusion is important. For other people to influence your emotion your level of physiological arousal cannot be too intense or else it will be experienced as aversive, regardless of the situation (Maslach, 1979; Zimbardo et al., 1993). Also, research shows that other people must be present as a possible explanation for arousal before its onset. Once people are aroused, they turn for an explanation to events that preceded the change in their physiological state (Schachter & Singer, 1979; Sinclair et al., 1994).

In subsequent chapters, we will see that the two-factor theory of emotion has far-reaching implications for passionate love, anger and aggression, and other affective experiences.

j Autobiographical Memories Philosopher James Mill once said, “The phenomenon of the Self and that of Memory are merely two sides of the same fact.” If the story of patient William Thompson at the start of this chapter is any indication, Mill was right. Without

two-factor theory of emotion The theory that the experience of emotion is based on two factors: physiological arousal and a cognitive interpretation of that arousal.

80220_ch03_ptg01_hr_052-103.indd 68 27/01/16 11:21 AM

69 The Self-Concept

autobiographical memories—recollections of the sequences of events that have touched your life (Bernsten & Rubin, 2012; Fivush & Haden, 2003; Rubin, 1996)— you would have no coherent self-concept. After all, who would you be if you could not remember your parents or childhood friends, your successes and fail- ures, the places you lived, the schools you attended, the books you read, and the teams you played for? Clearly, memories shape the self-concept. In this section, we’ll see that the self-concept shapes our personal memories as well (Conway & Pleydell-Pearce, 2000).

When people are prompted to recall their own experiences, they typically report more events from the recent past than from the distant past. There are, however, two consistent exceptions to this simple recency rule. The first is that older adults tend to retrieve a large number of personal memories from their adolescence and early adult years—a “reminiscence bump” found across many cultures that may occur because these are busy and formative years in one’s life (Conway et al., 2005; Fitzgerald, 1988; Jansari & Parkin, 1996). A second exception is that people tend to remember transitional “firsts.” Reflect for a moment on your own college career. What events pop to mind— and when did they occur? Did you come up with the day you arrived on campus or the first time you met your closest friend? What about notable classes, parties, or sports events? When David Pillemer and his colleagues (1996) asked college juniors and seniors to recount the most memorable experiences of their first year, 32% of all recollections were from the tran- sitional month of September. When college graduates were given the same task, they too cited a disproportionate number of events from the opening two months of their first year, followed by the next major transitional period, the last month of their senior year. Among students, these busy transitional periods are important regardless of whether their schools follow a semester calendar or some other academic schedule (Kurbat et al., 1998).

Obviously, not all experiences leave the same impression. Ask people old enough to remember November 22, 1963, and to this day they probably can tell you exactly where they were, who they were with, what they were thinking, and what was happening the moment they heard the news that President John F. Kennedy had been shot. Ask anyone old enough to remember the terrorist attacks of September 11, 2001, and they too could probably recount exquisite levels of detail about the event and their personal relation to it. Roger Brown and James Kulik (1977) coined the term flashbulb memories to describe these enduring, detailed, high-resolution recollections and speculated that humans are biologically equipped for survival purposes to “print” dramatic events in memory. These flashbulb memories are not necessarily accurate or even con- sistent over time. Still, these recollections “feel” special and serve as promi- nent landmarks in the biographies that we tell about ourselves (Conway, 1995; Luminet & Curci, 2009).

By linking the present to the past and providing us with a sense of inner continuity over time, autobiographical memory is a vital part of—and can be shaped by—our life story and sense of identity. The links between our sense of self and autobiographical memory are subtle, complex, and often not straightfor- ward (Prebble et al., 2013).

One complicating factor is that people tend to distort the past in ways that in- flate their own sense of importance and achievement. In one study, Harry Bahrick and others (1996) had college students try to recall all of their high school grades and then checked their reports against the actual transcripts. Overall, the majority of grades were recalled correctly. But most of the errors in memory were grade

“The nice thing about having memories is that you can choose.”

—William Trevor

Although adults recall more events from the recent than distant past, people are filled with memories from late adolescence and early adulthood. These formative years are nicely captured by high school photos, such as those of American actresses Angelina Jolie (top) and Jennifer Aniston (bottom).

Lo re

nz o

Ci ni

gl io

/S yg

m a/

Co rb

is Fo

to s

In te

rn at

io na

l/ G

et ty

Im ag

80220_ch03_ptg01_hr_052-103.indd 69 27/01/16 11:21 AM

70 Chapter 3 The Social Self

inflations—and most of these were made when the actual grades were low (see d Figure 3.3).

In a second study, Burcu Demiray and Steve Janssen (2015) asked hundreds of adults, 18 to 80 years old, to report the seven most impor- tant events from their lives and to rate these memories for how good, important, emotional, vivid, and close they were. The results showed that respondents felt psychologically “closer” to memories that were positive rather than negative. Suggesting that this bias serves an adap- tive purpose, the results also showed that this tendency that was par- ticularly evident in people who evaluate themselves favorably. These findings bring to mind George Herbert Mead’s (1934) suggestion that our visions of the past are like pure “escape fancies . . . in which we rebuild the world according to our hearts’ desires” (pp. 348–349). Or, as Anthony Greenwald (1980) put it: “The past is remembered as if it were a drama in which the self was the leading player” (p. 604).

Our autobiographies are so interconnected with our sense of who were are that as our self-concept changes, so does our visual perspective on the past. Think about an important way in which you have changed. Once you were a kid, now you are in college or working. Or maybe you were a smoker and stopped, or obese and lost weight. Or maybe you underwent a religious conversion, or had cancer and survived it—and now you feel “reborn.” Theorizing that our current self-concept col- ors how we see our past selves, Lisa Libby and Richard Eibach (2002) asked college students to write about one aspect of themselves that had changed a lot and another that had not changed since high school. Ana- lyzing the words used to describe these recollections, these researchers found that participants used more third-person pronouns to describe past actions that no longer fit their current selves—and they rated them- selves as more detached from these actions.

Finally, it is interesting to note that just as the contents of our auto- biographical memories are intertwined with our sense of who we are, the process of remembering can prove to be a positive emotional expe- rience. Have you ever lost yourself in a daydream, thinking back to a childhood vacation, a graduation, the time you spent at a sports camp, or the day you met a good friend? Nostalgia—defined as a sentimen-

tal longing for the past—is common and universal. Research shows that people often become nostalgic during distressing life events such as a breakup or di- vorce, a long distance move, feelings of loneliness, or serious illness (Wildschut et al., 2006). Research also shows that the effect of making people nostalgic is not merely inform or reinforce the self-concept but to boost their self-esteem and positive mood states, instill a sense that life is meaningful and worth living, and increase optimism about the future (Baldwin et al., 2015; Baldwin & Landau, 2014; Cheung et al., 2013; Routledge et al., 2011).

j Culture and the Self-Concept The self-concept is also heavily influenced by cultural factors. In America, it is said that “the squeaky wheel gets the grease”; in Japan, it is said that “the nail that stands out gets pounded down.” Thus, American parents try to raise their children to be independent, self-reliant, and assertive (a “cut above the rest”), whereas Japanese children are raised to fit into their groups and community.

d F I g u R e 3 . 3

Distortions in Memory of High School grades

College students were asked to recall their high school grades, which were then checked against their actual transcripts. These comparisons revealed that most errors in memory were grade inflations. Lower grades were recalled with the least accuracy (and the most inflation). It appears that people sometimes revise their own past to suit their current self-image. Based on Bahrick et al., “Accuracy and distortion in memory for high school grades,” Psychological Science vol 7 (pp. 265–271).

100

0 A B

Pe rc

en ta

ge o

f a cc

ur at

el y

re ca

lle d

gr ad

Grade

C D

80220_ch03_ptg01_hr_052-103.indd 70 27/01/16 11:21 AM

71 The Self-Concept

Differences in Cultural Orientation The preceding example illustrates two contrasting cultural orientations. One values individualism and the virtues of inde- pendence, autonomy, and self-reliance. The other orientation values collectivism and the virtues of interdependence, cooperation, and social harmony. Under the banner of individualism, one’s personal goals take priority over group allegiances. In collec- tivist cultures, by contrast, a person is first and foremost a loyal member of a family, team, company, church, and state, motivated to be part of a group—not different, better, or worse (Triandis, 1994). In what countries are these orientations the most extreme? In a worldwide study of 116,000 employees of IBM, Geert Hofstede (1980) found that the most fiercely individualistic people were from the United States, Australia, Great Britain, Canada, and the Netherlands—in that order. The most col- lectivist people were from Venezuela, Colombia, Pakistan, Peru, Taiwan, and China.

Individualism and collectivism are not opposites on a continuum; the similari- ties and differences between countries do not fit a simple pattern. Daphna Oyser- man and others (2002) conducted a meta-analysis of thousands of respondents in 83 studies. Within the United States, they found that African Americans were the most individualistic subgroup and that Asian Americans and Latino Americans were the most collectivistic. Comparing nations, they found that collectivist orien- tations varied within Asia, as the Chinese were more collectivistic than Japanese and Korean respondents.

On average, all humans are 99.5% similar to other humans. What is it about cultures, therefore, that creates such different orienta- tions among people? In Clash!: How to Thrive in a Multicultural World, Hazel Rose Markus and Alan Conner (2013) describe the conflicts that often arise between groups all over the world— East vs. West, rich vs. poor, urban vs. rural, coastal vs. heartland, and whites vs. people of color, to name a few. They note that national boundaries are not the only source of cultural differences, that each of us combines a special mix of biology and cultures to make us who we are. According to Markus and Conner, culture is made up of four I’s—ideas, institutions, and social interactions that shape how individuals think, feel, and act. In turn, how individuals act influences their ideas, institutions, and so- cial interactions. This dynamic culture cycle is depicted in d Figure 3.4.

Cultural Influences on the Self Individual- ism and collectivism are so deeply ingrained in a culture that they mold our very self-conceptions and identities. According to Hazel Markus and Shinobu Kitayama (1991), most North Ameri- cans and Europeans have an independent view of the self. In this view, the self is an entity that is distinct, autonomous, self-contained, and en- dowed with unique dispositions. Yet in much of Asia, Africa, and Latin America, people hold an

d F I g u R e 3 . 4

The Culture Cycle

The culture cycle shows that individuals are shaped by their interactions with others, by formal institutions, and by commonly shared ideas of what is a good and right way to be a person. In turn, through their actions and behaviors, individuals shape these aspects of their own world. Based on Markus, H.R., & Conner, A. (2013). Clash!: 8 cultural conflicts that make us who we are. New York, NY: Hudson Street Press.

Individuals

Interactions

Institutions

Ideas

(e.g. Thoughts, Feelings and Behavior)

(e.g. Family, Home Workplace, School)

(e.g. Goverment, Media, Sports,

Science, Education)

(e.g. What is good, Right, Natural)

80220_ch03_ptg01_hr_052-103.indd 71 27/01/16 11:21 AM

72 Chapter 3 The Social Self

interdependent view of the self. Here, the self is part of a larger network that includes one’s family, co-workers, and others with whom one is socially connected. People with an independent view say that “the only person you can count on is yourself” and “I enjoy being unique and different from others.” In contrast, those with an interdependent view are more likely to agree that “I’m partly to blame if one of my family members or co-workers fails” and “My happiness depends on the happiness of those around me” (Rhee et al., 1995; Triandis et al., 1998). These contrasting orientations—one focused on the personal self, the other on a collective self—are depicted in d Figure 3.5.

Research of various sorts confirms the close link between cultural orientation and conceptions of the self. In one study, David Trafimow and others (1991) had North American and Chinese college students complete 20 sentences beginning with “I am. . . .” Americans were more likely to fill in the blank with trait descrip- tions (“I am shy”), whereas the Chinese were more likely to identify themselves by group affiliations (“I am a college student”). Consistent with this finding, a sec- ond study has shown that when Chinese participants—but not Americans—think about themselves areas of the brain are activated that are also activated when they think about their mothers (Zhu et al., 2007b). A third study has shown that when it comes to making career decisions, Chinese students with interdependent selves are more likely to seek advice from others and compromise than are American students with independent selves (Guan et al., 2015).

Our cultural orientations can color the way people perceive, evaluate, and present themselves in relation to others. In this regard, Markus and Kitayama (1991) identified two interesting differences between East and West: The first is that people in individualistic cultures strive for personal achievement, whereas those living in collectivist cultures derive more satisfaction from the status of a

d F I g u R e 3 . 5

Cultural Conceptions of Self

As depicted here, different cultures foster different conceptions of the self. Many Westerners have an independent view of the self as an entity that is distinct, autonomous, and self-contained. Yet many Asians, Africans, and Latin Americans hold an interdependent view of the self that encompasses others in a larger social network. Based on Marcus, H. R. and Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98, 224–253.

Co-worker

Friend

A. Independent View of Self B. Interdependent View of Self

SiblingSibling

Mother

Mother Father

Co-worker

Friend

Friend Friend

SelfSelf

Father

80220_ch03_ptg01_hr_052-103.indd 72 27/01/16 11:21 AM

73 The Self-Concept

valued group. Thus, whereas North Americans tend to overestimate their own contributions to a team effort, seize credit for success, and blame others for failure, people from collectivist cultures tend to un- derestimate their own role and present themselves in more modest, self-effacing terms in relation to other members of the group (Heine et al., 2000).

A second consequence of these differing conceptions of the self is that American college students see themselves as less similar to other people than do Asian students. This difference reinforces the idea that individuals with independent self-conceptions believe they are unique. In fact, our cultural orientations toward conformity or independence may lead us to favor similarity or uniqueness in all things. In a study that illustrates the point, Heejung Kim and Hazel Markus (1999) showed abstract figures to subjects from the United States and Korea. Each figure contained nine parts. Most of the parts were identical in shape, position, and direction. One or more were different. Look at d Figure 3.6. Which of the nine subfigures within each group do you like most? The American subjects liked the sub- figures that were unique or in the minority, while Korean subjects preferred those that “fit in” as part of the group. In another study, these same researchers approached pedestrians of American and East Asian heritage at San Francisco’s airport to fill out a question- naire. Afterward, as a gift, they offered the participants a choice of one pen from a handful of pens, three or four of which had the same color barrel, green or orange. The result: 74% of the Americans chose a uniquely colored pen, and 76% of the East Asians selected a commonly colored pen! It seems that culturally ingrained orienta- tions to conformity and independence leave a mark on us, leading us to form preferences for things that “fit in” or “stand out.”

Are people from disparate cultures locked into thinking about the self in either personal or collective terms, or are both aspects present in everyone, to be expressed according to the situation? Reconsider the study noted earlier, where American students described themselves more in terms of personal traits and Chi- nese students cited more group affiliations. In a follow-up to that study, Trafimow and others (1997) tested students from Hong Kong, all of whom spoke English as a second language. One half of the students were given the “Who am I?” test in Chinese, and the other half took the test in English. Did this variation influence the results? Yes. Students who took the test in English focused more on personal traits, whereas those who took the test in Chinese focused more on group affilia- tions. It appears that each of us have both personal and collective aspects of the self to draw on—and that the part that comes to mind depends on the situation we are in.

The more closely social psychologists examine cultures and their impact on how people think, the more complex is the picture that emerges. Clearly, research documents the extent to which self-conceptions are influenced by the individual- ist and collectivist impulses within a culture. But there are other core differences as well. Kaiping Peng and Richard Nisbett (1999) note that people in East Asian cultures think in dialectical terms about contradictory characteristics—accepting, for example, that apparent opposites (such as black and white, friend and enemy, and strong and weak) can coexist within a single person either at the same time or as a result of changes over time. Grounded in Eastern traditions, dialecticism is a system of thought characterized by the acceptance of such contradictions

d F I g u R e 3 . 6

What’s Your Preference: Similarity or uniqueness?

Which subfigure within each set do you prefer? Kim and Markus (1999) found that Americans tend to like subfigures that “stand out” as unique or in the minority, while Koreans tend to like subfigures that “fit in” with the surrounding group. From Kim, H. and Marcus, H. R., “Deviance or unique- ness, harmony or conformity? A cultural analysis,” Journal of Personality and Social Psychology vol 77 (pp. 785–800). Copyright © 1999 by the American Psychological Association. Reprinted by permission.

dialecticism An Eastern system of thought that accepts the coexistence of contradictory characteristics within a single person.

80220_ch03_ptg01_hr_052-103.indd 73 27/01/16 11:21 AM

74 Chapter 3 The Social Self

through compromise, as implied by the Chinese proverb “Beware of your friends, not your enemies.” This thought style contrasts sharply with the American and European perspective, grounded in Western logic, by which people differentiate seeming opposites on the assumption that if one is right, the other must be wrong.

Wondering if a dialectical style of thought has implications for the self, Tammy English and Serena Chen (2007) conducted a series of studies in which they questioned American college students who were of European or Asian descent about what kind of person they are in such different everyday situations as a classroom, a cafeteria, a party, or the gym. Overall, they found that compared to European Americans who portray their “true selves” as stable across the board, Asian Americans vary their self-concepts more to suit different relationship situations—though they are consistent within these situa- tions. Other research too has shown that East Asians are more willing than Americans to see and accept contradic- tory aspects of themselves (Spencer-Rodgers et al., 2009)— as seen in their willingness to accept both positive and negative aspects of themselves at the same time (Boucher et al., 2009).

The study of cultural aspects of the self is also ex- panded by social psychologists interested in Latin Ameri- can cultures, where social and emotional relationships are a very important part of the collectivist orientation.

According to Renee Holloway and others (2009), Latino cultures prize the con- cept of simpatico, which emphasizes expressive displays of personable charm, graciousness, and hospitality. Does this cultural value become part of the Latino self-concept? Clearly, no two individuals are the same. But when these research- ers presented Latino and white Americans with the “Who am I?” task described earlier, they found that the Latino participants on average were more likely to describe themselves using simpatico-related terms such as likable, friendly, sym- pathetic, amiable, and gracious. Similarly, Nairán Ramírez-Esparza and colleagues (2012) found that Mexican participants were more likely to describe their own personality using words about relationships (e.g., parents, house, love, friends) and empathy (e.g., affectionate, honest, noble, tolerant).

Finally, it is important to realize that cultures themselves change over time, from one generation to another. Patricia Greenfield (2013) looked at the frequency of word usage from the year 1800 until 2000. She found that accompanying a shift from a more rural to urban population, there was also a shift in cultural values. As the occurrence of words such as “duty,” “obliged,” “give,” “obedience,” “au- thority,” “belong,” and “benevolence” decreased over time, the words “decision,” “choose,” “unique,” “individual,” “self,” and “acquisition” increased over time.

Generational changes within a culture can also be seen within smaller frames of time. Americans who were children in the 1940s and 1950s, a generation known as “baby boomers,” grew up in a very different culture than those who grew up in the 1960s and 1970s, a group commonly referred to as “GenX’ers,” and those who grew up in the 1980s and 1990s—a group known as “Millennials.” Analyz- ing questionnaire data collected from 9.2 million American high school seniors and college students from 1966 to 2009, Jean Twenge and her colleagues (2012) found that compared to boomers, subsequent generations were more focused on

Reflecting an independent view of the self, children in Japan are taught to fit into the community. Reflecting a more independent view of the self, children in the United States are encouraged to express their individuality.

BL O

O M

im ag

es /G

et ty

Im ag

es St

ef an

o Lu

na rd

i/ Ve

er .c

80220_ch03_ptg01_hr_052-103.indd 74 27/01/16 11:21 AM

75 The Self-Concept

money, fame, and self-image and less concerned with affiliation, community, and civic engagement (see d Figure 3.7). This change in values—often described as a shift from “Generation We” to “Generation Me”—suggests that American culture is more individualistic today than it was a half century ago.

Social Class as a Cultural Influence In all cultures of the world, individuals differ in their wealth, material possessions, education, and level of prestige. In a con- tinuum that ranges from people who live in poverty up through multibillionaires, the term social class is used to categorize people within a culture who have in common a low-, working-, middle-, or upper-class socioeconomic status.

Social class is another cultural factor that can influence the self-concept. In Western countries, people with more income, education, and status tend to have many opportunities to exhibit individualism by expressing their desires, their autonomy, and the pursuit of personal goals. They have more control over the lives, greater personal choice, and more independence and self-focus. In contrast, people with less income, education, and status are more constrained in terms of what they can and cannot do. Navigating a low-income world means having to rely more on others and fitting-in, fostering “hard interdependence” (Fiske & Markus, 2012; Kraus et al., 2012; Stephens et al., 2014). Recent studies confirm

d F I g u R e 3 . 7

generational Differences in American Cultural Orientation

Over the years, surveys have been administered to millions of Americans who grew up in the 1940s–1950s (“baby boomers”), 1960s–1970s (“GenX’ers”), and 1980s–1990s (“Millennials”). From the responses of high school and college students from 1966 to 2009, it seems that compared to boomers, later generations were more focused on money, fame, and self-image, and less concerned with affiliation, community, and civic engagement. These changes suggest that American culture is more individualistic today than it was a half century ago. Based on Twenge, J. M., Campbell, W. K., & Freeman, E. C. (2012). Generational differences in young adults’ life goals, concern for others, and civic orientation, 1966–009, Journal of Personality and Social Psychology, 102, 1045–062.

100

Being a leaderMoneyWell-off financially

19691966 1972 1975 1978 1981 1984 1987 19961990 1993 2005 20081999 2002

Philosophy of life

80220_ch03_ptg01_hr_052-103.indd 75 27/01/16 11:21 AM

76 Chapter 3 The Social Self

that the characteristics of the self that are associated with social class. In one study, for example, working-class men were more likely than middle-class men to see themselves in terms of their relationships to others (Markus & Conner, 2013). In a second study, people classified as low in social class were less likely to agree with statements of entitlement such as “I honestly feel I’m just more deserving than others” and with statements indicating narcissism such as “I like to look at myself in the mirror” (Piff, 2015).

How do you feel about yourself? Are you generally satisfied with the way you look, your personality, academic and athletic abilities, your accomplishments, and your friendships? Are you optimistic about your future? When it comes to the self, people are not cool, objective, dispassionate observers. Rather, we are judgmental, motivated, emotional, and highly protective of our self-esteem—an evaluative component of the self.

The word esteem comes from the Latin aestimare, which means “to estimate or appraise.” The term self-esteem thus refers to our positive and negative evalu- ations of ourselves (Coopersmith, 1967). Some individuals have a higher self- esteem than others do—an attribute that can have a profound impact on the way they think about, feel about, and present themselves. It’s important to keep in mind that although some of us have higher self-esteem than others, a feeling of self-worth is not a single trait etched permanently in stone. Rather, it is a state of mind that fluctuates up and down in response to success, failure, social relations, and other life experiences (Heatherton & Polivy, 1991).

With the self-concept made up of many self-schemas, people typically view parts of the self differently: Some parts they judge more favorably or see more clearly or as more important than other parts (Pelham, 1995). In fact, just as individuals differ ac- cording to how high or low their self-esteem is, they also differ in the extent to which their self-esteem is stable or unstable. As a general rule, self-esteem stays roughly the same from childhood through old age—people who are high or low in self-esteem remain in that relative position throughout life (Trzesniewski et al., 2003).

Although an individual’s self-esteem relative to others remains stable, the aver- age level of self-esteem in a population varies over the course of a lifetime. Longitudi- nal studies that track individuals over time, and cross-sectional studies that compare at one point in time people from different age groups, show that self-esteem declines from childhood to adolescence, gradually increases during the transition to adult- hood, continues to rise as adults get older, and declines in old age (Erol & Orth, 2011; Orth et al., 2010). Research with college students shows that average levels of self- esteem vary even over the course of a college career. Think for a moment about the whole college experience. What changes in self-esteem might you predict? In a lon- gitudinal study, 295 male and female students had their self-esteem tested six times over a four-year period. d Figure 3.8 shows that average levels dropped sharply dur- ing the very first semester, rebounded by the end of the first year, and then gradually increased from that point on (Chung et al., 2014).

j The Need for Self-esteem You, me, and just about everyone else on the planet seem to have a need for self- esteem, wanting to see ourselves in a positive light. As a result of who we are

Self-Esteem

self-esteem An affective component of the self, consisting of a person’s positive and negative self-evaluations.

80220_ch03_ptg01_hr_052-103.indd 76 27/01/16 11:21 AM

77 Self-Esteem

and the culture we live in, each of us may value different attributes and pursue this need in different ways. Some people derive a sense of worth from their ap- pearance; others value physical strength, professional accomplishments, wealth, people skills, or group affiliations. Whatever the source, it is clear that the pursuit of self-worth is an aspect of human motivation that runs deep. But let’s step back for a moment and ask, why? Why do we seem to need self-esteem the way we need food, air, sleep, and water?

At present, there are two social psychological answers to this question. Socio- meter Theory, proposed by Mark Leary and Roy Baumeister (2000), maintains that people are inherently social animals and that the desire for self-esteem is driven by a more primitive need to connect with others and gain their approval. As a result of this social connection for survival, people may have evolved a “sociometer”—a mechanism that enables us to detect acceptance and rejection and then translate these perceptions into high and low self-esteem. In this way, self-esteem serves as a rough indicator of how we’re doing in the eyes of others. The threat of social rejection thus lowers self-esteem, which activates the need to regain approval and acceptance. In an experiment aimed at identifying a “neural sociometer,” participants underwent fMRI while viewing positive and negative feedback words that a confederate ostensibly used to describe them (words such as interesting and boring). Participants also rated their self-es- teem in response to each feedback word. Results showed that increased activity in rejection-related brain regions was associated with lowered self-esteem (Eisenberger et al., 2011).

There is a second particularly important answer to this question. Jeff Greenberg, Sheldon Solomon, and Thomas Pyszczynski (1997) proposed Terror Management Theory to help explain our relentless need for self-esteem. According to this pro- vocative and influential theory, we humans are biologically programmed for life and

d F I g u R e 3 . 8

Self-esteem Over the Course of a College Career

In this longitudinal study of college students who were repeatedly tested, self-esteem dropped sharply during their first semester but then recovered and increased into their senior year. Chung et al., 2014 Chung, J. M., Robins, R. W., Trzesniewski, K. H., Noftle, E. E., Roberts, B. W., & Widaman, K. F. (2014). Continuity and change in self-esteem during emerging adulthood. Journal of Personality and Social Psychology, 106, 469-483.

3.3

3.9

3.8

3.7

3.6

3.5

3.4

Week 1 Year 1Semester 1 Year 4Year 2 Year 3

Le ve

l o f

Se lf-

Es te

Sociometer Theory The theory that self-esteem is a gauge that monitors our social interactions and sends us signals as to whether our behavior is acceptable to others.

Terror Management Theory The theory that humans cope with the fear of their own death by constructing worldviews that help to preserve their self-esteem.

80220_ch03_ptg01_hr_052-103.indd 77 27/01/16 11:21 AM

78 Chapter 3 The Social Self

self-preservation. Yet we are conscious of—and terrified by— the inevitability of our own death. To cope with this paralyzing, deeply rooted fear, we construct and accept cultural worldviews about how, why, and by whom the Earth was created; religious explanations of the purpose of our existence; and a sense of his- tory filled with heroes, villains, and momentous events. These worldviews provide meaning and purpose and a buffer against anxiety. In a series of experiments, these investigators found that people react to graphic scenes of death or to the thought of their own death with intense defensiveness and anxiety. When peo- ple are given positive feedback on a test, however, which boosts their self-esteem, that reaction is muted. Other research has since confirmed this type of result (Schmeichel et al., 2009). As we’ll see in later chapters, Terror Management Theory has been used to explain how people cope with traumas such as the ter- rorist attack of 9/11 (Pyszczynski et al., 2002) and why people all over the world seek solace in religion, strong leaders, and

cultural institutions (Pyszczynski et al., 2015). As for the need for self-esteem, Pyszczynski and his colleagues (2004) put it

this way:

Self-esteem is a protective shield designed to control the potential for terror that results from awareness of the horrifying possibility that we humans are merely transient animals groping to survive in a meaningless universe, designed only to die and decay. From this perspective, each individual human’s name and iden- tity, family and social identifications, goals and aspirations, occupation and title, and humanly created adornments are draped over an animal that, in the cosmic scheme of things, may be no more significant or enduring than any individual potato, pineapple, or porcupine. (p. 436)

Confirming folk wisdom, a good deal of research shows that high and low self-esteem can color our outlook on life. People with positive self-images tend to be happy, healthy, productive, and successful. They also tend to be confi- dent, bringing to new challenges a winning attitude that leads them to persist longer at difficult tasks, sleep better at night, maintain their independence in the face of peer pressure, and suffer from fewer ulcers. In contrast, people with negative self-images tend to be more depressed, pessimistic about the future, and prone to failure. Lacking confidence, they bring to new tasks a losing at- titude that traps them in a vicious, self-defeating cycle. Expecting to fail and fearing the worst, they become anxious, exert less effort, and “tune out” on important challenges.

Does high self-esteem ensure good life outcomes? This seemingly simple question is the subject of debate. On the one hand, based on their review of the research, Roy Baumeister and others (2003) conclude that although high self- esteem leads people to feel good, take on new challenges, and persist in the face of failure, the correlational evidence does not clearly support the strong conclu- sion that boosting self-esteem causes people to perform well in school or at work, to be socially popular, or to behave in ways that foster physical health. What’s more, Jennifer Crocker and Lora Park (2004) argue that the process of pursuing self-esteem itself can be costly. They point to research showing that in trying hard to boost and maintain their self-esteem, people often become anxious, avoid ac- tivities that risk failure, neglect the needs of others, and suffer from stress-related health problems. Self-esteem has its benefits, they concede, but striving for it can also be costly.

M ik

e Tw

oh y/

Th e

N ew

Y or

ke r C

ol le

ct io

n/ Ca

rt oo

n Ba

nk .c

80220_ch03_ptg01_hr_052-103.indd 78 27/01/16 11:21 AM

79 Self-Esteem

Challenging the conclusion that self-esteem is not worth striving for, William Swann and others (2007) note that although a person’s overall, or global, sense of self-worth may not be predictive of positive life outcomes, people with particu- lar domains of self-esteem benefit in more circumscribed ways. In other words, research suggests that individuals with high self-esteem—specifically for public speaking, mathematics, or social situations—outperform those who have less self-confidence in the domains of public speaking, math, and social situations, respectively. In addition to facilitating success in various life domains, research suggests that self-esteem is associated with a reduced risk for physical and men- tal health problems, substance abuse, and antisocial behavior (Orth et al., 2012; Swann et al., 2007).

j Are There gender and Race Differences? Just as individuals differ in their self-esteem, so too do social and cultural groups. If you were to administer a self-esteem test to thousands of people all over the world, would you find that some segments of the population score higher than others? Would you expect to see differences in the averages of men and women, blacks and whites, or inhabitants of different cultures? Believing that self-esteem promotes health, happiness, and success and concerned that some groups are disadvantaged in this regard, researchers have indeed made these types of comparisons.

Are there gender differences in self-esteem? Over the years, a lot has been written in the popular press about the inflated but fragile “male ego,” the low self- regard among adolescent girls and young women, and the resulting gender-related “confidence gap” (Orenstein, 1994). Does the research support this assumption? To find out, Kristin Kling and others (1999) statistically combined the results of 216 studies involving 97,000 respondents and then analyzed the surveys of 48,000 American students conducted by the National Center for Education Statistics. The result: Among adolescents and adults, males outscored females on various general measures of self-esteem. Contrary to popular belief, however, the difference was small, particularly among older adults. In fact, male–female differences are spe- cific to different aspects of self-esteem. In a meta-analysis of 115 studies involving 32,486 individuals, Brittany Gentile and her colleagues (2009) found, for example, that men have higher self-esteem with regard to their physical appearance and athletic abilities, while women have higher self-esteem when it comes to matters of ethics and personal morality.

Researchers have also wondered if low self-esteem is a problem for members of stigmatized minority groups who historically have been victims of prejudice and discrimination. Does membership in a minority group, such as African Ameri- cans, deflate one’s sense of self-worth? Based on the combined results of studies involving more than half a million respondents, Bernadette Gray-Little and Adam Hafdahl (2000) reported that black American children, adolescents, and adults consistently score higher—not lower—than their white counterparts on measures of self-esteem.

In a meta-analysis of hundreds of studies that compared all age groups and different American minorities, Jean Twenge and Jennifer Crocker (2002) con- firmed the African American advantage in self-esteem relative to whites but found that Hispanic, Asian, and Native American minorities have lower self- esteem scores. This self-esteem advantage is not easy to interpret. Surprised by the high African American scores, some researchers have suggested that perhaps African Americans—more than other minorities—are able to preserve their self- esteem in the face of adversity by attributing negative outcomes to the forces of

80220_ch03_ptg01_hr_052-103.indd 79 27/01/16 11:21 AM

80 Chapter 3 The Social Self

discrimination and using this adversity to build a sense of group pride. In this regard, Twenge and Crocker found that self-esteem scores of black Americans, relative to those of whites, have risen over time from the pre–civil rights days of the 1950s to the present.

j Self-Discrepancy Theory What determines how people feel about themselves? According to E. Tory Higgins (1989), our self-esteem is defined by the match or mismatch between how we see ourselves and how we want to see ourselves. To demonstrate, try the following exercise. On a blank sheet of paper, write down 10 traits that describe the kind of person you think you actually are (smart? easygoing? sexy? excitable?). Next, list 10 traits that describe the kind of person you think you ought to be, characteristics that would enable you to meet your sense of duty, obligation, and responsibility. Then make a list of traits that describe the kind of person you would like to be, an ideal that embodies your hopes, wishes, and dreams. If you follow these instructions, you should have three lists: your actual self, your ought self, and your ideal self.

Research has shown that these lists can be used to predict your self-esteem and your emotional well-being. The first list is your self-concept. The others rep- resent your personal standards, or self-guides. To the extent that you fall short of these standards, you will experience lowered self-esteem, negative emotion, and, in extreme cases, a serious affective disorder. The specific consequence depends on which self-guide you fail to achieve. If there’s a discrepancy between your actual and ought selves, you will feel guilty, ashamed, and resentful. You might even suffer from excessive fears and anxiety-related disorders. If the mismatch is between your actual and ideal selves, you’ll feel disappointed, frustrated, un- fulfilled, and sad. In the worst-case scenario you might even become depressed (Boldero & Francis, 2000; Higgins, 1999; Strauman, 1992). Each and every one of us must cope with some degree of self-discrepancy. In fact, recent research shows that people’s self-discrepancies—and the associated depression and anxiety—tend to remain stable over time, at least when retested over the course of three years (Watson et al., in press).

Our self-discrepancies may even set into motion a self-perpetuating process. Participating in a study of body images, college women with high rather than low discrepancies between their actual and ideal selves were more likely to compare themselves with thin models in TV commercials, which further increased their body dissatisfaction and depression (Bessenoff, 2006). Nobody’s perfect. Yet we do not all suffer from the emotional consequences. The reason, according to Higgins, is that self-esteem depends on a number of factors. One is simply the amount of discrepancy. The more of it there is, the worse we feel. Another factor is the im- portance of the discrepancy to the self. The more important the domain in which we fall short, again, the worse we feel. A third factor is how much we focus on our self-discrepancies. The more focused we are, the greater the harm. This last obser- vation raises an important question: What causes us to be more or less focused on our personal shortcomings? For an answer, we turn to self-awareness theory.

j The Self-Awareness “Trap” If you review your daily routine—your classes, work, errands, sports, leisure ac- tivities, social interactions, and meals—you will probably be surprised at how little time you actually spend thinking about yourself. In a study that illustrates this point, more than 100 people, ranging in age from 19 to 63, were equipped

80220_ch03_ptg01_hr_052-103.indd 80 27/01/16 11:21 AM

81 Self-Esteem

for a week with electronic beepers that sounded every two hours or so between 7:30 a.m. and 10:30 p.m. Each time the beepers went off, participants interrupted whatever they were doing, wrote down what they were thinking at that moment, and filled out a brief questionnaire. Out of 4,700 recorded thoughts, only 8% were about the self. For the most part, attention was focused on work and other ac- tivities. In fact, when participants were thinking about themselves, they reported feeling relatively unhappy and wished they were doing something else (Csikszent- mihalyi & Figurski, 1982).

State of Self-Awareness The finding that people may be unhappy while they think about themselves is interesting, but what does it mean? Does self-reflection bring out our personal shortcomings the way staring into a mirror draws our gaze to every blemish on the face? Is self-awareness an unpleasant mental state from which we need to retreat?

Many years ago, Robert Wicklund and his colleagues theorized that the an- swer is yes (Duval & Wicklund, 1972; Silvia & Duval, 2001; Wicklund, 1975). Ac- cording to their self-awareness theory, most people are not usually self-focused, but certain situations predictably force us to turn inward and become the objects of our own attention. When we talk about ourselves, glance into a mirror, stand before an audience or in front of a camera, watch ourselves on videotape, or be- have in a socially conspicuous manner, we enter into a state of heightened self- awareness that leads us naturally to compare our behavior to some high standard. Even a brightly lit room can make us feel self-aware and acutely “concerned about what other people think of me” (Steidle & Werth, 2014).

As you might expect, comparing oneself to some high standard often results in a negative discrepancy and a temporary reduction in self-esteem as we discover that we fall short. Thus, research participants who are seated in front of a mirror tend to react more negatively to their self-discrepancies, often slipping into a nega- tive mood state (Hass & Eisenstadt, 1990; Phillips & Silvia, 2005). Interestingly, Japanese people—whose culture already leads them to be highly concerned about their public “face”—are unaffected by the added presence of a mirror (Heine et al., 2008).

The real-life consequences of self-awareness can be substantial. The more self-focused people are in general, the more likely they are to find themselves in a bad mood (Flory et al., 2000) or depressed (Pyszczynski & Greenberg, 1987). People who are self-absorbed are also more likely to suffer from alcoholism, anxi- ety, and other clinical disorders (Mor & Winquist, 2002) and have self-destructive, suicide-related thoughts when they fail to meet their own standards (Chatard & Selimbegovic, 2011).

Is there a solution? Self-awareness theory suggests two basic ways of coping with such discomfort: (1) “shape up” by behaving in ways that help to reduce our self-discrepancies or (2) “ship out” by withdrawing from self-awareness. Accord- ing to Charles Carver and Michael Scheier (1981), the solution chosen depends on whether people think they can reduce their self-discrepancy and whether they’re pleased with the progress they make once they try (Duval et al., 1992). If so, they tend to match their behavior to personal or societal standards; if not, they will tune out, look for distractions, and turn attention away from the self. This process is depicted in d Figure 3.9.

In general, research supports the prediction that when people are self-focused, they are more likely to behave in ways that are consistent with their own per- sonal values or with socially accepted ideals. It’s why one team of researchers, after using bright lights to induce a state of self-awareness, has concluded that

“I have the true feeling of my- self only when I am unbearably unhappy.”

—Franz Kafka

self-awareness theory The theory that self-focused attention leads people to notice self-discrepancies, thereby motivating either an escape from self-awareness or a change in behavior.

80220_ch03_ptg01_hr_052-103.indd 81 27/01/16 11:21 AM

82 Chapter 3 The Social Self

“[g]ood lamps are the best police” (Zhong et al., 2010). Two interesting field stud- ies illustrate this point. In one, Halloween trick-or-treaters—children with masks, costumes, and painted faces—were greeted at a researcher’s door and left alone to help themselves from a bowl of candy. Although the children were asked to take only one piece, 34% violated the request. When a full-length mirror was placed behind the candy bowl, however, that number dropped to 12%. Apparently, the mirror forced the children to become self-focused, leading them to behave in a way that was consistent with public standards of desirable conduct (Beaman et al., 1979). In a second study, in England, customers at a lunch counter were trusted to pay for their coffee, tea, and milk by depositing money into an unsuper- vised “honesty box.” Hanging on the wall behind the counter was a poster that featured a picture of flowers or a pair of eyes. By calculating the ratio of money deposited to drinks consumed, researchers observed that people paid nearly three times more money in the presence of the eyes (Bateson et al., 2006).

Self-awareness theory states that if a successful reduction of self-discrepancy seems unlikely, individuals will take a second route: escape from self-awareness. Baumeister (1991) theorized that drug abuse, sexual masochism, spiritual ecstasy, binge eating, and suicide all serve this escapist function. Even television can serve as a form of escape. Sophia Moskalenko and Steven Heine (2003) brought college students into a laboratory and tested their actual–ideal self-discrepancies twice. Half watched a brief TV show on nature before the second test. In a second study, students were sent home with the questionnaire and instructed to fill it out either before or after watching TV. In both cases, those who watched TV had lower self- discrepancies on the second measure. In yet a third study, students who were told they had done poorly on an IQ test spent more time watching TV while waiting in the lab than those who were told they had succeeded. Perhaps TV and other forms of entertainment enable people to “watch their troubles away.”

One troubling health implication concerns the use of alcohol. According to Jay Hull, people often drown their sorrows in a bottle as a way to escape the negative implications of self-awareness. To test this hypothesis, Hull and Richard Young (1983) administered what was supposed to be an IQ test to male participants and

d F I g u R e 3 . 9

The Causes and effects of Self-Awareness

Self-awareness pressures people to reduce self-discrepancies either by matching their behavior to personal or societal standards or by withdrawing from self-awareness. © Cengage Learning®

Self- awareness

Self-focusing situations

Self-focusing persons

Accessibility of self- discrepancies

Match behavior to standards

High

Withdraw from self- awareness

Low

Expectation for discrepancy reduction

80220_ch03_ptg01_hr_052-103.indd 82 27/01/16 11:21 AM

83 Self-Esteem

gave false feedback suggesting that they had either succeeded or failed. Supposedly as part of a separate study, those participants were then asked to taste and rate different wines. As they did so, experimenters kept track of how much they drank during a 15-minute tasting period. As predicted, participants who were prone to self- awareness drank more wine after failure than after success, presumably to dodge the blow to their self-esteem. Among participants not prone to self-awareness, there was no difference in alcohol consumption. These results come as no surprise. Indeed, many of us expect alcohol to grant this form of relief (Leigh & Stacy, 1993) and help us manage our emotional highs and lows (Cooper et al., 1995).

Claude Steele and Robert Josephs (1990) proposed that alcoholic intoxication offers more than just a means of tuning out on the self. Alcohol can cause us to lose touch with who we are and to shed our inhibitions (Giancola et al., 2010). It can also evoke in us a state of “drunken self-inflation.” In one study, for example, participants rated their ac- tual and ideal selves on various traits—some important to self-esteem, others not important. After drinking either an 80-proof vodka cocktail or a harmless placebo, they re-rated themselves on the same traits. As measured by the perceived dis- crepancy between actual and ideal selves, those participants who were drinking expressed inflated views of themselves on traits they considered important (Banaji & Steele, 1989).

Trait of Self-Consciousness Just as situations evoke a state of self-awareness, some individuals are generally more self-focused than others. Research has revealed an important distinction between private self-consciousness—the tendency to in- trospect about our inner thoughts and feelings—and public self-consciousness— the tendency to focus on our outer public image (Buss, 1980; Fenigstein et al., 1975; Fenigstein, 2009). Table 3.1 presents a sample of items used to measure these traits.

Private and public self-consciousness are distinct traits. People who score high on a test of private self-consciousness tend to fill in incomplete sentences with first-person pronouns. They also make self-descriptive statements and recognize self-relevant words more quickly than other words (Mueller, 1982; Eichstaedt & Silvia, 2003). In contrast, those who score high on a measure of public self-con- sciousness are sensitive to the way they are viewed from an outsider’s perspective. Thus, when people were asked to draw a capital letter E on their foreheads, 43% of those with high levels of public self-consciousness, compared with only 6% of those with low levels, oriented the E so that it was backward from their own standpoint but correct for an observer (Hass, 1984). People who are high in pub- lic self-consciousness are also particularly sensitive to the extent to which others share their opinions (Fenigstein & Abrams, 1993).

The distinction between private and public self-awareness has implications for how to reduce self-discrepancies. According to Higgins (1989), people are motivated to meet either their own standards or the standards held for them by

private self-consciousness A personality characteristic of individuals who are introspective, often attending to their own inner states.

public self-consciousness A personality characteristic of individuals who focus on themselves as social objects, as seen by others.

From Fenigstein, A., Scheier, M. F., & Buss A. H. (1975). Public and private self-consciousness: Assessment and theory. Journal of Consulting and Clinical Psychology, 43, 522–527.

m T a b l e 3 . 1

How Self-Conscious are You?

These sample items appear in the Self-Consciousness Scale. How many of these statements indicating private or public self-consciousness would you use to describe yourself?

Items That Measure Private Self-Consciousness

d I’m always trying to figure myself out.

d I’m constantly examining my motives.

d I’m often the subject of my fantasies.

d I’m alert to changes in my mood.

d I’m aware of the way my mind works when I work on a problem.

Items That Measure Public Self-Consciousness

d I’m concerned about what other people think of me.

d I’m self-conscious about the way I look.

d I’m concerned about the way I present myself.

d I usually worry about making a good impression.

d One of the last things I do before leaving my house is look in the mirror.

80220_ch03_ptg01_hr_052-103.indd 83 27/01/16 11:21 AM

84 Chapter 3 The Social Self

significant others. If you’re privately self-conscious, you listen to an inner voice and try to reduce discrepancies relative to your own standards; if you’re publicly self-conscious, however, you try to match your behavior to socially accepted norms. As illustrated in d Figure 3.10, there may be “two sides of the self: one for you and one for me” (Scheier & Carver, 1983, p. 123).

god: like a Camera in the Sky? Across nations, cultures, and religions, more than 90% of the people on Earth believe in God or some other om- nipotent force. In most religions, people believe that God watches, evaluates, rewards, and punishes people for their moral and immoral behavior (Atran & Norenzayan, 2004).

For people of faith, thinking about God should trigger a state of self- focus—in the way that cameras, microphones, and other forms of surveil- lance do. As we saw, self-awareness theory predicts that self-focus should heighten a concern for our personal standards of good behavior. Does thinking about God have this effect? In a series of experiments, Will Gervais and Ara Norenzayan (2012) administered a religiosity scale to sort peo- ple into groups of high believers and low believers. All participants were then randomly assigned to complete a word task that required them to think about God, the presence of others, or something neutral. One study showed that high believers who were primed to think about God—but not low believers—became more self-aware, as if they were being observed by other people. A second study showed that thinking about God, as opposed to something neutral, led high believers to answer various questions in ways that were socially desirable—claiming, for example, that they are always good listeners and that they were never irritated by people who ask for favors (see d Figure 3.11). For individuals of faith, note Gervais and Norenzayan, God is “like a camera in the sky.”

As you might expect from this finding, other research has shown that thoughts of God can lead people of faith to behave more prosocially toward others. In one study, reported in an article entitled “God Is Watching You,” participants were given 10 $1 coins to keep or donate to an anonymous stranger. Before engaging in this “economic decision-making task,” participants were asked to unscramble sen- tences containing words that were neutral or religious (God, divine, spirit, prophet, sacred). How much of the $10 did the participants then leave for the stranger? Those who had worked with neutral words left an average of $1.84; those primed with God-related words left more than twice that amount, an average of $4.44 (Shariff & Norenzayan, 2007). Reasoning that thoughts of an all-knowing God should reduce people’s sense of anonymity, and increase their sense of account- ability, other researchers have found that reminders of God increase resistance to temptation—as measured, for example, by the number of bite-sized chocolate chip cookies participants ate in a taste-testing experiment (Laurin et al., 2012).

j Self-Regulation and Its limits To this point, we have seen that self-focused attention can motivate us to control our behavior and strive toward personal or social ideals. Achieving these goals— which enables us to reduce the self-discrepancies that haunt us—means that we must constantly engage in self-regulation—the processes by which we seek to control or alter our thoughts, feelings, behaviors, and urges in order to live an acceptable social life. From lifting ourselves out of bed in the early morning to diet- ing, running the extra mile, limiting how much we drink at a party, practicing safe

d F I g u R e 3 . 1 0

Revolving Images of Self

According to self-awareness theory, people try to meet either their own standards or standards held for them by others—depending, perhaps, on whether they are in a state of private or public self-consciousness. As Scheier and Carver (1983, p. 123) put it, there are “two sides of the self: one for you and one for me.” Snyder et al., 1983.

self-regulation The process by which people control their thoughts, feelings, or behavior in order to achieve a personal or social goal.

80220_ch03_ptg01_hr_052-103.indd 84 27/01/16 11:21 AM

85 Self-Esteem

sex, smiling politely at people we really don’t like, and working when we have more fun things to do, the exercise of self-control is something we are taught as children and do all the time (Bau- meister & Vohs, 2004; Bridgett et al., 2015; Carver & Scheier, 1998; Forgas et al., 2009; Heatherton, 2011).

Conflicts between our desires and the need for self-control are constant. When a sample of 205 adults wore Blackberry beepers for a week and reported their current states on cue, they indicated 7,827 episodes of desire—including, in order, biological desires to eat, sleep, and drink, followed by desires for media use, leisure activities, social contact, grooming and hygiene, sex, work, and sports activities. In nearly half of all in- stances, the desire was described as posing a conflict with other of the participants’ motivations, goals, and values (Hofmann et al., 2012).

Although the need is constant, Mark Muraven and Roy Baumeister (2000) theorized that self-control is a limited inner resource that can temporarily be depleted by usage. There are two components to their theory. The first is that all self-control efforts draw from a single common reservoir. The second is that exercising self-control is like flexing a muscle: Once used, it becomes fatigued and loses strength, making it more difficult to re-exert self-control—at least for a while, until the resource is replenished. Deny yourself the ice cream or donut that tickles your sweet tooth, and you’ll find it more difficult to hold your temper when provoked. Try to conceal your stage fright as you stand before an audience, and you’ll find it harder to resist the urge to watch TV when you should be studying.

Research has supported this provocative hypothesis. Muraven and Baumeister (1998) had participants watch a brief clip from an upsetting film that shows scenes of sick and dying animals exposed to radioactive waste. Some of the participants were in- structed to stifle their emotional responses to the clip, including their facial expressions; others were told to amplify or exaggerate their facial responses; a third group received no special instruc- tions. Both before and after the movie, self-control was measured by the length of time participants were able to squeeze a handgrip exerciser without letting go. As predicted, those who had to inhibit or amplify their emotions during the film—but not those in the third group—lost their strength it the handgrip task from the first time they tried it to the second (see d Figure 3.12). In laboratory experiments and in every- day life, other studies have since confirmed the point: After people exert self-control in one task, their capacity for self-regulation is weakened—causing them to talk too much, disclose too much, or brag too much in a later social situation (Hagger et al., 2010; Hofmann et al., 2012).

It appears that we can control ourselves only so much before self-regulation fatigue sets in, causing us to “lose it.” What might this mean, then, for people who constantly regulate their behavior? To find out, Kathleen Vohs and Todd Heatherton (2000) showed a brief and dull documentary to individual female college students, half of whom were chronic dieters. Placed in the viewing room— either within arm’s reach (high temptation) or 10 feet away (low temptation)— was a bowl filled with Skittles, M&Ms, Doritos, and salted peanuts that par- ticipants were free to sample. After watching the movie, they were taken to

d F I g u R e 3 . 1 1

Does Thinking About god Produce Self- Awareness effects?

Participants classified as high or low believers completed a word task that led them to think about God or something neutral. Afterward, they answered questions about themselves. Consistent with self-awareness theory, thinking about God—as opposed to something neutral—led High Believers (but not Lows) to answer the questions in ways that were more socially desirable. Based on Gervais, W. M., & Norenzayan, A. (2012). Like a cam- era in the sky? Thinking about God increases public self-aware- ness and socially desirable responding. Journal of Experimental Social Psychology, 48, 298–302.

High belief

So ci

al ly

D es

ir ab

le R

es po

ns e

Low belief

Control God

Belief in God

80220_ch03_ptg01_hr_052-103.indd 85 27/01/16 11:21 AM

86 Chapter 3 The Social Self

another room for an ice-cream taste test and told they could eat as much as they wanted. How much ice cream did they consume? The researchers predicted that dieters seated within reach of the bowl would have to fight the hardest to avoid snacking— an act of self-control that would cost them later. The predic- tion was confirmed. As measured by the amount of ice cream consumed in the taste test, dieters in the high-temptation con- dition ate more ice cream than did all nondieters and dieters in the low-temptation situation. What’s more, a second study showed that dieters who had to fight the urge in the high- temptation situation were later less persistent and quicker to give up on a set of impossible cognitive problems they were asked to solve.

Additional research has focused on the question of whether the resource limits of self-regulation are absolute and set in stone, or whether they can be overcome by various psychologi- cal factors—such as the exercise of willpower. Supporting the first hypothesis, some studies suggest that self-regulation fatigue invariably sets in because exerting self-control is physically tax- ing, as measured by the extent to which it consumes glucose, a vital source of bodily energy. Across a range of experiments, Matthew Gailliot and others (2007) had participants engage in an act of self-control—such as suppressing a word, thought, or emotion—before and after which blood samples were taken. Consistently, acts of self-control—relative to similar acts not requiring self-control—were followed by reduced blood glucose levels and a lessened capacity. What’s more, these researchers were able to counteract these adverse effects by feeding partici- pants sugared lemonade between tasks, which restored glucose to the bloodstream.

In contrast, other research indicates that psychological fac- tors can counteract self-regulation fatigue. In one program of research, for example, Veronika Job and others (2010, 2015) theorized that people who hold a nonlimited theory of will- power (as described by the statement, “After a strenuous mental activity, you feel energized for further challenging activities”) are more likely to maintain the ability to self- regulate after exertion than those who believe that self-control is limited and difficult to overcome (“After a strenuous mental activity, your energy is depleted and you must rest to get it refueled again”). In longitudinal studies that tracked college students over a semester, these researchers found that students who endorsed a nonlimited belief in willpower relative to those who endorsed a limited belief were better able to self-regulate during the difficult week before final exams: They ate fewer unhealthy foods, did less impulsive shopping, and spent more time studying. But does having a nonlimited theory of will- power create problems in everyday life when self-regulation demands are particularly high? Not necessarily. Among stu- dents who took a heavier-than-normal course load, and in other ways had a highly demanding semester, those who endorsed a nonlimited theory procrastinated less and ulti- mately scored higher grades (see d Figure 3.13).

d F I g u R e 3 . 1 2

Self-Control as a limited Inner Resource Participants were shown an upsetting film and told to amplify or suppress their emotional responses to it (a third group received no self-control instruction). Before and afterward, self-control was measured by persistence at squeezing a handgrip exerciser. As shown, the two groups that had to control their emotions during the film—but not those in the third group—later lost their willpower on the handgrip. Muraven, M., & Baumeister, R. F. (1998). Self-control as a limited resource: Regulatory depletion patterns. Journal of Personality and Social Psychology, 74, 774–789.

H an

dg ri

p pe

rs is

te nc

e (s

ec on

ds ) 80

50 Before movie After movie

Amplification Suppression No self-control

d F I g u R e 3 . 1 3

Does the Belief in Willpower Predict Self-Regulation? Tracking college students over time, researchers found that students who endorsed a non-limited belief in willpower, compared to those who endorsed a limited belief, were less likely to procrastinate when work demands were high. Job, V., Walton, G. M., Bernecker, K., & Dweck, C. S. (2015). Implicit theories about willpower predict self-regulation and grades in every- day life. Journal of Personality and Social Psychology, 108, 637–647

Pr oc

ra st

in at

io n

Low Demands High Demands 1

Nonlimited Theory Limited Theory

80220_ch03_ptg01_hr_052-103.indd 86 27/01/16 11:21 AM

87 Self-Esteem

j Ironic Mental Processes There’s another possible downside to self-control that is often seen in sports when athletes become so self-focused under pressure that they stiffen up and “choke.” Although many athletes rise to the occasion, the pages of sports history are filled with stories of basketball players who lose their shooting touch in the final min- ute of a championship game, of golfers who cannot sink a routine putt to win a tournament, and of tennis players who lose their serve, double-faulting when it matters most. “Choking” in these ways seems to be a paradoxical type of failure caused by trying too hard and thinking too much. When you learn a new motor activity like how to throw a curve ball or land a jump, you must think through the mechanics in a slow and cautious manner. As you get better, however, your move- ments become automatic, so you do not have to think about timing, breathing, the position of your head and limbs, or the distribution of your weight. You relax and just do it. Unless trained to perform while self-focused, athletes under pressure often try their hardest not to fail, become self-conscious, and think too much—all of which disrupts the fluid and natural flow of their performance (Baumeister, 1984; Beilock & Carr, 2001; Gray, 2004).

The paradoxical effects of attempted self-control are evident in other situ- ations, too. Studying what he calls ironic processes, Daniel Wegner (1994) has found that, at times, the harder you try to inhibit a thought, feeling, or behavior, the less likely you are to succeed. Try not to think about a white bear for the next 30 seconds, he finds, and that very image intrudes upon consciousness with remarkable frequency. Instruct the members of a jury to disregard inadmissible evidence, and the censored material is sure to pop to mind as they deliberate. Try not to worry about how long it’s taking to fall asleep, and you’ll stay awake. Try not to laugh in class, think about the chocolate cake in the fridge, or scratch the itch on your nose—well, you get the idea.

According to Wegner, every conscious effort at maintaining control is met by a concern about failing to do so. This concern automatically triggers an “ironic operating process” as the person, trying hard not to fail, searches his or her mind for the unwanted thought. The ironic process will not necessarily prevail, says Wegner. Sometimes we can put the imaginary white bear out of mind. But if the person is cognitively busy, tired, distracted, hurried, or under stress, then the ironic process, because it “just happens,” will prevail over the intentional process, which requires conscious attention and effort. As Wegner (1997) put it, “Any attempt at mental control contains the seeds of its own undoing” (p. 148).

“The highest possible stage in moral culture is when we rec- ognize that we ought to control our thoughts.”

—Charles Darwin

Why do some athletes choke under pressure and others rise to the occasion? Heading into the 2006 Winter Olympics in Turin, Italy, American downhill racer Bode Miller, a daredevil from New Hampshire, was favored to win several gold medals. Yet after slipping, falling, and missing gates, he failed to win a single medal, gold or otherwise (left). In contrast, American beach volleyball players Misty May- Treanor and Kerri Walsh Jennings were the twice-reigning gold medalists when they entered the 2012 Summer Olympics in London. Despite the pressure to defend their title, they turned in a series of clutch performances and won gold for the third straight Olympics (right).

A P

Im ag

es /T

ho m

as K

ie nz

M ar

ce o

D el

P oz

o/ Re

ut er

s/ La

nd ov

80220_ch03_ptg01_hr_052-103.indd 87 27/01/16 11:21 AM

88 Chapter 3 The Social Self

Ironic processes have been observed in a wide range of be- haviors. In an intriguing study of this effect on motor behavior, Wegner and others (1998) had participants hold a pendulum (a crystalline pendant suspended from a nylon fishing line) over the center of two intersecting axes on a glass grid, which formed a 1. Some participants were instructed simply to keep the pendulum steady; others were pointedly told not to allow it to swing back and forth along the horizontal axis. Try this yourself and you’ll see that it’s not easy to prevent all move- ment. In this experiment, however, the pendulum was more likely to swing horizontally when this direction was specifically forbidden. To further examine the role of mental distraction, the researchers had some participants count backward from a thousand by sevens while trying to control the pendulum. In this situation, the ironic effect was even greater. Among those who tried to prevent horizontal movement but could not con- centrate fully on the task, the pendulum swayed freely back and forth in the forbidden direction (see d Figure 3.14).

Applying this logic to keeping secrets, other researchers have found that instructing word-game players to conceal hid- den clues from a fellow player increased rather than decreased their tendency to leak that information (Lane et al., 2006). It may seem both comic and tragic, but at times our efforts at self- control backfire, thwarting even the best of intentions.

d F I g u R e 3 . 1 4

Ironic effects of Mental Control

In this study, participants tried to hold a pendulum motionless over a grid. As illustrated in the tracings shown here, they were better at the task when simply instructed to keep the pendulum steady (a) than when specifically told to prevent horizontal movement (c). Among participants who were mentally distracted during the task, this ironic effect was even greater (b and d). Wegner et al., 1998. © Cengage Learning

a. b.

c. d.

Sometimes the harder you try to control a thought, feeling, or behavior, the less likely you are to succeed.

TRue

j Mechanisms of Self-enhancement We have seen that self-awareness can lower self-esteem by focusing attention on self-discrepancies. We have seen that people often avoid focusing on themselves and turn away from unpleasant truths but that such avoidance is not always pos-

sible. And we have seen that efforts at self-regulation often fail and sometimes even backfire. How, then, does the average person cope with his or her faults, inadequacies, and uncertain future?

The Better-Than-Average effect At least in Western cultures, most people most of the time think highly of themselves. Consistently, and across a broad range of life domains, people see positive traits as more self-descriptive than negative traits, rate themselves more highly than they do others, rate themselves more highly than they are rated by others, exaggerate their control over life events, and predict that they have a bright future (Dunning et al., 2004; Sedikides & Gregg, 2008; Taylor, 1989).

The list of self-enhancement biases is long and impressive. Research shows that people overrate their effectiveness as speakers to an audience (Keysar & Henly, 2002); overestimate their own contributions to a group and how much they would be missed if absent (Savitsky et al., 2003); selectively recall positive feedback while neglecting the negative (Green et al., 2008); believe that they will achieve more in the future than they had in the past (Johnson, 2009); recognize themselves in self-portraits that were digitally morphed with 10% of another face that is highly attractive (Eply & Whitchurch, 2008); and credit themselves but not others for future but yet unrealized potential (Williams et al., 2012). Relative to others across a wide range of domains, it seems that people in general believe they

80220_ch03_ptg01_hr_052-103.indd 88 27/01/16 11:21 AM

89 Self-Esteem

are better, more honorable, more capable, and more compassionate. This pattern is known as the “better-than-average” effect.

What is particularly interesting is that people are more likely to see them- selves as better than average when it comes to personal traits that are important. Jonathon Brown (2012) presented participants with a list of characteristics and, for each one, he asked them to rate how important it is, how well it describes the self, and how well it describes most other people. He found that the effect is greater for traits that were rated as high in importance (such as honest, kind, responsible, and intelligent) than for traits of lesser importance (such as conscien- tious, agreeable, imaginative, and outgoing).

Implicit egotism It is clear and perhaps not surprising that people tilt positive when asked explicitly to evaluate themselves relative to others. But research suggests that people also exhibit implicit egotism, or implicit self-esteem—an unconscious and subtle expression of self-esteem. Implicit egotism is illustrated in studies showing that people are quicker to associate “self” words with positive traits than with negative traits. Particularly interesting in this regard is the finding that people evaluate the letters contained within their own names more favorably than other letters of the alphabet (Hoorens & Nuttin, 1993). This name-letter effect is also found in people’s preferences for their own birthday numbers (Jones et al., 2004).

In an article entitled “Why Susie Sells Seashells by the Seashore,” Brett Pelham and his colleagues (2002) theorized that the positive associations people form with the sight and sound of their own name may draw them toward other people, places, and entities that share this most personal aspect of “self.” In a thought-provoking series of studies, these researchers examined several important life choices that we make and found that people exhibit a small but statistically detectable tendency to gravitate toward things that contain the letters of their own name. For example, men and women are more likely than would be predicted by chance to live in places (Mike in Michigan, George in Georgia), attend schools (Kari from the University of Kansas, Preston from Penn State University), and choose careers (Dennis and Denise as dentists) whose names resemble their own. Marriage records found on genealogical websites also indicate that people are disproportionately likely to marry others with first or last names that resemble their own (Jones et al., 2004). Even more impressive, perhaps, is that people with high self-esteem are more likely to seek out romantic partners who are similar to themselves (Brown & Brown, 2015). In subtle but remarkable ways, it seems that we unconsciously seek out reflections of the self in other people and in our sur- roundings (Pelham et al., 2005).

Uri Simonsohn (2011) is not so sure. Although impressed with evidence for implicit egotism in the laboratory, Simonsohn re-analyzed the real-life data and concluded that too often the links resulted from a statistical fluke. Consider the finding that people are more likely than expected to marry others whose names resemble their own. A close look at these numbers reveals that the name effect may result from an “ethnic bias” so that people often marry others who share the same Hispanic, Asian, or other group’s last name. As for women named Georgia from that state, birth records show that more babies born in Georgia are named after the state of Georgia. In turn, Pelham and Carvallo (2011) point to ways in which Simonsohn’s criticism is overstated. To sum up: It’s clear that people prefer the letters of their own name and the number of their own birthday. What is less clear is whether these preferences influence the highest stakes decisions we make with regard to a marriage partner, or career, or a place to live.

implicit egotism A nonconscious form of self-enhancement.

80220_ch03_ptg01_hr_052-103.indd 89 27/01/16 11:21 AM

90 Chapter 3 The Social Self

Research on implicit egotism shows that, consciously or unconsciously, people hold themselves in high regard. We can’t all be perfect, however, nor can we all be better than average. So what supports this common illusion? In this section, we examine four methods that people use to rationalize or otherwise enhance their self-esteem: self-serving beliefs, self-handicapping, basking in reflected glory, and downward social comparisons.

Self-Serving Beliefs How well did you do on the SAT? When James Shepperd (1993b) asked college students about their performance on this infamous college entrance test, he uncovered two interesting patterns. First, the students overesti- mated their actual scores by an average of 17 points. This inflationary distortion was most pronounced among those with relatively low scores, and it persisted somewhat even when students knew that the experimenter would check their academic files. Second, a majority of students whose SAT scores were low described their scores as inaccurate and the test in general as invalid. In fact, the SATs for the group as a whole were predictive of their grade point averages. It’s not that people are delusional. But as memories fade, which occurs with the passage of time, the potential for self- enhancing recollections of test scores is increased (Willard & Gramzow, 2008).

When students receive exam grades, those who do well take credit for the success; those who do poorly complain about the instructor or test questions. When researchers have articles accepted for publication, they credit the quality of their work; when articles are rejected, they blame the editors and reviewers. When gamblers win a bet, they marvel at how skilled they are; when they lose, they blame fluke events that transformed near-victory into defeat. Whether people have high or low self-esteem, explain their own outcomes publicly or in private, and try to be honest or to make a good impression, there is bias. Across a range of cultures, people tend to take credit for success and distance themselves from failure (Mezulis et al., 2004; Schlenker et al., 1990)—all while seeing themselves as objective, not biased (Pronin et al., 2004).

Most of us are also unrealistically optimistic about the future. College students who were asked to predict their own futures compared with that of the average peer believed that they would graduate higher in their class, get a better job, have a happier marriage, and bear a gifted child. They also believed that they were less likely to get fired or divorced, have a car accident, become depressed, be victimized by crime, or suffer a heart attack (Weinstein, 1980). A number of later studies have shown that in matters of sports, politics, health and illness, romantic relationships, economics, and social issues, people exhibit an optimistic bias— essentially, an instance of wishful thinking—about their own future, judging desir- able events as more likely to occur than undesirable events (Krizan & Windschitl, 2007; Shepperd et al., 2013).

Perhaps one reason that people are eternally optimistic is that they harbor il- lusions of control, overestimating the extent to which they can influence personal outcomes that are not, in fact, within their power to control (Stefan & David, 2013). In a series of classic experiments on the illusion of control, Ellen Langer (1975) found that college students bet more money in a chance game of high- card when their opponent seemed nervous rather than confident and were more reluctant to sell a lottery ticket if they had chosen the number themselves than if it was assigned to them. Emily Pronin and others (2006) then tested the related hypothesis that imagining an event before it occurs can lead people to think they had influenced it. In one study, participants watched a trained confederate shoot hoops on a basketball court. Before each shot, they were instructed to visualize his success (“the shooter releases the ball and it swooshes through the net”) or an

“We don’t see things as they are, we see them as we are.”

—Anaïs Nín

80220_ch03_ptg01_hr_052-103.indd 90 27/01/16 11:21 AM

91 Self-Esteem

irrelevant event (“the shooter’s arm curls to lift a dumbbell”). After the confeder- ate’s successful shooting spree, spectators rated the extent of their influence over his performance. As if linking thoughts to outcomes, they exhibited an illusion of mental causation, taking more credit when they had visualized the shooter’s suc- cess than when they had not.

Self-Handicapping “My dog ate my homework.” “I had a flat tire.” “My alarm didn’t go off.” “My phone battery died.” “I had a bad headache.” “The referee blew the call.” On occasion, people make excuses for past performance. Sometimes we come up with excuses in anticipation of future performance. When people are afraid that they might fail in an important situation, they use illness, shyness, anxiety, pain, trauma, and other complaints as excuses for the possibility of failure (Snyder & Higgins, 1988). The reason people self-handicap is simple: By admitting to a limited physical or mental weakness, they can shield themselves from what could be the most shattering implication of failure—a lack of ability.

One form of excuse-making that many of us can relate to is procrastination —a purposive delay in starting or completing a task that is due at a particular time (Ferrari et al., 1995). Some people procrastinate chronically, whereas others do so only in certain situations. The phenomenon itself is found in a number of cultures—including Spain, Peru, Venezuela, England, Australia, and the United States, where nearly 15% of men and women identify themselves as chronic pro- crastinators (Ferrari et al., 2007). There are many reasons why someone might put off what needs to get done—whether it’s studying for a test, shopping for Christmas, or preparing for the April 15 tax deadline. According to Joseph Ferrari (1998), one “benefit” of procrastinating is that it helps to provide an excuse for possible failure.

Making excuses is one way to cope with the threatening implications of fail- ure. Sometimes, however, this strategy is taken one step further, as when people actually sabotage their own performance. It seems like the ultimate paradox, but there are times when we purposely set ourselves up for failure in order to pre- serve our precious self-esteem. As first described by Stephen Berglas and Edward Jones (1978), self-handicapping refers to actions people take to handicap their own performance in order to build an excuse for anticipated failure. To demon- strate, Berglas and Jones recruited college students for an experiment supposedly concerning the effects of drugs on intellectual performance. All the participants worked on a 20-item test of analogies and were told that they had done well, after which they expected to work on a second, similar test. For one group, the problems in the first test were relatively easy, leading participants to expect more success in the second test; for a second group, the problems were insoluble, leav- ing participants confused about their initial success and worried about possible failure. Before seeing or taking the second test, participants were given a choice of two drugs: Actavil, which was supposed to improve performance, and Pandocrin, which was supposed to impair it.

Although no drugs were actually administered, most participants who were confident about the upcoming test selected the Actavil. In contrast, most males (but not females) who feared the outcome of the second test chose the Pandocrin. By handicapping themselves, these men set up a convenient excuse for failure—an excuse, we should add, that may have been intended more for the experimenter’s benefit than for the benefit of the participants themselves. A follow-up study showed that although self-handicapping occurs when the experimenter witnesses the participants’ drug choice, it is reduced when the experimenter is not present while that choice is being made (Kolditz & Arkin, 1982).

self-handicapping Behaviors designed to sabotage one’s own performance in order to provide a subsequent excuse for failure.

80220_ch03_ptg01_hr_052-103.indd 91 27/01/16 11:21 AM

92 Chapter 3 The Social Self

Some people use self-handicapping as a defense more than others do—and in different ways (Rhodewalt, 1990). For example, some men self-handicap by taking drugs (Higgins & Harris, 1988) or neglecting to practice (Hirt et al., 1991), whereas women tend to report stress and physical symptoms (Smith et al., 1983). Another tactic is to set one’s goals too high, as perfectionists like to do, which sets up failure that is not interpreted to reflect a lack of ability (Hewitt et al., 2003). Yet another paradoxical tactic used to reduce performance pressure is for people to play down their own ability, lower expectations, and publicly predict that they will fail—a self-presentation strategy called sandbagging (Gibson & Sachau, 2000). People also differ in their reasons for self-handicapping. Dianne Tice (1991) found that people who are low in self-esteem use self-handicapping to set up a defensive, face-saving excuse in case they fail, whereas those who are high in self-esteem use it as an opportunity to claim extra credit if they succeed.

Whatever the tactics and whatever the goal, self-handicapping seems like an ingenious strategy: With the odds seemingly stacked against us, withholding effort, procrastinating, or otherwise pulling back can help to insulate the self from

failure and enhance the self in the case of success. But this strategy is not without a real cost. Sabotaging ourselves—by not practicing or by drinking too much, faking illness, or setting goals too high—objectively increases the risk of failure, even in important life domains. Indeed, re- search shows that self-handicapping is maladaptive when used in school and ultimately limits students’ academic achievement and potential (Urdan & Midgley, 2001; Schwinger et al., 2014).

Basking in Reflected glory To some extent, your self-esteem is influenced by individuals and groups with whom you identify. Accord- ing to Robert Cialdini and his colleagues (1976), people often bask in reflected glory (BIRg) by showing off their connections to successful others. Cialdini’s team first observed BIRGing on the university cam- puses of Arizona State, Louisiana State, Notre Dame, Michigan, Pitts- burgh, Ohio State, and Southern California. On the Monday mornings after football games, they counted the number of school sweatshirts worn on campus and found that more of them were worn if the team had won its game on the previous Saturday. In fact, the larger the margin of victory was, the more school shirts they counted.

To evaluate the effects of self-esteem on BIRGing, Cialdini gave students a general-knowledge test and rigged the results so half would succeed and half would fail. The students were then asked to describe in their own words the outcome of a recent football game. In these descriptions, students who thought they had just failed a test were more likely than those who thought they had succeeded to share in

their team’s victory by exclaiming that “we won” and to distance themselves from defeat by lamenting how “they lost.” In another study, participants coming off a recent failure were quick to point out that they had the same birth date as some- one known to be successful—thus BIRGing by a merely coincidental association (Cialdini & De Nicholas, 1989).

If self-esteem is influenced by our links to others, how do we cope with friends, family members, teammates, and co-workers of low status? Again, consider sports fans, an interesting breed. They loudly cheer their team in vic- tory, but they often turn and jeer their team in defeat. This behavior seems fickle, but it is consistent with the notion that people derive part of their self- esteem from associations with others. In one study, participants took part in

People often sabotage their own performance in order to protect their self- esteem.

TRue

bask in reflected glory (BIRg) To increase self-esteem by associating with others who are successful.

When Vietnam veterans returned home without victory more than 40 years ago, they were neglected, even scorned, by the American public. It seems that the tendency to bask in reflected glory is matched by an equally powerful need to cut off reflected failure.

To m

W ill

ia m

s/ CQ

R ol

l C al

l/ G

et ty

Im ag

80220_ch03_ptg01_hr_052-103.indd 92 27/01/16 11:21 AM

93 Self-Esteem

a problem-solving team that then succeeded, failed, or received no feedback about its performance. Participants were later offered a chance to take home a team badge. In the success and no-feedback groups, 68% and 50%, respec- tively, took badges; in the failure group, only 9% did (Snyder et al., 1986). For diehard sports fans—male and female alike—the tendency to bask in reflected glory is matched by an equally powerful tendency to CORF—that is, to cut off reflected failure (Ware & Kowalski, 2012).

Downward Social Comparisons Earlier, we discussed Festinger’s (1954) the- ory that people evaluate themselves by comparison with similar others. But con- template the implications. If the people around us achieve more than we do, what does that do to our self-esteem? Perhaps adults who shy away from class reunions in order to avoid having to compare themselves with former classmates are acting out an answer to that question. More to the point, perhaps the “Facebook Depression” link described earlier in this chapter provides a statistical answer to the question (Feinstein et al., 2013; Verduyn et al., 2015; Vogel et al., 2014).

Long ago, Festinger fully realized that people don’t always seek out objec- tive information and that social comparisons are sometimes made in self-defense. When a person’s self-esteem is at stake, he or she often benefits from making downward social comparisons with others who are less successful, less happy, or less fortunate (Wills, 1981; Wood, 1989). Research shows that people who suffer some form of setback or failure adjust their social comparisons in a downward direction (Gibbons et al., 2002) and that these comparisons have an uplifting effect on their mood and on their outlook for the future (Aspinwall & Taylor, 1993; Gibbons & McCoy, 1991).

Although Festinger never addressed the issue, Anne Wilson and Michael Ross (2000) note that in addition to making social com- parisons between ourselves and similar oth- ers, we make temporal comparisons between our past and present selves. In one study, these investigators had college students de- scribe themselves; in another, they analyzed the autobiographical accounts of celebrities appearing in popular magazines. In both cases, they counted the number of times the self-descriptions contained references to past selves, to future selves, and to others. The result was that people made more compari- sons to their own past selves than to others, and most of these temporal comparisons were favorable. Other research has confirmed the basic point. Keenly aware of how “I’m better today than when I was in the past,” people use downward temporal comparisons the way they use downward social comparisons as a means of self-enhancement (Zell & Alicke, 2009).

Whether people make upward or downward social comparisons can have striking health implications. When victimized by tragic life events (perhaps a crime, an accident, a disease, or the death of a loved one), people like to affiliate with others in the same predicament who have adjusted well, role models who

downward social comparison The defensive tendency to compare ourselves with others who are worse off than we are.

av id

S ip

re ss

/C on

de N

as t C

ol le

ct io

80220_ch03_ptg01_hr_052-103.indd 93 27/01/16 11:21 AM

94 Chapter 3 The Social Self

offer hope and guidance. But they tend to compare themselves with others who are worse off, a form of downward social comparison (Taylor & Lobel, 1989).

Clearly it helps to know that life could be worse, which is why most cancer patients tend to compare themselves with others in the same predicament but who are adjusting less well than they are. In a study of 312 women who had early-stage breast cancer and were in peer support groups, Laura Bogart and Vicki Helgeson (2000) had the patients report every week for seven weeks on instances in which they talked to, heard about, or thought about another patient. They found that 53% of all the social comparisons made were downward, to others who were worse off; only 12% were upward, to others who were better off (the rest were “lateral” comparisons to similar or dissimilar others). In fact, the more often patients made these social comparisons, the better they felt. Downward social comparison is also associated with an ability to cope with the kinds of life regrets that sometimes haunt people as they get older. Adult development researchers have observed that aging adults often experience intense feelings of regret over decisions made, contacts lost, opportunities passed up, and the like— and these regrets can compromise the quality of their lives. Isabelle Bauer and others (2008) asked adults ranging from 18 to 83 years old to disclose their biggest regret and then indicate whether their same-age peers had regrets that were more or less severe. Among the older adults in the sample, those who tended to see oth- ers as having more severe regrets than their own felt better than those who saw others as less regretful.

j Are Positive Illusions Adaptive? Psychologists used to maintain that an accurate perception of reality is vital to mental health. In recent years, however, this view has been challenged by research on the mechanisms of self-enhancement. Consistently, as we have seen, people preserve their self-esteem through biased beliefs, self-handicapping, BIRGing, and downward comparisons. The result: Most people see themselves as better than average. Are these strategies a sign of health and well-being or are they symptoms of disorder?

When Shelley Taylor and Jonathon Brown (1988) first reviewed the research, they found that individuals who are depressed or low in self-esteem actually have more realistic views of themselves than do most others who are better adjusted. Their self-appraisals are more likely to match appraisals of them made by neutral observers, they make fewer self-serving attributions to account for success and failure, they are less likely to exaggerate their control over uncontrollable events, and they make more balanced predictions about their future. Based on these re- sults, Taylor and Brown reached the provocative conclusion that positive illusions promote happiness, the desire to care for others, and the ability to engage in pro- ductive work—hallmark attributes of mental health: “These illusions help make each individual’s world a warmer and more active and beneficent place in which to live” (p. 205). People with high self-esteem appear to be better adjusted in personality tests and in interviews rated by friends, strangers, and mental health professionals (Taylor et al., 2003).

Drawing on evolutionary theory, William von Hippel and Robert Trivers (2011) offered a new and provocative perspective on the adaptive advantages of self-de- ception. Over the years, evolutionary psychologists have noted that deceit is a com- munication skill that animals use to curry favor, attract a mate, and influence others to share food, shelter, and other resources. Just as humans have evolved ways to deceive others, we have also evolved ways to detect deception in others. The more skilled we are in these interactions, the more we flourish. The problem is that when

80220_ch03_ptg01_hr_052-103.indd 94 27/01/16 11:21 AM

95 Self-Esteem

people lie, they become nervous, try to suppress the nervousness, and work hard to make the lie sound believable—especially when talking to others who know them. Here’s the hitch: If deceivers can convince themselves that their deception is true, they will not be as nervous, not have to suppress, not have to work so hard, and, therefore, be more successful. As Seinfeld character George Costanza advised Jerry, who was scheduled to take a lie-detector test, “It’s not a lie if you believe it.”

When it comes to the illusion of control, unrealistic optimism, and the other self-enhancement biases that enable people to see themselves as better than aver- age, von Hippel and Trivers (2011) extend their argument: By deceiving ourselves in ways that create positive illusions, we are able to display greater confidence in public than we may actually feel, making us more successful in our social relations. Perhaps this explains the research finding that people tend to overestimate the self- esteem of other people—independent of how well we know them (Kilianski, 2008).

Not everyone agrees with the notion that it is adaptive in the long run to wear rose-colored glasses. Roy Baumeister and Steven Scher (1988) warned that posi- tive illusions can give rise to chronic patterns of self-defeating behavior, as when people escape from self-awareness through alcohol and other drugs, self-handicap themselves into failure and underachievement, deny health-related problems un- til it’s too late for treatment, and rely on the illusion of control to protect them from the inescapable odds of the gambling casino. Others have noted that people sometimes need to be self-critical in order to improve. In a study on success and failure feedback, Heine and others (2001) found that whereas North American college students persisted less on a task after an initial failure than after success, Japanese students persisted more in this situation. Sometimes we have to face up to our shortcomings in order to correct them.

From an interpersonal standpoint, C. Randall Colvin and others (1995) found that people with inflated rather than realistic views of themselves were rated less favorably on certain dimensions by their own friends. In their studies, self- enhancing men were seen as boastful, condescending, hostile, and less consider- ate; self-enhancing women were seen as more hostile, more defensive and sensi- tive to criticism, more likely to overreact to minor setbacks, and less well liked. People with inflated self-images may make a good first impression on others, but they are liked less and less as time wears on (Paulhus, 1998).

Realism or illusion—which orientation is more adaptive? As social psycholo- gists debate the short-term and long-term effects of positive illusions and their evolutionary advantages, it seems that there is no simple answer. For now, the picture that has emerged is this: People who harbor positive illusions of them- selves are likely to enjoy the benefits and achievements of high self-esteem and social influence. But these same individuals may pay a price in other ways, as in their relations with others. Do positive illusions motivate personal achievement but alienate us socially from others? Is it adaptive to see oneself in slightly inflated terms but maladaptive to take a view that is too biased? It will be interesting to see how this thorny debate is resolved in the years to come.

j Culture and Self-esteem Earlier we saw that inhabitants of individualistic cultures tend to view themselves as distinct and autonomous, whereas those in collectivist cultures view the self as part of an interdependent social network. Do these different orientations have implications for self-esteem? This turns out to be a tricky question.

Steven Heine and his colleagues (1999) argued that cultures have differing effects on the pursuit of self-esteem. Comparing the distribution of self-esteem

80220_ch03_ptg01_hr_052-103.indd 95 27/01/16 11:21 AM

96 Chapter 3 The Social Self

test scores in Canada and Japan, they found that while most Canadians’ scores clustered in the high-end range, a majority of Japanese respondents scored in the center of that same range. In other studies, they also observed that Japa- nese respondents can sometimes be quite self-critical, talking about themselves in negative, self-effacing terms.

Do Japanese people have a less positive self-esteem compared to Americans? Or do Japanese respondents have a positive self-esteem but feel compelled to present themselves modestly to others (as a function of the collectivist need to “fit in” rather than “stand out”)? To answer this question, some researchers have used less direct, subtle tests that measure implicit self-esteem—a person’s unconscious tendency to positively evaluate people and objects that reflect upon themselves (Falk & Heine, 2015). In a timed word-association study, researchers found that despite their lower scores on overt self-esteem tests, Asian Americans—just like their European American counterparts—are quicker to associate themselves with positive words like happy and sunshine than with negative words such as vomit and poison (Greenwald & Farnham, 2000; Kitayama & Uchida, 2003). In keep- ing with the Eastern dialectical perspective described earlier, other implicit self- esteem research has shown that although East Asians, like most Westerners, are quick to associate the self with positive traits, they are more likely to associate the self with contradictory negative traits as well (Boucher et al., 2009).

Drawing on these results, Constantine Sedikides and colleagues (2003) maintained that people from individualist and collectivist cultures are similarly motivated to think highly of themselves—that the burning need for positive self-regard is universal, or “pancultural.” The observed differences, they argue, stem from the fact that cultures influence how we seek to fulfill that need: Individualists present themselves as unique and self-confident, while collectivists present themselves as modest, equal members of a group. From this perspective, people are tactical in their self-enhancements, ex- hibiting self-praise or humility depending on what is desirable within their cultural surroundings (J. D. Brown, 2003; Lalwani et al., 2006; Sedikides et al., 2005). As for what is desirable—when Japanese and American students evaluated a fictitious per- son from his or her high or low responses on a self-esteem questionnaire, Americans showed a stronger preference for the high self-esteem person (Brown, 2010).

Heine and his colleagues agree only in part with this interpretation of the re- search. They too argue that all people have a need for positive self-regard, wanting to become “good selves” within their own culture. They note, however, that in the effort to achieve this goal, individualists tend to use self-enhancement tactics to stand out, confirm, and express themselves, whereas East Asians and other collectivists tend to maintain face in order to fit in, improve the self, and adjust to the standards set by their groups. To sum up: Although the underlying need for positive self-regard is universal, the specific drive toward self-enhancement is culturally ingrained (Heine, 2005; Heine & Hamamura, 2007).

Self-Presentation The human quest for self-knowledge and esteem tells us about the inner self. The portrait is not complete, however, until we paint in the outermost layer, the behavioral expression of the social self. Most people are acutely concerned about the image they present to others. The fashion industry, tattoos, piercings, and cosmetic surgeries designed to reshape everything from eyelids and noses to butts and breasts, and the endless search for miracle drugs that grow hair, remove hair, whiten teeth, freshen breath, and smooth out wrinkles all exploit our preoccupation

80220_ch03_ptg01_hr_052-103.indd 96 27/01/16 11:21 AM

97 Self-Presentation

with physical appearance. In a similar manner, we are concerned about the impressions we convey through our public behavior both in person and on Facebook and other social networking sites.

Thomas Gilovich and others (2000) found that people are so self-conscious in public settings that they are often subject to the spotlight effect, a tendency to believe that the social spotlight shines more brightly on them than it really does. In one set of studies, participants were asked to wear a T-shirt with a flatter- ing or embarrassing image into a room full of strangers, after which they estimated how many of those strangers would be able to identify the image. Demonstrating that people self-consciously feel as if all eyes are upon them, the T-shirted participants overestimated by 23% to 40% the number of observers who had noticed and could recall what they were wearing. Follow-up studies have similarly shown that when people commit a pub- lic social blunder, they later overestimate the negative impact of their behavior on those who had observed them (Savitsky et al., 2001).

In As You Like It, William Shakespeare wrote, “All the world’s a stage, and all the men and women merely players.” This insight was first put into social science terms by sociologist Erving Goffman (1959), who argued that life is like a theater and that each of us acts out certain lines, as if from a script. Most important, said Goffman, is that each of us assumes a certain face, or social identity, that others politely help us maintain. This “dramaturgical” perspective on social behavior—in which society is seen as an elaborately scripted play in which individuals enact different roles, as if performing in a reality TV show—has a long tradition in social psychology (Sandstrom et al., 2009; Sullivan et al., 2014).

Inspired by Goffman’s theory, social psychologists study self-presentation— the process by which we try to shape what other people think of us and what we think of ourselves (Schlenker, 2003). An act of self-presentation may take many different forms. It may be conscious or unconscious, accurate or misleading, or intended for an external audience or for ourselves. In The Manipulation of Online Self-Presentation: Create, Edit, Re-edit and Present, Alison Attrill (2015) describes how this process plays out on social networking sites, where people determine how to present their selves to a vast online public. In this section, we look at the goals of self-presentation and the ways that people try to achieve these goals.

j Strategic Self-Presentation There are basically two types of self-presentation, each serving a different motive. Strategic self-presentation consists of our efforts to shape others’ impressions in specific ways in order to gain influence, power, sympathy, or approval. Prominent examples of strategic self-presentation are everywhere: in personal ads, in online message boards, in political campaign promises, in defendants’ appeals to the jury. The specific goals vary and include the desire to be seen as likable, competent, moral, dangerous, or helpless. Whatever the goal may be, people find it less effortful

self-presentation Strategies people use to shape what others think of them.

When Ellen DeGeneres hosted the Academy Awards in 2014, she invited stars from the audience to join in this famous celebrity selfie. As you try to recognize as many stars as you can, note that people in general, whether posing for pictures or not, are acutely aware of the image they present to others.

El le

n D

eG en

er es

/T w

itt er

/G et

ty Im

ag es

E nt

er ta

in m

en /G

et ty

Im ag

80220_ch03_ptg01_hr_052-103.indd 97 27/01/16 11:21 AM

98 Chapter 3 The Social Self

to present themselves in ways that are accurate rather than contrived (Vohs et al., 2005).

To illustrate this point, Beth Pontari and Barry Schlenker (2000) instructed research participants who tested as introverted or extroverted to present themselves to a job interviewer in a way that was consistent or inconsistent with their true personality. Without distraction, all participants successfully pre- sented themselves as introverted or extroverted, de- pending on the task they were given. But could they present themselves as needed if, during the interview, they also had to keep an eight-digit number in mind for a memorization test? In this situation, cognitively busy participants self-presented successfully when asked to convey their true personalities but not when asked to portray themselves in a way that was out of character.

The specific identities that people try to pres- ent may vary from one person and situation to an- other. However, two strategic self-presentation goals are very common. The first is ingratiation, a term used to describe acts that are motivated by the desire to “get along” with others and be liked. The other is self-promotion, a term used to describe acts that

are motivated by a desire to “get ahead” and gain respect for one’s competence (Arkin, 1981; Jones & Pittman, 1982). As shown in Table 3.2, observations of employment interviews reveal that ingratiation and self-promotion are the most common self-presentation tactics that Western job applicants use (Stevens & Kristof, 1995) and that these tactics lead recruiters to form positive impressions (Higgins & Judge, 2004). Of course, as you might expect, research shows that job applicants will present themselves in very different ways depending on the ideals of their culture (Sandal et al., 2014).

On the surface, it seems easy to achieve these goals. When people want to be liked, they put their best foot forward, smile a lot, nod their heads, express agree- ment, and, if necessary, use favors, compliments, and flattery. When people want to be admired for their competence, they try to impress others by talking about themselves and immodestly showing off their status, knowledge, and exploits. In both cases, there are trade-offs. As the term brown-nosing graphically suggests, ingratiation tactics need to be subtle or else they will backfire (Jones, 1964). People also do not like those who relentlessly trumpet and brag about their own achievements (Godfrey et al., 1986) or who exhibit a “slimy” pattern of being friendly to superiors but not to subordinates (Vonk, 1998).

Self-presentation may give rise to other problems as well. Suggesting that “Self-Presentation Can Be Hazardous to Your Health,” Mark Leary and others (1994) reviewed evidence suggesting that the need to project a favorable public image can lure us into unsafe patterns of behavior. For example, self-presentation concerns can increase the risk of AIDS (as when men are too embarrassed to buy condoms and talk openly with their sex partners), skin cancer (as when people bake under the sun for an attractive tan), eating disorders (as when women over- diet to stay thin), drug abuse (as when teenagers smoke, drink, and use drugs to impress their peers), and accidental injury (as when young men drive recklessly to look fearless to others).

Higgins & Judge, 2004; Stevens & Kristof, 1995. © Cengage Learning

m T a b l e 3 . 2

Strategic Self-Presentation in the employment Interview

In studies of the influence tactics that job applicants report using in employment interviews, the following uses of ingratiation and self- promotion were commonly reported.

Ingratiation

d I complimented the interviewer or organization.

d I discussed interests I shared in common with the recruiter.

d I indicated my interest in the position and the company.

d I indicated my enthusiasm for working for this organization. I smiled a lot or used other friendly nonverbal behaviors.

Self-Promotion

d I played up the value of positive events that I took credit for.

d I described my skills and abilities in an attractive way.

d I took charge during the interview to get my main points across.

d I took credit for positive events even if I was not solely responsible.

d I made positive events I was responsible for appear better than they actually were.

80220_ch03_ptg01_hr_052-103.indd 98 27/01/16 11:21 AM

99 Self-Presentation

j Self-Verification In contrast to strategic self-presentation is a second motive, self-verification: the de- sire to have others perceive us as we truly perceive ourselves. According to William Swann (1987), people are highly motivated in their social encounters to confirm or verify their existing self-concept in the eyes of others. Swann and his colleagues have gathered a great deal of evidence for this hypothesis and have found, for example, that people selectively elicit, recall, and accept personality feedback that confirms their self-conceptions. In fact, people sometimes bend over backward to correct others whose impressions are positive but mistaken. In one study, partici- pants interacted with a confederate who later said that they seemed dominant or submissive. When the comment was consistent with the participant’s self-concept, it was accepted at face value. Yet when it was inconsistent, participants went out of their way to prove the confederate wrong: Those who perceived themselves as dominant but were labeled submissive later behaved more assertively than usual; those who viewed themselves as submissive but were labeled dominant subse- quently became even more docile (Swann & Hill, 1982).

Self-verification seems desirable, but do people who harbor a negative self- concept want others to share that impression? Nobody is perfect, and everyone has some faults. But do we really want to verify these faults in the eyes of others? Do those of us who feel painfully shy, socially awkward, or insecure about an abil- ity want others to see these weaknesses? Or would we prefer to present ourselves in public as bold, graceful, or competent? What happens when the desire for self- verification clashes with the need for flattery and self-enhancement?

Seeking to answer this question, Swann and his colleagues (1992) asked each student participant in a laboratory study to fill out a self-concept questionnaire and then choose an interaction partner from two other participants—one who supposedly had evaluated them favorably and a second who had supposedly evaluated them unfavorably. The result? Although participants with a positive self-concept chose part- ners who viewed them in a positive light, a majority of those with a negative self-concept preferred part- ners who confirmed their admitted shortcomings. In a later study, 64% of participants with low self-esteem (compared with only 25% with high self-esteem) sought clinical feedback about their weaknesses rather than strengths when given a choice (Giesler et al., 1996). Research also suggests that people prefer to interact with others who verify their group mem- berships, a key aspect of the collective self (Chen et al., 2009). Indeed, the desire for self-verification appears to be universal, observed in both individual- ist and collectivist cultures (Seih et al., 2013).

If people seek self-verification from laboratory partners, it stands to reason that they would want the same from their close relationships. In a study of married couples, husbands and wives separately an- swered questions about their self-concepts, their spouse, and their commitment to the marriage. As predicted, people who had a positive self-concept expressed more commitment to partners who appraised them favorably, whereas those with a negative self-concept felt more committed to partners who appraised them un- favorably (Swann et al., 1992).

People often distinguish between their public and private self. However, research on self- verification suggests that this cartoon is wrong—that people do want to be defined by who they are.

Pe te

r C . V

ey /T

he N

ew Y

or ke

r C ol

le ct

io n/

Ca rt

oo n

Ba nk

.c om

80220_ch03_ptg01_hr_052-103.indd 99 27/01/16 11:21 AM

100 Chapter 3 The Social Self

On important aspects of the self-concept, research shows that people would rather reflect on and learn more about their positive qualities than negative ones. Still, it appears that the de- sire for self-verification is powerful and can even, at times, trump the need for self-enhancement. We all want to make a good impression, but we also want others in our lives to have an accurate impression (Swann & Bosson, 2010). Studying on- line social networking sites, Mitja Back and others (2010) found strong support for this proposition. These researchers studied 236 Facebook users in the United States and Germany and found that the impressions conveyed by their posted profiles correlated highly with a combination of objective personality tests, self-reports, and reports from a sample of well-acquainted friends.

concerns, they are poised, ready, and able to modify their behavior as they move from one setting to another. As measured by the Self-

Monitoring Scale (Snyder, 1974; Snyder & Gangestad, 1986), they are likely to agree with such statements as “I would probably make a good actor” and “In different situations and with different people, I often act like very different per- sons.” In contrast, low self-monitors are self-verifiers by nature, appearing less concerned about the social acceptability of their behavior. Like character actors always cast in the same role, they express themselves in a consistent manner from one situation to the next, exhibiting what they regard as their true and hon- est self. On the Self-Monitoring Scale, low self-monitors say, “I can only argue for ideas which I already believe” and “I have never been good at games like charades or improvisational acting” (see Table 3.3).

Social psychologists disagree on whether the Self-Monitoring Scale measures one global trait or a combination of two or more specific traits. They also dis- agree about whether high and low self-monitors represent two discrete types of people or just varying points along a continuum—a position supported by a recent analysis of test scores (Wilmott, 2015). Either way, the test scores do appear to predict important social behaviors (Gangestad & Snyder, 2000).

j Self-Monitoring Although self-presentation is a way of life for all of us, it differs considerably among individu- als. Some people are generally more conscious of their public image than others. Also, some people are more likely to engage in strategic self- presentation, whereas others seem to prefer self- verification. According to Mark Snyder (1987), these differences are related to a personality trait he called self-monitoring: the tendency to regu- late one’s own behavior to meet the demands of social situations.

Individuals who are high in self-monitoring appear to have a repertoire of selves from which to draw. Sensitive to strategic self-presentation

self-monitoring The tendency to change behavior in response to the self-presentation concerns of the situation.

m T a b l e 3 . 3

Self-monitoring Scale

Are you a high or low self-monitor? For each statement, answer True or False. When you are done, give yourself one point if you answered T to items 4, 5, 6, 8, 10, 12, 17, and 18. Then give yourself one point if you answered F to items 1, 2, 3, 7, 9, 11, 13, 14, 15, and 16. Count your total number of points. This total represents your Self-Monitoring Score. Among North American college students, the average score is about 10 or 11.

1. I find it hard to imitate the behavior of other people.

2. At parties and social gatherings, I do not attempt to do or say things that others will like.

3. I can only argue for ideas that I already believe.

4. I can make impromptu speeches even on topics about which I have almost no information.

5. I guess I put on a show to impress or entertain others.

6. I would probably make a good actor.

7. In a group of people I am rarely the center of attention.

8. In different situations and with different people, I often act like very different persons.

9. I am not particularly good at making other people like me.

10. I’m not always the person I appear to be.

11. I would not change my opinions (or the way I do things) in order to please someone or win their favor.

12. I have considered being an entertainer.

13. I have never been good at games like charades or improvisational acting.

14. I have trouble changing my behavior to suit different people and different situations.

15. At a party I let others keep the jokes and stories going.

16. I feel a bit awkward in company and do not show up quite as well as I should.

17. I can look anyone in the eye and tell a lie with a straight face (if for a right end).

18. I may deceive people by being friendly when I really dislike them.

From Snyder, M., and Gangstad, S., “On the nature of self-monitoring: Matters of assessment, matters of validity,” Journal of Personality and Social Psychology vol 51 (pp. 125–139). Copyright © 1986 by the American Psychological Association. Reprinted by permission.

80220_ch03_ptg01_hr_052-103.indd 100 27/01/16 11:21 AM

101 Reflections: The Multifaceted Self

Concerned with public image, high self-monitors go out of their way to learn about others with whom they might interact and about the rules for appropriate conduct. Then, once they have the situation sized up, they modify their behavior ac- cordingly. If a situation calls for conformity, high self-monitors conform; if the same situation calls for autonomy, they refuse to conform. By contrast, low self-monitors maintain a relatively consistent posture across a range of situations (Snyder & Monson, 1975). Unconsciously adapting to social situations, high self-monitors are likely to mimic the demeanor of others in subtle ways that facilitate smooth social interactions (Cheng & Chartrand, 2003). They are also more likely to switch dia- lects according to their local surroundings, facilitating “linguistic adaptation” (Blank et al., 2012). Consistent with the finding that high self-monitors are more concerned than lows about what other people think, research conducted in work settings shows that high self-monitors receive higher performance ratings and more promotions and that they are more likely to emerge as leaders (Day et al., 2002).

Throughout this textbook, we see that because so much of our behavior is in- fluenced by social norms, self-monitoring is relevant to many aspects of social psy- chology. There are also interesting developmental implications. A survey of 18- to 73-year-olds revealed that self-monitoring scores tend to drop with age, presumably because people become more settled and secure about who they are as they get older (Reifman et al., 1989). For now, however, ponder this question: Is it better to be a high or low self-monitor? Is one orientation inherently more adaptive than the other?

The existing research does not enable us to make this kind of value judgment. Consider high self-monitors. Quite accurately, they regard themselves as pragmatic, flexible, and adaptive and as able to cope with the diversity of life’s roles. But they could also be described as fickle or phony opportunists, more concerned with appearances than with reality and willing to change colors like a chameleon just to fit in. Now think about low self-monitors. They describe themselves as principled and forthright; they are without pretense, always speaking their minds so others know where they stand. Of course, they could also be viewed as stubborn, insensi- tive to their surroundings, and unwilling to compromise in order to get along. Con- cerning the relative value of these two orientations, then, it is safe to conclude that neither high nor low self-monitoring is necessarily undesirable—unless carried to the extreme. Goffman (1955) made the same point many years ago, when he wrote:

Too little perceptiveness, too little savoir faire, too little pride and considerateness, and the person ceases to be someone who can be trusted to take a hint about him- self or give a hint that will save others embarrassment. . . . Too much savoir faire or too much considerateness and he becomes someone who is too socialized, who leaves others with the feeling that they do not know how they really stand with him, nor what they should do to make an effective long-term adjustment. (p. 227)

Reflections: The Multifaceted Self Throughout human history, writers, poets, philosophers, and personality theorists have portrayed the self as an enduring aspect of personality, as an invisible “inner core” that is stable over time and slow to change. The struggle to find your “true self” is based on this portrait. Indeed, when people older than 85 years were asked to reflect on their lives, almost all said that despite having changed in certain ways, they had remained essentially the same person (Troll & Skaff, 1997). In recent years, however, social psychologists have focused on change. In doing so, they have discov- ered that at least part of the self is malleable—molded by life experiences and varying from one situation to the next, online and offline, in public and in private, and de- pending on cultural context. From this perspective, the self has many different faces.

It’s more adaptive to alter one’s behavior than to stay consistent from one social situation to the next.

FAlSe

80220_ch03_ptg01_hr_052-103.indd 101 27/01/16 11:21 AM

102 Chapter 3 The Social Self

1. People are especially attentive to information of rel- evance to the self and are self-conscious, as if under a “spotlight,” in the presence of other people.

2. Your self-concept represents the sum total of beliefs you have about the kind of person you are—your traits, abili- ties, motivation, and so on.

3. Often people learn about themselves not by introspec- tion but by observing their own behavior and comparing themselves to other people.

4. While most Westerners have an independent view of themselves as distinct and autonomous, people in many Asian cultures and elsewhere hold an interdependent view of the self as part of a larger social network.

5. All over the world, people have a strong need for high self-esteem and want to see themselves in a positive light.

6. When people enter a state of self-awareness—which hap- pens in front of an audience, for example, or looking in a mirror—they become their own worst critics and experi- ence a temporary drop in self-esteem.

7. In striving to meet our personal ideals, we often engage in self-regulation—trying to control our own thoughts, urges, and behaviors—which can be physically taxing.

8. Despite our shortcomings and self-critical tendencies, most people think highly of themselves thanks to vari- ous adaptive means of self-enhancement (for example, by taking credit for success but denying blame for failure or comparing to others who are less well off).

9. Through the processes of self-presentation, people put on a public face that may or may not be consistent with their private self in an effort to get others to see them in a positive light.

10. In public, individuals differ in their level of self- monitoring—some people modify what they do to suit the social situation; others behave more consistently regardless of the situation.

Top 10 Key Points in Chapter 3

Review

When you look into the mirror, what do you see: one self or many? Do you see a person whose self-concept is enduring or one whose identity seems to change from time to time? Do you see a person whose strengths and weaknesses are evaluated with an objective eye or one who is insulated from unpleasant truths by mechanisms of self-defense? Do you see a person who has an inner, hidden self that is different from the face shown to others?

Based on the material presented in this chapter, the answers seem always to be the same: The self has all these characteristics. Long before social psychology was born, William James (1890) remarked that the self is not simple but multifaceted. Based on current theories and research, we can now appreciate just how right James was. Sure, there’s an aspect of the self-concept that we can come to know only by in- trospection and that is stable over time. But there’s also an aspect that changes with the company we keep and the information we get from others. When it comes to self- esteem, there are times when we are self-focused enough to be acutely aware of our shortcomings. Yet there are also times when we guard ourselves through self-serving cognitions, self-handicapping, BIRGing, and downward social comparisons. Then there is the matter of self-presentation. It’s clear that each of us has a private self that consists of our inner thoughts, feelings, and memories. But it is equally clear that we also have an outer self, portrayed by the roles we play and the masks we wear in public. As you read through the later pages of this text, you will see that the cogni- tive, affective, and behavioral components of the self are not separate and distinct but interrelated. They are also of great significance for the rest of social psychology.

80220_ch03_ptg01_hr_052-103.indd 102 27/01/16 11:22 AM

103 Review

The Self-Concept

Humans are the only animals who recognize themselves in the mirror. F False. Studies have shown that the great apes

(chimpanzees, gorillas, and orangutans) are also capable of self-recognition.

Smiling can make you feel happier. T True. Consistent with the facial feedback hypothesis,

facial expressions can trigger or amplify the subjective experience of emotion.

Self-esteem

Sometimes the harder you try to control a thought, feeling, or behavior, the less likely you are to succeed. T True. Research on ironic processes in mental control

have revealed that trying to inhibit a thought, feeling, or behavior often backfires.

People often sabotage their own performance in order to protect their self-esteem. T True. Studies have shown that people often handicap

their own performance in order to build an excuse for anticipated failure.

Self-Presentation

It’s more adaptive to alter one’s behavior than to stay consistent from one social situation to the next.

F False. High and low self-monitors differ in the extent to which they alter their behavior to suit the situation they are in, but neither style is inherently more adaptive.

Answers to Putting Common Sense to the Test

affective forecasting (59) bask in reflected glory (bIRG) (92)

dialecticism (73) downward social comparisons (93)

facial feedback hypothesis (62) implicit egotism (89) overjustification effect (64)

private self-consciousness (83) public self-consciousness (83) self-awareness theory (81) self-concept (55) self-esteem (76) self-handicapping (91) self-monitoring (100) self-perception theory (60)

self-presentation (97) self-regulation (84) self-schema (55) social comparison theory (66) Sociometer Theory (77) Terror management Theory (77)

two-factor theory of emotion (68)

Key Terms

80220_ch03_ptg01_hr_052-103.indd 103 27/01/16 11:22 AM

Perceiving Persons

This chapter examines how people come to know (or think that they know) other persons. First, we introduce the elements of social perception—those aspects of persons, situations, and behavior that guide initial observations. Next, we examine how people make explanations, or attributions, for the behavior of others and how they form integrated impressions based on initial perceptions and attributions. We then consider various confirmation biases, the subtle ways that initial impressions resist change by leading people to distort later information, setting in motion a self-fulfilling prophecy.

80220_ch04_ptg01_hr_104-153.indd 104 30/01/16 10:53 AM

Perceiving Persons

4 Observation: The Elements of Social Perception | 107

A Person’s Physical Appearance

Perceptions of Situations

Behavioral Evidence

Detecting Truth and Deception

Attribution: From Elements to Dispositions | 120

Attribution Theories

Attribution Biases

Culture and Attribution

Motivational Biases

Integration: From Dispositions to Impressions | 134

Information Integration: The Arithmetic

Deviations From the Arithmetic

Perceptions of Moral Character

Confirmation Biases: From Impressions to Reality | 142

Perseverance of Beliefs

Confirmatory Hypothesis Testing

The Self-Fulfilling Prophecy

Social Perception: The Bottom Line | 149 Review

105

Pa tr

ik G

ia rd

in o/

Cr av

e/ Co

rb is

80220_ch04_ptg01_hr_104-153.indd 105 30/01/16 10:53 AM

106 Chapter 4 Perceiving Persons

Oscar Pistorius is a South African sprinter. Although both his legs were amputated below the knee when he was a baby, he went on with artificial limbs to become a Paralympic record-setting gold medalist. In summer of 2012, he became the first amputee to compete in the summer Olympic games. Pistorius was a national hero and an inspiration to disabled people all over the world.

Then on Valentine’s Day of 2013, Pistorius shot and killed his girlfriend, Reeva Steenkamp, in his home. He said he thought she was an intruder hiding in the locked bathroom when he fired four shots through the door. Pistorius was arrested, charged, and tried for murder. The prosecutor noted that he and Steenkamp had argued that day, that neighbors heard a woman screaming before the gunshots, and that the story of an intruder made no sense. Pistorius was emotional at trial, often breaking down in tears. Ultimately, the judge convicted Pistorius of “cul- pable homicide” by finding that he had not committed premeditated murder but rather that his killing of Steenkamp was unintentional. Pistorius was then sen- tenced to only five years in prison. While the prosecutor appealed the verdict, and over objections from Steenkamp’s parents, Pistorius was granted a release on parole. Then in November 2015, the South Africa Supreme Court overturned the original verdict and convicted Pistorius of murder.

The world was riveted to the Pistorius trial. Is it possible that a hero, so ad- mired, had shot his girlfriend on purpose, in a moment of passion, or was it an ac- cident for which a good person had suffered enough and deserved mercy? When Pistorius broke down and cried at trial, repeatedly, did his tears betray genuine heartfelt emotion or an attempt to manipulate the judge and public opinion? What information did the judge use to determine his state of mind that night? Who is Oscar Pistorius—a hero, a villain, or both?

More than a year later, on July 17, 2014, a Malaysia Airlines plane carrying 298 passengers from Amsterdam to Kuala Lumpur was shot down over eastern Ukraine. No one survived. Ukrainian officials blamed Russian-backed separatists who were in the area; Russia blamed the Ukraine. Even after incriminating tape recordings and physical evidence retrieved from the crash site suggested that the plane was shot down by a Russian surface-to-air missile, President Vladimir Putin adamantly denied direct or indirect involvement—not only in downing the plane but even in supporting the separatist uprising. As a European Union investigation into the crash proceeded, the world wondered: Did Putin know what had hap- pened? Was he telling the truth or lying about Russia’s lack of involvement? Did the people of the world, and the media, react on the basis of evidence or national bias?

Po ol

/G et

ty Im

ag es

A le

xe y

N ik

ol sk

y/ A

FP /R

ia N

ov os

ti Pr

es id

en tia

l Pr

es s

Se rv

ic e

- P oo

l/ G

et ty

Im ag

El sa

/G et

ty Im

ag es

S po

rt /G

et ty

Im ag

Recent events involving Oscar Pistorius, Vladimir Putin, and Tom Brady illustrate the kinds of questions that social perceivers often ask in trying to understand other people.

80220_ch04_ptg01_hr_104-153.indd 106 30/01/16 10:53 AM

107 Observation: The Elements of Social Perception

On January 18, 2015, the Indianapolis Colts and New England Patriots met in Foxborough, Massachusetts, to play the AFC Championship game. The winner would go on to play in the NFL Super Bowl. The Patriots led 17–7 after the first half when a columnist for Indiana television station tweeted: “Breaking: A league source tells me the NFL is in- vestigating the possibility the Patriots deflated footballs. More to come.” Although the Patriots went on to win the game by a one-sided margin of 45–7, the scandal that broke, that the Patriots had deflated their footballs, making it easier for the quarterback to grip and pass was referred to as “Deflategate” and investigated by the NFL.

For two weeks leading up to the Super Bowl, sports fans and the news media were laser-focused. The balls used by the Patriots—but not those used by the Colts—were lighter at half time than at the start of the game. The Patriots went on to win the Super Bowl and become NFL Champions. After complet- ing its investigation, the NFL concluded that the balls were deflated and that Brady probably knew about it. The NFL sus- pended Brady for four games the following year. Reactions to the results were quick and mixed. The Patriots appealed the suspension and in September of 2015 a federal judge threw it out. In response, the NFL appealed. To this day, questions linger: Were the balls intentionally deflated? If so, who did it and why? Did quarterback Tom Brady ask for it, or know about it? Throughout the ordeal, Brady denied “knowledge of wrongdoing.” Was his denial believable? What did Patriots fans think? What about fans of opposing teams?

Whatever the topic—crime, world politics, sports, business, entertainment, or personal events closer to home—we are all engaged and interested participants in social perception, the processes by which people come to understand one an- other. This chapter is divided into four sections. First we look at the “raw data” of social perception: persons, situations, and behaviors. Second, we examine how people explain and analyze behavior. Third, we consider how people integrate their various observations into a coherent impression of other persons. Fourth, we discuss some of the subtle ways that our impressions create a distorted picture of reality, often setting in motion a self-fulfilling prophecy. As you read this chapter, you’ll notice that we consider the various processes from a perceiver’s vantage point. Keep in mind, however, that in the events of life, you are both a perceiver and a target of others’ perceptions.

Observation: The Elements of Social Perception As our opening examples suggest, understanding other people may not be easy, but it is an essential and adaptive part of everyday life. How do we do it? What kinds of information do we use? We cannot actually “see” someone’s mental or emotional state or his or her motives or intentions any more than a detective can see a crime that has already been committed. So like a detective who tries to reconstruct a crime by turning up witnesses, fingerprints, blood samples, and

social perception A general term for the processes by which people come to understand one another.

Circle Your Answer

T F The impressions we form of others are influenced by superficial aspects of their appearance.

T F Adaptively, people are skilled at knowing when someone is lying rather than telling the truth.

T F Like social psychologists, people are sensitive to situational causes when explaining the behavior of others.

T F People are slow to change their first impressions on the basis of new information.

T F The notion that we can create a “self- fulfilling prophecy” by getting others to behave in ways we expect is a myth.

T F People are more accurate at judging the personalities of friends and acquaintances than of strangers

Putting Common SenSe to the Test

80220_ch04_ptg01_hr_104-153.indd 107 30/01/16 10:53 AM

108 Chapter 4 Perceiving Persons

other evidence, the social perceiver comes to know others by relying on indirect clues—the elements of social perception. These clues arise from an interplay of three sources: persons, situations, and behavior.

j A Person’s Physical Appearance Have you ever met someone for the first time, or visited their Facebook page for the first time, and formed a quick impression based only on a quick “snapshot” of information? As children, we were told that we should not judge a book by its cover, that things are not always what they seem, that surface appearances are deceiving, and that all that glitters is not gold. Yet as adults we can’t seem to help ourselves.

To illustrate the rapid-fire nature of the process, Janine Willis and Alexander Todorov (2006) showed college students photographs of unfamiliar faces for one- tenth of a second, half a second, or a full second. Whether the students judged the faces for how attractive, likable, competent, trustworthy, or aggressive they were, their ratings—even after the briefest exposure—were quick and highly correlated with judgments that other observers made without time limits (see Table 4.1). Additional research has shown that people evaluate quickly, spontaneously, and unconsciously whether a face indicates that a person is dominant or submissive, and trustworthy or untrustworthy (Stewart et al., 2012). This tendency to infer personal characteristics from the face is not only quick but early to develop. For example, young children as young as 3 and 4 years show this adult-like tendency to judge whether a person is “mean” or “nice” based on a brief exposure to the face (Cogsdill et al., 2014). There’s no doubt about it: If you flip through the pages of a magazine, or browse Snapchat, Instagram, or Pinterest, you will see for your- self that sometimes it takes a mere fraction of a second for you to form impres- sions of a stranger from his or her face.

If first impressions are quick to form, then on what are they based? In 500 b.c.e., the mathematician Pythagoras looked into the eyes of prospective students to see if they were gifted. At about the same time, Hippocrates, the founder of modern medicine, used facial features to make diagnoses of life and

death. In the nineteenth century, Viennese physician Franz Gall introduced a carnival-like science called phrenology and claimed that he could assess a person’s character by the shape of his or her skull. And in 1954, psychologist William Sheldon concluded from flawed studies of adult men that there is a strong link between physique and personality.

People may not measure each other by bumps on the head, as phrenologists used to do, but our first impressions are influenced in subtle ways by a person’s height, weight, skin color, hair color, tattoos, piercings, eyeglasses, and other aspects of physical appearance. As social perceivers, we also form impressions of people that are often accurate based on a host of indirect telltale cues. In Snoop: What Your Stuff Says About You, Sam Gosling (2008) de- scribes research he has conducted showing that people’s personalities can be revealed

Willis & Todorov, 2006.

m T a b l e 4 . 1

First Impressions in a Fraction of a Second

Traits Being Judged 0.10 sec 0.50 sec 1 sec

Trustworthy .73 .66 .74

Competent .52 .67 .59

Likable .59 .57 .63

Aggressive .52 .56 .59

Attractive .69 .57 .66

Participants rated unfamiliar faces based on pictures they saw for one-tenth of a second, half a second, or a full second. Would their impressions stay the same or change with unlimited time? As measured by the correlations of these ratings with those made by observers who had no exposure time limits, the results showed that ratings were highly correlated even at the briefest exposure times. Giving participants more time did not increase these correlations.

80220_ch04_ptg01_hr_104-153.indd 108 30/01/16 10:53 AM

109 Observation: The Elements of Social Perception

in the knick-knacks found in their offices and dormitory rooms, the identity claims they make on Facebook pages, the books that line their shelves, and the types of music that inhabit their iPods.

Other superficial cues can also lead us to form quick impres- sions. In one study, fictional characters with “old-generation” names such as Harry, Walter, Dorothy, and Edith were judged to be less popular and less intelligent than those with younger-generation names such as Kevin, Michael, Lisa, and Michelle (Young et al., 1993). In a second study, both men and women were seen as more feminine when they spoke in high-pitched voices than in lower- pitched voices (Ko et al., 2006). In a third study, cleverly entitled “I like you but I don’t know why,” people rated computer- generated faces more positively when these faces—unbeknownst to them—were created to resemble a boyfriend or girlfriend in a satisfying relationship (Günaydin et al., 2012). In a fourth study, people formed impressions of strangers they had never met— seeing them as introverted or extroverted, calm or anxious, and agreeable or disagreeable—on the basis of the cartoon-like avatars these strangers had created to represent themselves online (see d Figure 4.1) (Fong & Mar, 2015).

The human face in particular attracts more than its share of attention (Todorov et al., 2015). Since the time of ancient Greece, people have attended to physiognomy—the art of reading character from faces. Although we may not realize it, this tendency persists today. For example, Ran Hassin and Yaacov Trope (2000) found that people pre- judge others in photographs as kind-hearted rather than mean-spirited based on such features as a full, round face, curly hair, long eyelashes, large eyes, a short nose, full lips, and an upturned mouth. Interestingly, these researchers also found that just as people read traits from faces, at times they read traits into faces based on prior information. In one study, for example, participants who were told that a man was kind—compared to those told he was mean—later judged his face to be fuller, rounder, and more attractive.

In social perception studies of the human face, researchers have found that adults who have baby-faced features—large, round eyes; high eye- brows; round cheeks; a large forehead; smooth skin; and a rounded chin— tend to be seen as warm, kind, naive, weak, honest, and submissive. In contrast, adults who have mature features—small eyes, low brows and a small forehead, wrinkled skin, and an angular chin—are seen as stronger, more dominant, and more competent (Berry & Zebrowitz-McArthur, 1986). In small claims court, judges are more likely to favor baby-faced defen- dants who are accused of intentional wrongdoing but rule against them when accused of negligence. In the work setting, baby-faced job applicants are more likely to be recommended for employment as day-care teachers, whereas mature-faced adults are considered to be better suited for work as bankers. Recent research even shows that this link between baby-faced appearance and personal characteristics is seen not only in Western cul- tures but also among the Tsimané people living in a Bolivian rainforest (Zebrowitz et al., 2012). Leslie Zebrowitz and Joann Montepare (2005) thus conclude that baby-facedness “profoundly affects human behavior in the blink of an eye” (p. 1565).

d F i g u r e 4 . 1

Sample avatars used in a study showing that people form impressions of unknown others—for example, as introverted or extroverted—on the basis of what they “see” in their online visual representations (Fong & Mar, 2015).

en ga

ge L

ea rn

in g®

Ju lio

C es

ar A

gu ila

r/ A

FP /G

et ty

Im ag

What is your first impression of this person? Posing during an International Tattoo Exposition in Mexico, in 2013, Maria Jose Cristerna, also known as Vampire Woman, set the Guinness World Record for the most changes made to her body.

80220_ch04_ptg01_hr_104-153.indd 109 30/01/16 10:53 AM

110 Chapter 4 Perceiving Persons

What accounts for these findings? And why, in general, are people so quick to judge others by appearances? One explanation is that human beings are pro- grammed by evolution to respond gently to babyish features so that real babies are treated with tender loving care. Many years ago, animal behaviorist Konrad Lorenz noted that babyish features in many animal species seem to trigger a spe- cial nurturing response to cuteness. Recently, this old idea derived new support from a brain-imaging study showing that a frontal brain region associated with love and other positive emotions is activated when people are exposed, even fleet- ingly, to pictures of babies’ faces but not to pictures of the faces of other adults (Kringelbach et al., 2008).

Our reflex-like response to babies is understandable. But why would we re- spond in the same way to baby-faced adults? Leslie Zebrowitz believes that we associate babyish features with helplessness traits and then overgeneralize this ex- pectation to baby-faced adults. Consistent with this point, she and her colleagues found in a brain-imaging study that the region of the brain activated by pictures of babies’ faces was also activated by pictures of baby-faced men (Zebrowitz et al., 2009).

Other researchers also believe that people as social perceivers overgeneralize in making snap judgments. Alexander Todorov and others (2008) find that people are quick to perceive unfamiliar faces as more or less trustworthy—a judgment we must often make—and that we do so by focusing on features that resemble the expressions of happiness and anger (a trustworthy face has a U-shaped mouth and raised eyebrows; in an untrustworthy face, the mouth curls down and the eyebrows form a V). In other words, faces are seen as trustworthy if they look happy, an emotion that signals a person who is safe to approach, and untrust- worthy if they look angry, an emotion that signals danger to be avoided. Facial expressions may be temporary, but they too can influence our perceptions. When people smile, their faces look lighter and brighter than when they frown (Song et al., 2012).

j Perceptions of Situations In addition to the beliefs we hold about persons, we also have preset notions about certain types of situations—”scripts” that enable us to anticipate the goals, behaviors, and outcomes likely to occur in a particular setting (Abelson, 1981; Read, 1987). Based on past experience, people can easily imagine the sequences of events likely to unfold in a typical greeting or at a shopping mall, the dinner table, or a tennis match. The more experience you have in a given situation, the more detail your scripts will contain.

In Do’s and Taboos Around the World, Roger Axtell (1993) described scripts that are culture specific. In Bolivia, he notes, dinner guests are expected to fully clean their plates to prove that they enjoyed the meal. Eat in an Indian home, how- ever, and you’ll see that many native guests will leave some food on the plate to show the host that they had enough to eat. Social scripts of this nature can influ- ence perceptions and behavior. As we’ll see in Chapter 11 on aggression, in places that foster a culture of honor, men are expected to defend against all insult, women are expected to remain modest and loyal, and indications of female infidelity can trigger domestic violence (Vandello & Cohen, 2003). According to Angela Leung and Dov Cohen (2011), similar socials scripts can be found in other cultures, where a greater value is placed on face (the notion that people are expected to show defer- ence to others of higher status and humility in public situations) and dignity (the notion that everyone at birth has intrinsic and equal value).

“Our faces, together with our lan- guage, are social tools that help us navigate the social encounters that define our “selves” and fashion our lives.”

—Alan J. Fridlund

The impressions we form of others are influenced by superficial aspects of their appearance.

True

80220_ch04_ptg01_hr_104-153.indd 110 30/01/16 10:53 AM

111 Observation: The Elements of Social Perception

Behavioral scripts can be quite elaborate. Studying the “first date” script, John Pryor and Thomas Merluzzi (1985) asked U.S. college students to list the se- quence of events that take place in this situation. From these lists, a picture of a typical American first date emerged. Sixteen steps were identified, including (1) male arrives; (2) female greets male at door; (3) female introduces date to parents or roommate; (4) male and female discuss plans and make small talk; (5) they go to a movie; (6) they get something to eat or drink; (7) male takes female home; (8) if interested, he remarks about a future date; (9) they kiss; (10) they say good night. Sound familiar? Pryor and Merluzzi then randomized their list of events and asked participants to arrange them into the appropriate order. They found that those with extensive dating experience were able to organize the statements more quickly than those who had less dating experience. For peo- ple who are familiar with a script, the events fall into place like pieces of a puzzle. In fact, almost 30 years later, despite changes in gender and dating norms, re- search shows that this basic script has remained essentially the same (Eaton & Rose, 2011; Morr Serewicz & Gale, 2008).

Knowledge of social settings provides an important context for understand- ing other people’s verbal and nonverbal behavior. For example, this knowledge leads us to expect someone to be polite during a job interview, playful at a picnic, and rowdy at a NASCAR rally. Often our expectations for how situa- tions affect us can influence the way we interpret other people’s facial expres- sions. In one study, participants looked at photographs of human faces that had ambiguous expressions. When told that the person in the photo was being threatened by a vicious dog, they saw the facial expression as fearful; when told that the individual had just won money, participants interpreted the same expression as a sign of happiness (Trope, 1986). In other studies, scowling faces were seen as afraid when the surrounding situation was described as dangerous or as determined when said to be in a medal race on an Olympic rowing team (Carroll & Russell, 1996). Context effects on the perception of joy, anger, fear, pride, disgust, surprise, and other emotions in a facial expression—and even whether someone is audibly laughing or crying—are quick and automatic (Aviezer et al., 2008; Barrett et al., 2011; Lavan et al., 2015). d Figure 4.2 illustrates this point.

j Behavioral evidence An essential first step in social perception is recognizing what some- one is doing at a given moment. Identifying actions from move- ment is surprisingly easy. Even when actors dressed in black move about in a dark room with point lights attached only to the joints of their bodies, people quickly and easily recognize such complex acts as walking, running, jumping, and falling (Johansson et al., 1980). This ability is found in people of all cultures (Barrett et al., 2005), and it enables us to recognize specific individuals, such as friends, strictly by their movements (Loula et al., 2005).

More interesting, perhaps, is that we derive meaning from our observations by dividing the continuous stream of human behavior into discrete “units.” By having participants observe someone on videotape and press a button whenever they detect a “meaningful action,” Darren Newtson and others (1987) found that some perceivers break the behavior stream into a large num- ber of fine units, whereas others break it into a small number of

d F i g u r e 4 . 2

Judging emotions in Context

Look at the face of tennis star Serena Williams (left). How is she feeling—angry, perhaps, or in agony? Now look at her in a fuller context (right). You can see that Williams was actually euphoric, clenching her fist in victory at the 2008 U.S. Open.

Ti m

ot hy

A C

la ry

/A FP

/G et

ty Im

ag es

80220_ch04_ptg01_hr_104-153.indd 111 30/01/16 10:53 AM

112 Chapter 4 Perceiving Persons

gross units. While watching a baseball game, for example, you might press the button after each pitch, after each batter, after every inning, or only after runs are scored. The manner in which people divide a stream of behavior can influ- ence their perceptions. Research participants who were told to break an event into fine units rather than gross units attended more closely, detected more meaningful actions, and remembered more details about the actor’s behavior than did participants who were told to break events into gross units (Lassiter et al., 1988).

Mind Perception In a developing area of research, social psychologists are now interested in mind perception, the process by which people attribute humanlike mental states to various animate and inanimate objects, including other people. Stud- ies show that people who identify someone’s actions in high-level terms rather than low-level terms (for example, by describing the act of “painting a house” as “trying to make a house look new,” not just “applying brush strokes”) are also more likely to attribute humanizing thoughts, feelings, intentions, consciousness, and other states of mind to that actor (Kozak et al., 2006).

Although people do not tend to attribute mental states to inanimate objects, in general the more humanlike a target object is, the more likely we are to at- tribute to it qualities of “mind.” In a series of studies, Carey Morewedge and others (2007) found that whether people are asked to rate different animals in nature (such as a sloth, turtle, housefly, deer, wolf, and hummingbird); cartoon robots or human beings whose motion was presented in slow, medium, and fast speeds; or a purple blob oozing down a city street at the same, slower, or faster pace than the people around it, the result is always the same: People see inner qualities of mind in target objects that superficially resemble humans in their speed of movement.

“What kinds of things have minds?” To answer this question, Heather Gray and others (2007) conducted an online survey in which they presented more than 2,000 respondents with an array of human and nonhuman characters such as a 7-week-old fetus, a 5-month-old infant, an adult man, a man in a vegetative state, a dead woman, a frog, the family dog, a chimpanzee, God, and a sociable robot. They then asked respondents to rate the extent to which each character possessed various mental capacities such as pleasure, pain, fear, pride, embarrass- ment, memory, self-control, and morality. Once statistically combined, the results showed that people perceive minds along two dimensions: agency (a target’s abil- ity to plan and execute behavior) and experience (the capacity to feel pleasure, pain, and other sensations). Overall, the more “mind” respondents attributed to a character, the more they liked it, valued it, wanted to make it happy, and wanted to rescue it from destruction.

The perception of mind is an important aspect in how we perceive, connect with, and behave toward one another; that we sometimes see human mindlike qualities in nonhuman beings, and in certain inanimate objects; and that people are more likely to ascribe mind to others with whom they share a social connec- tion than to distant others (Waytz et al., 2010). In a series of studies, for example, Adam Waytz and Nicholas Epley (2012) found that research participants who were asked to reflect on someone in their lives to whom they are close—say, a good friend, significant other, or family member—were then less likely to attribute humanizing mental qualities to other people.

What about the gadgets we tuck into our hip pockets? Does an iPhone have qualities of mind? In 2012, Apple introduced a series of television commercials

mind perception The process by which people attribute humanlike mental states to various animate and inanimate objects, including other people.

80220_ch04_ptg01_hr_104-153.indd 112 30/01/16 10:53 AM

113 Observation: The Elements of Social Perception

for the iPhone that are focused on Siri, the virtual assistant that is programmed to answer questions. In one commercial, actor John Malkovich asks Siri about the meaning of life. She advises him to be nice to people, avoid eating fat, and read a good book every now and then. “And try to live together in peace and harmony with people of all creeds and nations.” Malkovich thanks Siri for the advice: “I enjoyed this chat im- mensely. You are very eloquent.” “That’s nice of you to say,” Siri responds.

As technology becomes more and more sophisticated, with robots programmed with artificial intelligence, will people come to see a humanlike mind in machines? For ex- ample, what about autonomous vehicles—self-driving cars of the near future, equipped with special cameras, maps, and collision avoidance software—that will control their own steering and speed? This technology is currently being developed and tested. In a fascinating study, Waytz, Heafner, and Epley (2014) brought 100 people into a driving simulator to “drive” a normal car; an autonomous ve- hicle that controlled steering and speed; or an autonomous vehicle enhanced by a name (Iris), gender (female), and voice (captured by human audio files). Those placed in the enhanced autonomous vehicle were more likely to see their car as smart and to sense that it could feel, anticipate, and plan a route. They were also more likely to trust the vehicle.

The Silent Language of Nonverbal Behavior Behavioral cues are used not only to identify a person’s physical actions but also to determine his or her inner states. Knowing how someone is feeling can be tricky because people often try to conceal their true emotions from others. Have you ever had to suppress your rage at someone, mask your disappointment after failure, feign surprise, make excuses, or pretend to like something just to be polite? Sometimes people come right out and tell us how they feel. But sometimes they do not tell us, they are themselves not sure, or they actively try to hide their true feelings. For these reasons, we often tune in to the silent language of nonverbal behavior.

What kinds of nonverbal cues do people use in judging how someone else is feeling? In The Expression of the Emotions in Man and Animals, naturalist Charles Darwin (1872)—whose theory of evolution transformed our understand- ing of human history—proposed that the face expresses emotion in ways that are both innate and understood by people all over the world. Contemporary re- search supports this notion. Many studies have shown that when presented with photographs similar to those on page 114, people can reliably identify at least six “primary” emotions: happiness, sadness, anger, fear, surprise, and disgust. In one study, participants from 10 different countries—Estonia, Germany, Greece, Hong Kong, Italy, Japan, Scotland, Sumatra, Turkey, and the United States— exhibited high levels of agreement in their recognition of these emotions (Ekman et al., 1987).

From one end of the world to the other, it’s clear that a smile is a smile and a frown is a frown and that just about everyone knows what they mean, even when the expressions are “put on” by actors and are not genuinely felt. But do the results fully support the claim that basic emotions are “universally” recognized from the face, or is the link culturally specific? To answer this question, Hillary Elfenbein and Nalini Ambady (2002) meta-analyzed 97 studies involving a total

If people see mind in machines, what about autonomous vehicles, like Google’s experimental self-driving car? Waytz and others (2014) put subjects into a simulated autonomous vehicle and found that they saw the car as smart and trustworthy—and as able to feel, anticipate, and plan.

TN S/

La nd

nonverbal behavior Behavior that reveals a person’s feelings without words, through facial expressions, body language, and vocal cues.

80220_ch04_ptg01_hr_104-153.indd 113 30/01/16 10:53 AM

114 Chapter 4 Perceiving Persons

of 22,148 social perceivers from 42 different countries. As shown in d Figure 4.3, they found support for both points of view. On the one hand, people all over the world are able to recognize the primary emotions from photographs of facial expressions. On the other hand, people are 9% more accurate at judging faces from their own national, ethnic, or regional groups than from members of less familiar groups—indicating that we enjoy an “in-group advantage” when it comes to knowing how those who are closest to us are feeling.

In a study that illustrates the point, Elfenbein and Ambady (2003) showed pic- tures of American faces to groups with varying degrees of exposure to Americans. As predicted, more life exposure was associated with greater accuracy, from a low of 60% among Chinese participants living in China up to 83% among Chinese liv- ing in the United States and 93% among non–Chinese Americans. When it comes to recognizing emotions in the face, it appears that familiarity breeds accuracy.

Darwin believed that the ability to recognize emotion in others has survival value for all members of a species. This hypothesis suggests that it is more impor- tant to identify some emotions than others. For example, it may be more adaptive to be wary of someone who is angry, and hence prone to lash out in violence, than of someone who is happy, a nonthreatening emotion. Indeed, studies have shown that angry faces arouse us and cause us to frown even when presented sublimi- nally and without our awareness (Dimberg et al., 2000). Illustrating what Chris- tine and Ranald Hansen (1988) called the “anger superiority effect,” researchers have found that people are quicker to spot—and slower to look away from—angry faces in a crowd than faces with neutral, non-threatening emotions (Fox et al., 2002; Horstmann & Bauland, 2006).

It is interesting that people are not always quicker to spot the threatening face of anger in a crowd than, say, a face of happiness. Our tendencies are more com- plicated in two ways. First, what we search for may be conditioned by our current motivational state. In a visual search task resembling “Where’s Waldo?” research participants who were induced to fear social rejection and loneliness were quicker

Can you tell how these individuals are feeling? If you are like most people, regardless of culture, you will have little trouble recognizing the emotions portrayed.

G ui

do A

lb er

to R

os si

/A G

E Fo

to st

oc k

A la

n W

ei ne

or bi

W ar

ne r B

ro th

er s/

Ev er

et t C

ol le

ct io

M ic

ha el

H ae

ge le

/F lir

t/ Co

rb is

Vl ad

im ir

W ra

ng el

/F ot

ol ia

L LC

80220_ch04_ptg01_hr_104-153.indd 114 30/01/16 10:53 AM

115 Observation: The Elements of Social Perception

to spot faces in diverse crowds that wore welcoming smiles than other expressions (DeWall et al., 2009). Second, while static angry faces do not necessarily capture our attention, presentations of faces in motion, which is more realistic, may well have this effect. Studies that compared happy and angry faces in a crowd that were dynamic rather than static con- firm this hypothesis. When shown neutral faces in a crowd that gradually turn positive or negative in their expression, people are quicker to spot those faces that turn angry (Ceccarini & Caudek, 2013).

Disgust is another basic emotion that has adap- tive significance. When confronted with an offensive stimulus such as a foul odor, spoiled food, feces, rotting flesh, or the sight of mutilation, people react with an aversion that shows in the way they wrinkle the nose, raise the upper lip, and gape. This visceral reaction is often accompanied by nausea; in the case of bad food, this can facilitate gagging and vomiting (Rozin & Fallon, 1987). In nature, food poisoning is a real threat, so it is adaptive for us to recognize disgust in the face of others. To illustrate, Bruno Wicker and others (2003) had 14 men watch video clips of people smelling pleasant, disgusting, or neutral odors. Af- terward, these same men were exposed to the odors themselves. If you’ve ever inhaled the sweet, floury aroma of a bakery or inserted your nose into a carton of soured milk, you’ll appre- ciate the different reactions that would appear on your face. Using fMRI, research- ers monitored activity in the participants’ brains throughout the experiment. They found that a structure in the brain known as the insula was activated not only when participants sniffed the disgusting odor but also when they watched others sniffing it. This result suggests that people more than recognize the face of dis- gust; they experience it at a neural level.

The social value of the human face is evident when we communicate by tex- ting, Twitter, or e-mail. When e-mail first became popular, the written word was often misinterpreted, especially when the writer tried to be funny or sarcastic, because we lacked the nonverbal cues that normally animate and clarify live interactions. To fill in this gap, e-mailers created smiley faces and other “emoti- cons” (emotion icons) from standard keyboard characters. A sampling of routinely used emoticons, which are meant to be viewed with one’s head tilted 90 degrees to the left, are shown in d Figure 4.4. To simplify the task, Gmail, Hotmail, AOL, Yahoo!, and other e-mail providers now offer a choice of still and animated emoti- con faces to communicate an assortment of emotions and other mental states.

Other nonverbal cues can also influence social perception, enabling us to make quick and sometimes accurate judgments of others based on thin slices of expressive behavior (Ambady & Rosenthal, 1993). But wait. Can you really tell how much eye contact a person gives to people, or how often they smile, or nod, or gesture, from only a 5-second sample of behavior? Wouldn’t it help to have more time? To answer these questions, Nora Murphy and others (2015) conducted several studies in which they taped college students, medical interns, and job ap- plicants engaged in various social interactions. After fully coding their nonverbal behaviors, they took thin slices to determine if these small samples accurately

d F i g u r e 4 . 3

How good Are People at identifying emotions in the Face?

A meta-analysis of emotion recognition studies involving 22,148 participants from 42 countries confirmed that people all over the world can recognize the six basic emotions from posed facial expressions. Based on Elfenbein, H.A., and Ambady, N. (2002). How good are people at identifying emotions in the face? Psychological Bulletin vol 128 (pp. 203–235). Copyright © 2002 by the American Psychological Association.

Happiness

Surprise

Fear

Sadness

Anger

Disgust

Overall accuracy percentages

40 1000 20 8060

80220_ch04_ptg01_hr_104-153.indd 115 30/01/16 10:53 AM

116 Chapter 4 Perceiving Persons

captured the whole. The answer was yes. You can tell how much eye contact a person tends to give in a conversation—or how much smiling, nodding, and gesturing—from a mere 30, 60, and 90 seconds of observation.

Sometimes the impressions people form from thin slices of behavior are quite accurate. In one study, people accurately judged the intelligence of strangers, as measured by actual IQ test scores, based only on hearing them read short sentences (Borkenau et al., 2004). In a second study, people rated 8- to 12-year-old children on personality characteristics based on mere thin slices of behavior; these ratings too matched the personality assessments of these children made by their own parents (Tackett et al., 2015). In a third study, people watched brief video clips from interviews of maximum security inmates. Even with only 5 seconds of exposure, their ratings significantly correlated with clinical diagnoses of the inmates’ personality disorders (Fowler et al., 2009). Taking these findings as a whole, it’s no wonder that Nalini Ambady (2010) described the quick judgments

we often make as “intuitive” and “efficient.” “Thin slicing is not an exotic gift,” notes Malcolm Gladwell (2005), author of the best seller Blink. “It is a central part of what it means to be human” (p. 43).

Eye contact, or gaze, is a particularly powerful form of nonverbal communi- cation. As social beings, people are highly attentive to eyes, often following the gaze of others. Look up, down, left, or right, and someone observing you will likely follow the direction of your eyes (Langton et al., 2000). Even 1-year-old infants tend to follow gaze, looking toward or pointing at the object of an adult researcher’s attention (Brooks & Meltzoff, 2002). Clearly, each of us is drawn like a magnet to the gaze of others. Controlled laboratory studies of this “eye contact effect” show that people who look us straight in the eye quickly draw and hold our attention, increase arousal, and activate key social areas of the brain—and that this sensitivity is present at birth (Senju & Johnson, 2009) and even when the other person’s gazing eyes are hidden behind dark sunglasses (Myllyneva & Hietanen, 2015).

Eyes have been called the “windows of the soul.” In many cultures, people assume that someone who avoids eye contact is evasive, cold, fearful, shy, or indifferent; that frequent gazing signals intimacy, sincerity, self-confidence, and respect; and that the person who stares is tense, angry, and unfriendly. If you’ve ever conversed with someone who kept looking away, as if uninterested, then you would understand why people might form negative impressions from “gaze disengagement” (Mason et al., 2005). Sometimes eye contact is interpreted in light of a preexisting relationship. If two people are friendly, frequent eye con- tact elicits a positive impression. If a relationship is not so friendly, that same eye contact is seen in negative terms. Hence, it is said that if two people lock eyes for more than a few seconds, they will either make love or kill each other (Kleinke, 1986).

Before concluding our discussion of nonverbal behavior, it is important to note Axtell’s (1993) advice, useful for travel to other countries, that there is a great

d F i g u r e 4 . 4

Some Common e-mail “emoticons”

In order to clarify meaning of their written words, e-mailers often add smiles, winks, and other face-like symbols, or emoticons, to their electronic messages. One set of emoticons is shown here; you may be familiar with others. Based on “Smileys” © 1996 O’Reilly & Associates, Inc. All rights reserved. www.oreilly.com

Wink Smirk Said smiling

Said frowning

Sardonic incredulity

Disgusted Kiss, kiss Clowning around

Said late at night

Said tongue- in-cheek

80220_ch04_ptg01_hr_104-153.indd 116 30/01/16 10:53 AM

117 Observation: The Elements of Social Perception

deal of cultural variation. In Bulgaria, nodding your head means “no,” and shaking your head sideways means “yes.” In Germany and Brazil, the American “okay” sign (forming a circle with your thumb and forefinger) is an obscene gesture. Personal-space habits also vary across cultures. Japanese people like to maintain a comfortable distance while interact- ing. But in Puerto Rico and much of Latin America, people stand very close and backing off is consid- ered an insult. Also beware of what you do with your eyes. In Latin America, locking eyes is a must, yet in Japan, too much eye contact shows a lack of respect. If you’re in the habit of stroking your cheek, you should know that in Italy, Greece, and Spain it means that you find the person you’re talking to attractive. And don’t ever touch someone’s head in predominantly Buddhist countries, especially Thai- land. The head is sacred there. Even a smile does not convey a universal impression: While smiling tends to make you look smart in Germany and China, it can have the opposite effect in Iran (Krys et al., 2014).

Different cultures also have different rules for the common greeting. In Finland, you should give a firm handshake; in France, you should loosen the grip; in Zambia, you should use your left hand to support the right; and in Bolivia, you should extend your arm if your hand is dirty. In Japan, people bow; in Thailand, they put both hands together in a praying position on the chest; and in Fiji, they smile and raise their eye- brows. In certain parts of Latin America, it is common for people to hug, embrace, and kiss upon meeting. And in most Arab countries, men greet one another by saying salaam alaykum, then shaking hands, saying kaifhalak, and kissing on the cheek.

j Detecting Truth and Deception Social perception can be tricky because people often try to hide or stretch the truth about themselves. Poker players bluff to win money, witnesses lie to pro- tect themselves, public officials make campaign promises they know they can’t keep, and acquaintances pass compliments to each other to be polite. On occasion, everyone tells something less than “the truth, the whole truth, and nothing but the truth.” Can social perceivers tell the difference? Can you tell when someone is lying?

Sigmund Freud, the founder of psychoanalysis, once said that “no mortal can keep a secret. If his lips are silent, he chatters with his finger- tips; betrayal oozes out of him at every pore” (1905, p. 94). Paul Ekman and Wallace Friesen (1974) later revised

After the 9/11 terrorist attacks in 2001, the Transportation Security Administration has trained its agents to observe the facial expressions and demeanor of airport passengers in an effort to detect signs of malintent. Thus far, there is little evidence that the judgments they make by this type of observations are accurate.

People often stretch or conceal the truth, which can make social perception a tricky endeavor.

G ar

y C.

C as

ke y/

U PI

/L an

do v

Ki m

W ar

p/ N

ew Y

or ke

r C ar

to on

/C on

de N

as t C

ol le

ct io

80220_ch04_ptg01_hr_104-153.indd 117 30/01/16 10:53 AM

118 Chapter 4 Perceiving Persons

Freud’s observation by pointing out that some pores “ooze” more than others. Ekman and Friesen proposed that some channels of communication are difficult for deceivers to control, whereas others are relatively easy. To test this hypothesis, they showed a series of films— some pleasant, others disgusting—to a group of female nurses. While watching, the nurses were instructed either to report their honest impressions of these films or to conceal their true feelings. Through the use of hidden cameras, these participants were videotaped. Others, acting as observers, then viewed the tapes and judged whether the participants had been truthful or deceptive. The results showed that judgment accuracy rates were influenced by which types of nonverbal cues the observers were exposed to. Observers who watched tapes that focused on the body were better at detecting deception than were those who saw tapes

focused on the face. The face can communicate emotion but is relatively easy for deceivers to control, unlike nervous movements of the hands and feet. Clearly, there is nothing like the wooden Pinocchio’s nose to reveal whether someone is lying or telling the truth.

This study was the first of hundreds that have been conducted to this day. In all this research, one group of participants makes truthful or deceptive statements, while another group reads the transcripts, listens to audiotapes or watches videotapes, and then tries to judge the statements. Consistently, in laboratories all over the world, results show that people are only about 54% accurate in judging truth and deception, in part because they too often ac- cept what others say at face value (Bond & DePaulo, 2006; Vrij, 2008). In fact, research shows that professionals who are specially trained and who make these kinds of judgments for a living—such as police detectives, judges, psy-

chiatrists, customs inspectors, and those who ad- minister lie-detector tests for the CIA, the FBI, and the military—are also highly prone to error (Ekman & O’Sullivan, 1991; Meissner & Kassin, 2002; Vrij, 2008; see Table 4.2).

What seems to be the problem? One hypothesis is that there is a mismatch between the behavioral cues that actually signal deception and those we use to detect deception (Zuckerman et al., 1981; DePaulo et al., 2003). Think about it. There are four channels of communication that provide potentially relevant information: the spoken word, the face, the body, and the voice. Yet when people have a reason to lie, the words they choose cannot be trusted, and they are generally able to control both the face and body (the voice is the most telling channel; when people lie, they tend to hesitate, then speed up and raise the pitch of their voice). In a survey of approximately 2,500 adults in 58 countries, an international team of researchers found that more than 70% believed that liars tend to avert their eyes—a cue that is not supported by any research. Similarly, most survey

Research on lying and its detection has shown that there is no one behavioral cue, like Pinocchio’s growing wooden nose, that can be used to signal deception.

Ev er

et t C

ol le

ct io

m T a b l e 4 . 2

Can the “experts” Distinguish Truth and Deception?

Observer groups Accuracy rates (%)

College students 52.82

CIA, FBI, and military 55.67

Police investigators 55.79

Trial judges 56.73

Psychiatrists 57.61

U.S. Secret Service agents 64.12

Lie-detection experts with experience at making judgments of truth and deception were shown brief videotapes of 10 women telling the truth or lying about their feelings. Considering that there was a 50–50 chance of guessing correctly, the accuracy rates were remarkably low. Only a sample of U.S. Secret Service agents posted a better-than-chance performance.

80220_ch04_ptg01_hr_104-153.indd 118 30/01/16 10:53 AM

119 Observation: The Elements of Social Perception

respondents believed that people squirm, stutter, fidget, and touch themselves when they lie—also cues that are not supported by the research (Global Deception Research Team, 2006).

Perhaps people rely on the wrong cues to deception, but a recent analysis of this same literature shows that the problem should be somewhat differently stated: It’s not that people make truth and lie judgments on the basis of the wrong cues, but rather the problem is that none of the behavioral cues people look for are very telling (Hartwig & Bond, 2011). In a recent meta-analysis of 144 samples containing 9,380 speakers, providing a total of 26,866 messages, and spanning more than forty years, Hartwig and Bond (2014) found that the low level of detect- ability of deception did not differ much regardless of whether the speaker was a college student or nonstudent; whether or not the speaker was highly motivated to evade detection; or whether the truths and lies told were accompanied by high or low levels of emotion.

Throughout history, people have assumed that the way to spot a liar is to watch for outward signs of stress or anxiousness in his or her behavior. Yet in important real-life situations—for example, at a high-stakes poker table, the secu- rity screening area of an airport, or a police interrogation room—innocent truth tellers are also likely to exhibit signs of stress. For this reason, researchers today are seeking a different approach. Most no- tably, Aldert Vrij, Anders Granhag, and their colleagues have theorized that lying is harder to do and requires more thinking than telling the truth (Vrij et al., 2010; Vrij & Granhag, 2012). Therefore, they argue, we should both induce and focus on behavioral cues that betray cognitive effort.

This realization has led researchers to create more chal- lenging types of interviews that could outsmart liars, force them to think harder, and struggle more, and thereby expose their deception. In one study, for example, they asked truth tellers and liars to recount their stories in reverse chronologi- cal order. This task was more effortful for the deceivers to do, which made the interviewers better able to distinguish between truths and lies (Vrij et al., 2008). In a second study, truth tellers and liars were urged to maintain eye contact with their interviewer. This added burden taxed those who were lying more it did those who were tell- ing the truth, which enabled observers to better distinguish between the two groups (Vrij et al., 2010).

In the wake of the September 11 terrorist attacks and heightened worldwide concerns about security at airports, train stations, and other public places, the ability to distinguish truths and lies is essential, potentially a matter of life and death. Yet research shows that social perceivers are misguided. Too easily se- duced by the silver tongue, the smiling face, and the restless body, we often fail to notice the quivering voice or the verbal content of what is said. Too focused on how stressed a person seems while speaking—an emotional state that afflicts not only guilty liars but innocent truth tellers who stand falsely accused—we fail to notice how much effort it takes someone to recite their story or answer a question. With social psychologists in hot pursuit of ways to improve upon human lie-detection skills—for example, by preventing individuals from think- ing too much before making their judgments (Reinhard et al., 2013) or, by con- trast, having individuals come together in groups to discuss and make these judgments (Klein & Epley, 2015)—stay tuned for further developments in years to come.

Sweden’s Martin Jacobson won $10 million as the 2014 Champion of the World Series of Poker. As with all great players, Jacobson wears the proverbial “poker face” while playing so as not to betray how he feels about his hand.

Adaptively, people are skilled at knowing when someone is lying rather than telling the truth.

FALSe

Jo e

G iro

n/ Co

rb is

E nt

er ta

in m

en t/

Co rb

80220_ch04_ptg01_hr_104-153.indd 119 30/01/16 10:53 AM

120 Chapter 4 Perceiving Persons

To interact effectively with others, we need to know how they feel and when they can be trusted. But to understand people well enough to predict their future be- havior, we also try to identify their inner dispositions—stable characteristics such as personality traits, attitudes, and abilities. Since we cannot actually see disposi- tions, we infer them indirectly from what a person says and does. In this section, we look at the processes that lead us to make these inferences.

j Attribution Theories Do you ever think about the influence you have on other people? What about the roles of heredity, childhood experiences, and social forces? Do you wonder why some people succeed while others fail? Individuals differ in the extent to which they feel a need to explain the events of human behavior (Weary &

Edwards, 1994). Although there are vast differences among us, people in general tend to ask “why?” when confronting events that are important; harm- ful, immoral, or otherwise negative; or unexpected (Chakroff & Young, 2015; Weiner, 1985)—and when understanding these events has personal relevance (Malle & Knobe, 1997).

To make sense of our social world, we try to under- stand the causes of other people’s behavior. But what kinds of explanations do we make, and how do we go about making them? In a classic book entitled The Psy- chology of Interpersonal Relations, Fritz Heider (1958) took the first step toward answering these questions. To Heider, we are all scientists of a sort. Motivated to understand others well enough to manage our social

lives, we observe, analyze, and explain their behavior. The explanations we come up with are called attributions, and the theory that describes the process is called attribution theory. The questions posed at the beginning of the chapter regarding the behaviors of Vladimir Putin, Oscar Pistorius, and Tom Brady are all questions of attribution.

Ask people to explain why their fellow human beings behave as they do—why they succeed or fail, laugh or cry, work or play, or help or hurt others—and you’ll see that they come up with complex explanations focused on whether the behav- ior they observe is intentional or unintentional—and whether it reflects on the actor’s desire, belief, and personality (Malle & Holbrook, 2012). Interested in how people answer these kinds of why questions, Heider found it particularly useful to group the causal attributions people give into two categories: personal and situ- ational. In the Oscar Pistorius shooting incident, everyone wanted to know what caused him to pick up his gun in the middle of the night and spray four bullets into the bathroom door. The prosecution pointed the finger of blame at Pistorius, an intensely driven young man overheard yelling at his girlfriend (a personal attribution). Yet others speculated that his actions were provoked by what he thought to be the sound of an intruder inside his house and a state of confusion that ensued (a situational attribution).

The task for the attribution theorist is not to determine the true causes of such an event but rather to understand people’s perceptions of causality. Heider’s

Attribution: From Elements to Dispositions

People make personal and situational attributions all the time in an effort to make sense of their social world. But what kind of attribution is being made here?

Ro be

rt M

an ko

ff /T

he N

ew Y

or ke

r C ol

le ct

io n/

Ca rt

oo n

Ba nk

.c om

“It’s not you, Frank, it’s me—I don’t like you.”

attribution theory A group of theories that describe how people explain the causes of behavior.

personal attribution Attribution to internal characteristics of an actor, such as ability, personality, mood, or effort.

situational attribution Attribution to factors external to an actor, such as the task, other people, or luck.

80220_ch04_ptg01_hr_104-153.indd 120 30/01/16 10:53 AM

121 Attribution: From Elements to Dispositions

insights provided an initial spark for a number of for- mal models that together came to be known as attribu- tion theory (Weiner, 2008). For now, we describe two of these theories.

Jones’s Correspondent inference Theory Accord- ing to Edward Jones and Keith Davis (1965), each of us tries to understand other people by observing and ana- lyzing their behavior. Jones and Davis’s correspondent inference theory predicts that people try to infer from an action whether the act corresponds to an enduring personal trait of the actor. Is the person who commits an act of aggression a beast? Is the person who donates money to charity an altruist? To answer these kinds of questions, people make inferences on the basis of three factors.

The first factor is a person’s degree of choice. Be- havior that is freely chosen is more informative about a person than behavior that is coerced by the situ- ation. In one study, participants read a speech, pre- sumably written by a college student, which either favored or opposed Fidel Castro, then the former com- munist leader of Cuba. Some participants were told that the student had freely chosen this position and others were told that the student had been assigned the position by a professor. When asked to judge the student’s true attitude, participants were more likely to assume a correspondence between the essay (be- havior) and the student’s attitude (disposition) when the student had had a choice than when he or she had been assigned to the role (Jones & Harris, 1967; see d Figure 4.5). Keep this study in mind. It supports correspondent inference theory, but as we’ll see later, it also demonstrates one of the most tenacious biases of social perception.

A second factor that leads us to make dispositional inferences is the expect- edness of behavior. As previously noted, an action tells us more about a person when it departs from the norm than when it is typical, part of a social role, or oth- erwise expected under the circumstances (Jones et al., 1961). Thus, people think they know more about a student who wears three-piece suits to class or a citizen who openly refuses to pay taxes than about a student who wears blue jeans to class or a citizen who dutifully files tax returns on April 15.

Third, social perceivers take into account the intended effects or consequences of someone’s behavior. Acts that produce many desirable outcomes do not reveal a person’s specific motives as clearly as acts that produce only a single desir- able outcome (Newtson, 1974). For example, you are likely to be uncertain about exactly why a person stays on a job that is enjoyable, high paying, and in an at- tractive location—three highly desirable outcomes, each sufficient to explain the behavior. In contrast, you may feel more certain about why a person stays on a job that is tedious and low paying but is in an attractive location—only one desir- able outcome.

d F i g u r e 4 . 5

What Does This Speechwriter really Believe?

As predicted by correspondent inference theory, participants who read a student’s speech (behavior) were more likely to assume that it reflected the student’s true attitude (disposition) when the position taken was freely chosen (left) rather than assigned (right). But also note the evidence for the fundamental attribution error. Even participants who thought the student had been assigned a position inferred the student’s attitude from the speech. Based on Jones, E. E., & Harris, V. A. (1967). The attribution of attitudes. Journal of Experimental Social Psychology, 3, 1–24. Copyright © 1967 Elsevier.

Student chooses position

Student is assigned position

Ra tin

g of

th e

St ud

en ts

' pr

o- C

as tr

o at

tit ud

Pro-Castro speech Anti-Castro speech

80220_ch04_ptg01_hr_104-153.indd 121 30/01/16 10:53 AM

122 Chapter 4 Perceiving Persons

Kelley’s Covariation Theory Correspondent inference theory seeks to describe how perceivers try to discern an individual’s personal characteristics from a slice of behavioral evidence. However, behavior can be attributed not only to personal fac- tors but to situational factors as well. How is this distinction made? In the opening chapter, we noted that the causes of human behavior can be derived only through experiments. That is, one has to make more than a single observation and compare behavior in two or more settings in which everything stays the same except for the independent variables. Like Heider, Harold Kelley (1967) theorized that people are much like scientists in this regard. They may not observe others in a controlled laboratory, but they too search for clues, make comparisons, and think in terms of “experiments.”

According to Kelley, people make attributions by using the covariation principle: In order for something to be the cause of a behavior, it must be present when the behavior occurs and absent when it does not. Three kinds of covariation information in particular are useful: consensus, distinctiveness, and consistency. To illustrate these concepts, imagine you are standing on a street corner one hot, steamy evening minding your own business, when all of a sudden a stranger comes out of an air-conditioned movie theater and blurts out, “Great flick!” Look- ing up, you don’t recognize the movie title, so you wonder what to make of this “recommendation.” Was the behavior (the rave review) caused by something about the person (the stranger), the stimulus (the film), or the circumstances (say, the comfortable theater)? If you are possibly interested in a night at the movies, how would you proceed to explain what happened? What kinds of information would you want to obtain?

Thinking like a scientist, you might seek out consensus information to see how different persons react to the same stimulus. In other words, what do other moviegoers think about this film? If others also rave about it, then this stranger’s behavior is high in consensus and is attributed to the stimulus. If others are criti- cal of this film, however, then the behavior is low in consensus and is attributed to the person.

Still thinking like a scientist, you might also want distinctiveness information to see how the same person reacts to different stimuli. In other words, what does this moviegoer think of other films? If the stranger is generally critical of other films, then the target behavior is high in distinctiveness and is attributed to the stimulus. If the stranger raves about everything he or she sees, however, then the behavior is low in distinctiveness and is attributed to the person.

Finally, you might want to seek consistency information to see what happens to the behavior at another time when the person and the stimulus both remain the same. How does this moviegoer feel about this film on other occasions? If the stranger raves about the film on video as well as in the theater, regardless of surroundings, then the behavior is high in consistency. If the stranger does not always enjoy the film, the behavior is low in consistency. According to Kelley, behavior that is consistent is attributed to the stimulus when consensus and dis- tinctiveness are also high and to the person when they are low. In contrast, behav- ior that is low in consistency is attributed to transient circumstances, such as the temperature of the movie theater.

Does Kelley’s attribution theory describe the kinds of information you seek when you try to determine what causes people to behave as they do? Often it does. Research shows that when people are asked to make attributions for various events, they tend to follow the logic of covariation so long as it is socially safe to do so (Fosterling, 1992; McArthur, 1972; Quayle & Naidoo, 2012). However, this research also shows that individuals have their own attributional styles, so

covariation principle A principle of attribution theory that holds that people attribute behavior to factors that are present when a behavior occurs and are absent when it does not.

80220_ch04_ptg01_hr_104-153.indd 122 30/01/16 10:53 AM

123 Attribution: From Elements to Dispositions

people often disagree about what caused a particular behavior (Robins et al., 2004). There are two ways that social perceivers differ. First, we vary in the extent to which we believe that human behaviors are caused by personal traits that are fixed (“Everyone is a certain kind of person; there is not much that can be done to really change that”) or by characteristics that are malleable (“People can change even their most basic qualities”) (Dweck, 2012). Second, some of us are more likely than others to process new information in ways that are colored by self- serving motives (von Hippel et al., 2005).

j Attribution Biases Daniel Kahneman won a Nobel Prize in Economics for work on the psychology of judgment and decision making. In his recent book, Thinking, Fast and Slow, Kahneman (2011) summarizes a lifetime of research showing that the human mind operates by two different systems of thought. System 1 is quick, easy, and automatic—using a process that one might call “intuitive.” Determining which of two objects is more distant, detecting anger in a face, adding 2 1 2, and un- derstanding a simple sentence are the kinds of automatic activities engaged by this system. In contrast, System 2 is slow, controlled, and requires attention and effort—using a process that feels more reasoned. Looking for a specific face in a crowd, parking in a narrow space, counting the number of letters on a page, figuring out how a magic trick works, and filling out taxes are the kinds of activi- ties that require your focused and undivided attention. According to Kahneman, Systems 1 and 2 are both active when people are awake. System 1 runs automati- cally and guides us until it runs into difficulty, as when something unexpected happens. At that point the more effortful System 2 is activated.

When the theories of attribution were first proposed, they were represented by elaborate flow charts, formulas, and diagrams, leading many social psycholo- gists to wonder: Do people really analyze behavior in the way that one might expect of scientists, or computers? Do people have the time, the motivation, or the cognitive capacity for such elaborate, mindful, System 1 processes? The answer is sometimes yes, sometimes no. As social perceivers, we are limited in our ability to process all relevant information, or we lack the kinds of training needed to employ fully the principles of attribution theory. More important, we often don’t bother to think carefully about the attributions we make. With so much to explain and not enough time in a day, people take mental shortcuts, cross their fingers, hope for the best, and get on with life. The problem is that speed brings bias and perhaps even a loss of accuracy. In this section, we examine some of these shortcuts and their consequences.

Cognitive Heuristics According to Kahneman, Tversky, and others, people often make attributions and other types of social judgments using “cognitive heuristics”— information-processing rules of thumb that enable us to think in ways that are quick and easy but that often lead to error (Gigerenzer et al., 2011; Gilovich et al., 2002; Kahneman et al., 1982; Nisbett & Ross, 1980).

One rule of thumb that has particularly troublesome effects on attribution is the availability heuristic, a tendency to estimate the odds that an event will occur by how easily instances of it pop to mind. To demonstrate, Tversky and Kahneman (1973) asked research participants: Which is more common, words that start with the letter r or words that contain r as the third letter? In actuality, the English language has many more words with r as the third letter than as the first. Yet most people guessed that more words begin with r. Why? Because it’s

availability heuristic The tendency to estimate the likelihood that an event will occur by how easily instances of it come to mind.

80220_ch04_ptg01_hr_104-153.indd 123 30/01/16 10:53 AM

124 Chapter 4 Perceiving Persons

easier to bring to mind words in which r appears first. Ap- parently, our estimates of likelihood are heavily influenced by events that are readily available in memory (MacLeod & Campbell, 1992).

Research shows that the availability heuristic can lead us astray in two ways. First, it gives rise to the false-consensus effect, a tendency for people to overestimate the extent to which others share their opinions, attributes, and behaviors. This bias is pervasive. Regardless of whether people are asked to predict how others feel about military spending, abortion, gun control, Campbell’s soup, certain types of music, favorite celebrities, or norms for appropriate behavior, they exagger- ate the percentage of others who behave similarly or share their views (Bui, 2012; Krueger, 1998; Ross et al., 1977). To illustrate the effect, Joachim Krueger (2000) asked partici- pants in a study to indicate whether or not they had certain personality traits. Then they were asked to estimate the per- centage of people in general who have these same traits. As shown in Table 4.3, participants’ beliefs about other people’s personalities were biased by their own self-perceptions. In part, the false-consensus bias is a by-product of the availabil-

ity heuristic. We tend to associate with others who are like us in important ways, so we are more likely to notice and recall instances of similar rather than dissimi- lar behavior.

A second consequence of the availability heuristic is that social perceptions are influenced more by one vivid life story than by hard statistical facts. Have you ever wondered why so many people buy lottery tickets despite the astonish- ingly low odds or why so many travelers are afraid to fly even though they are more likely to perish in a car accident? These behaviors are symptomatic of the base-rate fallacy—the fact that people are relatively insensitive to numerical base rates, or probabilities; they are influenced more by graphic, dramatic events such as the sight of a multimillion-dollar lottery winner celebrating on TV or a photo- graph of bodies being pulled from the wreckage of a plane crash.

The base-rate fallacy can lead to various misperceptions of risk. For that rea- son, people overestimate the number of those who die in shootings, fires, floods, and terrorist bombings and underestimate the death toll caused by heart attacks, strokes, diabetes, and other mundane events. Perceptions of risk seem more rel- evant now than in the past by newly acquired fears of terrorism, or economic col- lapse, and research shows that such perceptions are affected more by fear, anxiety, and other emotions than by cold and objective probabilities (Loewenstein et al., 2001; Pachur et al., 2012; Slovic, 2000). At times the effect of strong emotion can be capricious and downright irrational. For example, consistent with the fact that people tend to fear things that sound unfamiliar, participants in one study rated fictional food additives as more hazardous to health when the names were hard to pronounce—such as Hnegripitrom, than when they were easier to pronounce— such as Magnalroxate (Song & Schwarz, 2009).

Every day, we are besieged by both types of information: We read about the jobless and unemployment rates, and we watch personal interviews with people who were recently laid off; we read the casualty figures of war, and we witness the agony of a parent who has lost a child in combat. Logically, statistics that summarize the experiences of large numbers of people are more informative than a single and perhaps atypical case, but perceivers are not influenced by numbers.

“A single death is a tragedy; a million is a statistic.”

—Joseph Stalin

false-consensus effect The tendency for people to overestimate the extent to which others share their opinions, attributes, and behaviors.

base-rate fallacy The finding that people are relatively insensitive to consensus information presented in the form of numerical base rates.

Krueger (2000).

m T a b l e 4 . 3

The False-Consensus effect

Traits Self Yes (%) Self No (%)

Alert 75 65

Discontented 48 33

Loud 46 43

Meticulous 52 41

Sly 36 28

Smug 41 33

In this study, participants who did and participants who did not rate various personality traits as descriptive of themselves estimated the percentage of other people who had these traits. As shown below, participants’ estimates of the population consensus were biased by their own self-perceptions.

80220_ch04_ptg01_hr_104-153.indd 124 30/01/16 10:53 AM

125 Attribution: From Elements to Dispositions

As long as the personal anecdote is seen as relevant (Schwarz et al., 1991) and the source as credible (Hinsz et al., 1988), and particularly to the extent that it involves someone we know (Hills & Pachur, 2012), it seems that one vivid image is worth a thousand numbers.

Counterfactual Thinking Interestingly, people can also be influenced by how easy it is to imagine events that did not occur. As thoughtful and curious beings, we are often not content to accept what happens to us or to others without wondering, at least in private, “What if . . . ?”

According to Kahneman and Miller (1986), people’s emotional reactions to events are often colored by counterfactual thinking, a tendency to imagine alter- native outcomes that might have occurred but did not. There are different types of counterfactual thoughts. If we imagine a result that is better than the actual result, then we’re likely to experience disappointment, regret, and frustration. If the imagined result is worse, then we react with emotions that range from relief and satisfaction to elation. Thus, the psychological impact of positive and nega- tive events depends on the way we think about “what might have been” (Roese, 1997; Roese & Olson, 1995).

What domains of life trigger the most counterfactual thinking—and the regret that may follow? Summarizing past research, Neal Roese and Amy Summerville (2005) found that people’s top three regrets center, in order, on education (“I should have stayed in school”), career (“If only I had applied for that job”), and romance (“If only I had asked her out”)—all domains that present us with opportunities that we may or may not realize. Of course, counterfactual thinking does not necessarily lead people into a sense of regret. In fact, Laura Kray and others (2010) note that reflecting on “what might have been” can sometimes help us to define ourselves as well as the meaning in our lives. Think about it. What would you be like today if you had chosen to attend a different school, or if you had not met your best friend? These researchers found that participants who were asked to reflect on these counterfactual “what if” questions later saw their college choices and friendships as more meaningful than those who did not stop to ask these questions. Perhaps by thinking about what might have been, people can more fully appreciate what they have.

Some individuals think in counterfactual terms more than others do. As you might expect, for example, people who strongly believe in free will— the idea that we can control our own fate by the choices we make—are more likely to wonder “what if” than those who believe that their fate is predetermined or unpredictable (Alquist et al., 2015). Also, some types of experiences lead people to engage in counterfactual thought more than others. Research shows that we are more likely to think about what might have been, often with feelings of regret, after negative outcomes that result from actions we take rather than from actions we don’t take (Byrne & McEleney 2000). Consider an experience that may sound familiar: You take a multiple-choice test, and after reviewing an item you struggled over, you want to change the answer. What do you do? Research has shown that most changes in test answers are from incorrect to correct. Yet most college students harbor the “first instinct fallacy” that it is best to stick with one’s original answer. Why? Justin Kruger and his colleagues (2005) found that this myth arises from counterfactual thinking: that students are more likely to react with regret and frustration (“If only I had . . .”) after changing a correct answer than after failing to change an incorrect answer.

Country singer Lee Brice wrote a song about counterfactual thinking. In A Woman Like You, his girlfriend asks, “Honey, what would you do if you’d never met me?” Here’s part of his response:

I’d do a lot more offshore fishin’ I’d probably eat more drive-thru chicken Take a few strokes off my golf game If I’d have never known your name I’d still be driving that old green Nova I probably never would have heard of yoga Be a better football fan But if I was a single man Alone and out there on the loose Well I’d be looking for a woman like you.

Ri ck

D ia

m on

d/ G

et ty

Im ag

es E

nt er

ta in

m en

t/ G

et ty

Im ag

counterfactual thinking The tendency to imagine alternative events or outcomes that might have occurred but did not.

80220_ch04_ptg01_hr_104-153.indd 125 30/01/16 10:53 AM

126 Chapter 4 Perceiving Persons

According to Victoria Medvec and Kenneth Savitsky (1997), certain situations—such as being on the verge of a better or worse outcome, just above or below some cutoff point—also make it especially easy to conjure up images of what might have been. The implications are intriguing. Imagine, for example, that you are an Olympic athlete and have just won a silver medal—a remarkable feat. Now imagine that you have just won the bronze medal. Which situation would make you feel better? Rationally speaking, you should feel more pride and satisfaction with a silver medal. But what if your achievement had prompted you to engage in counterfactual thinking? What alternative would haunt your mind if you had finished in second place? Where would your focus be if you had placed third? Is it possible that the athlete who is better off objectively will feel worse?

To examine this question, Medvec and others (1995) videotaped 41 athletes in the 1992 summer Olympic Games at the moment they realized that they had won a silver or a bronze medal and again, later, during the medal ceremony. Then they showed these tapes, without sound, to people who did not know the order of finish. These participants were asked to observe the medalists and rate their emotional states on a scale ranging from “agony” to “ecstasy.” The intriguing result, as you might expect, was that the bronze medalists, on average, seemed happier than the silver medalists. Was there any more direct evidence of counterfactual thinking? In

a second study, participants who watched interviews with many of these same athletes rated the silver medalists as more negatively focused on finishing second rather than first and the bronze medalists as more positively focused on finish- ing third rather than fourth. For these world-class athletes, feelings of satisfaction were based more on their thoughts of what might have been than on the reality of what was.

The Fundamental Attribution error By the time you finish reading this text- book, you will have learned the cardinal lesson of social psychology: People are profoundly influenced by the situational contexts of their behavior—or, as Samuel Sommers (2011) put it, Situations Matter. This point is not as obvious as it may seem. For instance, parents are often surprised to hear that their impossibly mischievous child is a perfect angel in the classroom. And students are often surprised to observe that their favorite professor, so eloquent in the lecture hall, may stumble over words in less formal gatherings. These reactions are symptomatic of a well-documented aspect of social perception. When people explain the behavior of others, they tend to overestimate the role of personal factors and overlook the impact of situations. Because this bias is so pervasive (and sometimes so misleading) it has been called the fundamental attribution error (Ross, 1977).

Evidence of the fundamental attribution error was first reported in the Jones and Harris (1967) study described earlier, in which participants read an essay presumably written by a student. In that study, participants were more likely to infer the student’s true attitude when the position taken had been freely chosen than when they thought that the student had been assigned to it. But look again at Figure 4.5, and you’ll notice that even when participants thought that the student had no choice but to assert a position, they still used the speech to infer his or her attitude. This finding has been repeated many times. Whether the essay topic is nuclear power, abortion, drug laws, or the death penalty, the results are essentially the same (Jones, 1990).

“During the 1996 Olympics, Nike ran a counterfactual (and contro- versial) ad: ‘You don’t win silver, you lose gold.’”

Entering the 2012 Summer Olympics in London, U.S. gymnast McKayla Maroney was the best female vaulter in the world. She went on to win a silver medal. In what fast became an iconic image, here she is during the podium ceremony. Thinking about what might have been, Maroney, clearly, was focused on not winning gold.

Ro na

ld M

ar tin

ez /G

et ty

Im ag

fundamental attribution error The tendency to focus on the role of personal causes and underestimate the impact of situations on other people’s behavior.

80220_ch04_ptg01_hr_104-153.indd 126 30/01/16 10:53 AM

127 Attribution: From Elements to Dispositions

People fall prey to the fundamental attribution error even when they are fully aware of the situation’s impact on behavior. In one experiment, the participants were themselves assigned to take a position, whereupon they swapped essays and rated each other. Remarkably, they still jumped to conclusions about each other’s attitudes (Miller et al., 1981). In another experiment, participants inferred atti- tudes from a speech even when they were the ones who had assigned the position to be taken (Gilbert & Jones, 1986).

A fascinating study by Lee Ross and his colleagues (1977) demonstrates the fundamental attribution error in a familiar setting, the TV quiz show. By a flip of the coin, participants in this study were randomly assigned to play the role of either the questioner or the contestant in a quiz game while spectators looked on. In front of the contestant and spectators, the experimenter instructed each questioner to write 10 challenging questions from his or her own store of general knowledge. If you are a trivia buff, you can imagine how esoteric such questions can be: Who was the founder of eBay? What team won the NHL Stanley Cup in 1976? It is no wonder that contestants correctly answered only about 40% of the questions asked. When the game was over, all participants rated the questioner’s and contestant’s general knowledge on a scale of 0 to 100.

Picture the events that transpired. The questioners appeared more knowledgeable than the contestants. After all, they knew all the answers. But a moment’s reflection should remind us that the situation put the questioner at a distinct advantage (there were no dif- ferences between the two groups on an objective test of general knowledge). Did participants take the question- er’s advantage into account, or did they assume that the questioners actually had greater knowledge? The results were startling. Spectators rated the questioners as above average in their general knowledge and the contestants as below average. The contestants even rated themselves as inferior to their partners. Like the spectators, they too were fooled by the loaded situation (see d Figure 4.6).

What’s going on here? Why do social perceivers consistently make assumptions about persons and fail to appreciate the impact of situations? According to Daniel Gilbert and Patrick Malone (1995), the problem stems in part from how we make attributions. Attribution theo- rists used to assume that people survey all the evidence and then decide on whether to make a personal or a situ- ational attribution. Instead, it appears that social percep- tion is a two-step process: First, we identify the behavior and make a quick personal attribution; then we correct or adjust that inference to account for situational influ- ences. At least for those raised in a Western culture, the first step is simple and automatic, like a reflex; the sec- ond requires attention, thought, and effort.

Several research findings support this hypothesis. First, without realizing it, people often form impres- sions of others based on a quick glimpse at a face or fleeting sample of behavior (Kressel & Uleman, 2015; Uleman et al., 2012). Second, perceivers are more likely

d F i g u r e 4 . 6

Fundamental Attribution error and the TV Quiz Show

Even though the simulated quiz show situation placed questioners in an obvious position of advantage over contestants, observers rated the questioners as more knowledgeable (right). Questioners did not overrate their general knowledge (left), but contestants rated themselves as inferior (middle) and observers rated them as inferior as well. These results illustrate the fundamental attribution error. Based on Ross, L., Amabile, T.M., & Steinmetz, J.L. (1977). Social roles, social control, and biases in social-perception processes. Journal of Personality and Social Psychology, 35, 485–494.

Questioners' ratings

Contestants' ratings

Observers' ratings

Ra tin

g of

g en

er al

k no

w le

dg e

Questioner Contestant

Midpoint

80220_ch04_ptg01_hr_104-153.indd 127 30/01/16 10:53 AM

128 Chapter 4 Perceiving Persons

to commit the fundamental attribution error when they are cognitively busy, or distracted, as they observe the target person than when they pay full attention (Gilbert et al., 1992; Trope & Alfieri, 1997). Since the two-step model predicts that personal attributions are automatic but that the later adjustment for situational factors requires conscious thought, it makes sense to suggest that when attention is divided, when the attribu- tion is made hastily, or when perceivers lack motivation, the second step suffers more than the first. As Gilbert and his colleagues (1988) put it, “The first step is a snap, but the second one’s a doozy” (p. 738).

Why is the first step such a snap, and why does it seem so natural for people to assume a link between acts and personal dispositions? One possible reason is based on Heider’s (1958) insight that people see others’ dispositions in behavior because of a perceptual bias, something like an optical illusion. When you listen to a speech or watch a show, the actor is the conspicuous figure of your attention; the situation fades into the back- ground (“out of sight, out of mind”). According to Heider, people attribute events to factors that are perceptually conspicuous, or salient. To test this hypothesis, Shelley Taylor and Susan Fiske (1975) varied the seating ar- rangements of observers who watched as two actors engaged in a care- fully staged conversation. In each session, the participants were seated so that they faced actor A, actor B, or both actors. When later questioned about their observations, most participants rated the actor they faced as

the more dominant member of the pair, the one who set the tone and direction.

j Culture and Attribution In the fifth century b.c.e., Herodotus, a Greek historian, argued that Greeks and Egyptians thought differently because the Greeks wrote from left to right and the Egyptians from right to left. Many years later, inspired by anthropologist Edward Sapir, Benjamin Lee Whorf (1956) theorized that the language people speak—the words, the rules, and so on—can influence the way they conceptualize the world. To illustrate, he pointed to cultural variations in the use of words to represent reality. He noted that the Hanunoo of the Philippines have 92 different terms for rice, in contrast to the crude distinction North Americans make between “white rice” and “brown rice.” Similarly, while English speakers have one word for snow, Eskimos have sev- eral words, which, Whorf argued, enables them to make distinctions that others may miss between “falling snow, snow on the ground, snow packed hard like ice, slushy snow, wind-driven flying snow—whatever the situation may be” (p. 216).

As a result of many years of research, it is now clear that language and cul- ture can influence the way people think about time, space, objects, and other aspects of the physical world around them. Consider our perceptions of color. The rainbow is a continuum of light varying smoothly between the shortest and lon- gest wavelengths of the visible spectrum. Yet when we look at it, we see distinct classes of color that correspond to “red,” “orange,” “yellow,” “green,” “blue,” and so on. Languages differ in the parts of the color spectrum that are named. In Papua, New Guinea, where Berinmo speakers distinguish between green and brown (they single out a form of “khaki” as the color of dead leaves), an object re- flecting light at 450 nanometers would be called green. Yet many English speakers, who distinguish between colors that cross the blue-green part of the spectrum, might see that same object as blue (Özgen, 2004).

Just as culture influences the way we perceive the physical world, it can also influence the way we view individuals and their place in the social world around

How knowledgeable is this man? Alex Trebek has hosted the TV quiz show Jeopardy! since 1984. As host, Trebek reads questions to contestants and then reveals the correct answers. In light of the quiz show study by Ross and others (1977), which illustrates the fundamental attribution error, viewers probably see Trebek as highly knowledgeable, despite knowing that the answers he recites are provided to him as part of his job.

Ro th

S to

ck /E

ve re

tt C

ol le

ct io

80220_ch04_ptg01_hr_104-153.indd 128 30/01/16 10:53 AM

129 Attribution: From Elements to Dispositions

them. Hence, although attribution researchers used to assume that people all over the world explained human behavior in the same ways, it is now clear that cul- tures shape in subtle but profound ways the kinds of attributions we make about people, their behavior, and social situations (Nisbett, 2003).

Culture and the Fundamental Attribution error Consider the contrasting orientations between Western “individualist” cultures (whose members tend to be- lieve that persons are autonomous, motivated by internal forces, and responsible for their own actions) and non-Western “collectivist” cultures (whose members take a more holistic view that emphasizes the relationship between persons and their sur- roundings). Do these differing worldviews influence the attributions we make? Is the fundamental attribution error is a uniquely Western phenomenon?

To answer these questions, Joan Miller (1984) asked Americans and Asian Indians of varying ages to describe the causes of positive and negative behaviors they had observed in their lives. Among young children, there were no cultural differences. With increasing age, however, the American participants made more personal attributions, while the Indians made more situational attributions (see d Figure 4.7). Testing this hypothesis in different ways, other studies as well have revealed that people form habits of thought, learning to make attributions accord- ing to culturally formed beliefs about the causes of human behavior (Miller et al., 2011; Miyamoto & Kitayama, 2002; Na & Kitayama, 2011).

Like social psychologists, people are sensitive to situational causes when explaining the behavior of others.

FALSe

d F i g u r e 4 . 7

Fundamental Attribution error: A Western Bias?

American and Asian Indian participants of varying ages described the causes of negative actions they had observed. Among young children, there were no cultural differences. With increasing age, however, Americans made more personal attributions and Indian participants made more situational attributions. Explanations for positive behaviors followed a similar pattern. This finding suggests that the fundamental error is a Western phenomenon. Based on Miller, J. G. (1984). Culture and the development of everyday social explanation. Journal of Personality and Social Psychology, 46, 961–978. Copyright © 1984 by the American Psychological Association.

Age Age

Adult

Pr op

or tio

n of

p er

so na

l a nd

tu at

io na

l a ttr

ib ut

io ns

Situational Attributions Personal Attributions

11 158 8 Adult11 15

American participants Indian participants

80220_ch04_ptg01_hr_104-153.indd 129 30/01/16 10:53 AM

130 Chapter 4 Perceiving Persons

In all cultures of the world, individual members vary in their wealth, material possessions, education, and prestige. These differences in social class can be seen as yet another cultural influence on attributions. People in upper social classes, relative to those less fortunate, have more choices to make, more opportunities, and greater control over the lives (Fiske & Markus, 2012; Kraus et al., 2012). Consistent with these differences, recent research shows that people in the upper social classes are more likely than those in the lower classes to see behavior in general as caused by internal personal traits (Varnum et al., 2012).

Focal Objects and Backgrounds Ara Norenzayan and Richard Nisbett (2000) note that cultural differences in attribution are founded on varying folk theories about human causality. Western cultures, they note, emphasize the individual person and his or her attributes, whereas East Asian cultures focus on the back- ground or field that surrounds that person. To test this hypothesis, they showed American and Japanese college students underwater scenes featuring a cast of small fish, small animals, plants, rocks, and coral and one or more large, fast- moving focal fish, the stars of the show. Moments later, when asked to recount what they had seen, both groups recalled details about the focal fish to a nearly equal extent, but the Japanese reported far more details about the supporting cast in the background. Other researchers have also observed cultural differences in the extent to which people notice, track, think about, and remember the details of focal objects versus their contexts (Kitayama et al., 2003; Masuda & Nisbett, 2006; Savani & Markus, 2012).

Cultural differences in terms of what people focus on can be seen in natural settings outside the psychology laboratory. For example, Takahiko Masuda and others (2008) found that the art created in Western countries—whether the artists

were professionals or schoolchildren— tends to highlight individual people and objects, whereas art created in East Asia dedicates more space to backgrounds and scenery (see d Figure 4.8). Today, this difference in emphasis can be seen on the In- ternet. Comparative analyses show Western websites tend to present a small number of central details on a page, while East Asian websites con- tain more information overall, includ- ing more peripheral details (Wang et al., 2012). Similarly, while the focal face tends to dominate Facebook pro- file pictures posted in Western coun- tries, more background space appears in Facebook profile pictures in East Asian cultures (Huang & Park, 2013).

Moving from visual representa- tions into the world of sports, Hazel Rose Markus and others (2006) com- pared the way Olympic performances were described in the United States and Japan. These researchers ana- lyzed newspaper and TV coverage in

d F i g u r e 4 . 8

east Asian and Western Art

As demonstrated by these pieces of thirteenth-century art, East Asian art tends to have higher horizons and smaller people-to-scenery ratios (left). By contrast, Western art tends to occupy more space focusing on people (right).

Ic on

o f

Sa nt

a M

ar in

a an

d st

or ie

s of

h is

li fe

, 1 3t

h ce

nt ur

y By

za nt

in e

ar t.

Ch ur

of t

he H

ol y

Cr os

s, P

ed ho

ul as

, N ic

os ia

, C yp

ru s.

/ D

e A

go st

in i P

ic tu

re L

ib ra

ry /

A .

D ag

li O

rt i /

B rid

ge m

an Im

ag es

RM N

-G ra

nd P

al ai

s/ A

rt R

es ou

rc e,

N Y

80220_ch04_ptg01_hr_104-153.indd 130 30/01/16 10:53 AM

131 Attribution: From Elements to Dispositions

these countries and discovered that although everyone attributed victory and defeat to the athletes, American media were more likely to focus on an athlete’s unique personal attributes such as speed, strength, health, and determination): “I just stayed focused,” said Misty Hyman, American gold medalist swimmer. “It was time to show the world what I could do.” In contrast, Japanese media were also more likely and more fully to report on an athlete’s background and the role of others such as parents, coaches, and competitors. Woman’s mara- thon gold medalist Naoko Takahashi explained her own success this way: “Here is the best coach in the world, the best manager in the world, and all of the people who support me—all of these things were getting together and became a gold medal.”

Increasingly, the world is becoming a global village characterized by more and more racial and ethnic diversity within countries. Many people who migrate from one country to another become bicultural in their identity, retaining some ancestral manners of thought while adopting some of the lifestyles and values of their new homeland. How might these bicultural individuals make attributions for human behavior? Is it possible that they view people through one cultural frame or the other, depending on which one is brought to mind? It’s interesting that when shown a picture of one fish swimming ahead of a group, and asked why, Americans see the lone fish as leading the others (a personal attribution), while Chinese see the same fish as being chased by the others (a situational at- tribution). But what about bicultural social perceivers? In a study of China-born students attending college in California, researchers presented images symboliz- ing one of the two cultures (such as the U.S. and Chinese flags), administered the fish test, and found that compared to students exposed to the American images, those who saw the Chinese images made more situational attributions, seeing the lone fish as being chased rather than as leading. Apparently, it is possible for us to hold differing cultural worldviews at the same time and to perceive others through either lens, depending on which culture is brought to mind (Hong et al., 2000; Oyserman & Lee, 2008).

j Motivational Biases As objective as we try to be, our social perceptions are sometimes colored by personal hopes, needs, wishes, and preferences. This tendency shows it- self in the officiating controversies of the Olympics every four years, in other competitive sports, and in talent contests such as American Idol, The Voice, and The X Factor.

Wishful Seeing To illustrate wishful seeing in action, look at the object in d Figure 4.9. What do you see? In a series of studies, Emily Balcetis and David Dun- ning (2006) showed stimuli like this one to college students who thought that they were participating in a taste-testing experiment. The students were told that they would be randomly assigned to taste either freshly squeezed orange juice or a green- ish, foul-smelling “organic” drink—depending on whether a letter or a number was flashed on a laptop computer. For those told that a letter would assign them to the orange juice condition, 72% saw the letter B. For those told that a number would assign them to the orange juice, 61% saw the number 13. In another study, research participants in a laboratory were quicker to identify rapidly presented food words on a computer screen when they were hungry than when they had recently eaten (Radel & Clement-Guillotin, 2012). In some very basic ways, people have a tendency to see what they want to see (Dunning & Balcetis, 2013).

d F i g u r e 4 . 9

Motivated Visual Perception: How People See What They Want to See

Look at the image below. What do you see, the letter B or the number 13? The stimulus itself is ambiguous and can plausibly be seen either way. Research participants who thought they were in a taste-testing experiment were told that they would be assigned to taste orange juice or a foul-smelling green drink depending on whether a letter or a number was flashed on a laptop computer. For those told that a letter would yield orange juice, 72% saw the image as B. For those told that a number would yield orange juice, 61% saw a 13. This difference shows that sometimes people see what they want to see. From Balcetis, E., & Dunning, D. (2006). See what you want to see: Motivational influences on visual perception. Journal of Personality and Social Psychology, 91, 612–625. Copyright © 2006 by the American Psychological Association. Reprinted by permission.

80220_ch04_ptg01_hr_104-153.indd 131 30/01/16 10:53 AM

132 Chapter 4 Perceiving Persons

In a particularly ingenious program of research on “wishful seeing,” Balcetis and Dunning (2010) wondered if people would judge objects that they want to be physically closer than more neutral objects. In one study, college students who were thirsty (fed pretzels without water) compared to those who were quenched (drank as much water as they wanted) estimated that a bottle of water across a table was 3 inches closer to them. In a second study, students estimated their distance in a room from a $100 bill as 8 inches closer when they thought they could win it than when it had no potential value. In a third study, participants took part in a beanbag toss in which they tried to hit a target on the floor that was 13 feet away. When the target was said to be worth $25 compared to when it had $0 value, participants underthrew the beanbag by an average of 9 inches—suggesting that they perceived the target to be closer than it was.

Need for Self-esteem Motivations can bias our social perceptions in other ways as well. In Chapter 3, we saw that people have a strong need for self-esteem, a motive that can lead us to make favorable, self-serving, and one-sided attribu- tions for our own behavior. Research with students, teachers, parents, workers, athletes, and others shows that people tend to take more credit for success than they do blame for failure. This self-serving bias in attribution holds even when people outperform a competitor who is incompetent (Zell et al., 2015)! Similarly, other research shows that people seek more information about their strengths than about weaknesses, overestimate their contributions to group efforts, exaggerate their control, and predict an optimistic and rosy future. This positivity bias in attri- butions is ubiquitous. Based on a meta-analysis of 266 studies involving thousands of participants, Amy Mezulis and others (2004) found that except in some Asian cultures, “the self-serving bias is pervasive in the general population” (p. 711).

According to Dunning (2005), the need for self-esteem can bias social percep- tions in other subtle ways too, even when we don’t realize that the self is impli- cated. For example, do you consider yourself to be a “people person,” or are you more of a “task-oriented” type? And which of the two styles do you think makes for great leadership? It turns out that students who describe themselves as people- oriented see social skills as necessary for good leadership, whereas those who are more task-focused see a task orientation as better for leadership. Hence, people tend to judge favorably others who are similar to themselves rather than different on key characteristics (McElwee et al., 2001).

Sometimes ideological motives can color our attributions for the behavior of oth- ers. In the United States, it is common for political conservatives to blame poverty, crime, and other social problems on an “underclass” of people who are uneducated, immoral, lazy, or self-indulgent; in contrast, liberals often attribute these same prob- lems to social and economic institutions that favor powerful groups over others. Do conservatives and liberals think differently about the causes of human behavior, or do the attributions they make depend on whether the particular behavior they’re trying to explain fits with their ideology? Linda Skitka and others (2002) had college students who identified themselves as conservative or liberal make attributions for various events. They found that even though participants in general made personal attributions, as Westerners reflexively tend to do, they corrected for situational factors when ideologically motivated to do so. To explain why a prisoner was paroled, con- servatives were more likely to believe that the facility was overcrowded (a situational attribution) than that the prisoner had reformed (a personal attribution); to explain why a man lost his job, liberals were more likely to blame the company’s finances (a situational attribution) than the worker’s poor performance (a personal attribution).

Belief in a Just World At times, personal defensive motives lead us to blame others for their misfortunes. Consider the following classic experiment. Participants

80220_ch04_ptg01_hr_104-153.indd 132 30/01/16 10:53 AM

133 Attribution: From Elements to Dispositions

thought they were taking part in an emotion-perception study. One person, actually a confederate, was selected randomly to take a memory test while the others looked on. Each time the confederate made a mistake, she was jolted by a painful electric shock (actually, there was no shock; what participants saw was a staged videotape). Since participants knew that only the luck of the draw had kept them off the “hot seat,” you might think they would react with sympathy and compassion. Not so. In fact, they belittled the hapless confederate (Lerner & Simmons, 1966).

Melvin Lerner (1980) argues that the tendency to be critical of victims stems from our deep-seated belief in a just world. According to Lerner, people need to view the world as a just place in which we “get what we deserve” and “deserve what we get”—a world where hard work and clean living always pay off and where laziness and a sinful lifestyle are punished. To believe otherwise is to concede that we, too, are vulnerable to the cruel twists and turns of fate. A good deal of research supports this theory (Hafer & Bègue, 2005). Research also suggests that the belief in a just world can help victims cope and serves as a buffer against stress. But how might this belief system influence perceptions of others? If people cannot help or compensate the victims of misfortune, they turn on them. Thus, it is often assumed that poor people are lazy, that crime victims are careless, that battered wives pro- voke their abusive husbands, and that gay men and women with AIDS are promis- cuous. As you might expect, cross-nation comparisons reveal that people in poorer countries are less likely than those in more affluent countries to believe in a just world (Furnham, 2003). Similarly, as you might expect, political ideology can skew the belief in a just world. In a study of whether U.S. citizens believe in the ideal that America is a land of opportunity in which upward social mobility enables poor children to become successful adults, politically liberal respondents believe that there is less social mobility relative to conservative respondents (Chambers et al., 2015).

The tendency to disparage victims may seem like a symptom of the fundamental attribution error: too much focus on the person and not enough on the situation. But the conditions that trigger this tendency suggest there is more to it. Over the years, studies have shown that acci- dent victims are held more responsible for their fate when the damages from the accident are severe rather than mild (Walster, 1966), when the victim’s situation is similar to the perceiver’s (Shaver, 1970), when the perceiver is gener- ally anxious about threats to the self (Thornton, 1992), and when the perceiver identifies with the victim (Aguiar et al., 2008). Apparently, the more threatened we feel by an ap- parent injustice, the greater is the need to protect ourselves from the dreadful implication that it could happen to us—an implication that we defend by disparaging the victim. Ironically, recent research shows that people may also satisfy their belief in a just world by enhancing members of disadvan- taged groups—for example, by inferring that poor people are happy and that obese people are sociable, both attributes that restore justice by compensation (Kay & Jost, 2003; Kay et al., 2005).

In a laboratory experiment that reveals part of this process at work, participants watched a TV news story about a boy who was robbed and beaten. Some were told that the boy’s assailants were captured, tried, and sent to prison. Others were told that the assailants fled the country, never to be brought to trial—a story that strains one’s belief in a just world. Afterward, participants were asked to name as quickly as they could the colors in which various words in a list were typed (for

belief in a just world The belief that individuals get what they deserve in life, an orientation that leads people to disparage victims.

Attributions of blame are often biased by self-serving motivations.

Be rn

ar d

Sc ho

en ba

um /T

he N

ew Y

or ke

r C ol

le ct

io n/

Ca rt

oo n

Ba nk

.c om

“And see that you place the blame where it will do the most good.”

80220_ch04_ptg01_hr_104-153.indd 133 30/01/16 10:53 AM

134 Chapter 4 Perceiving Persons

example, the word chair may have been written in blue, floor in yellow, and wide in red). When the words themselves were neutral, all participants—regardless of which story they had seen—were equally fast at naming the colors. But when the words pertained to justice (words such as fair and unequal), those who had seen the justice-threatened version of the story were more distracted by the words and hence slower to name the colors. In fact, the more distracted they were, the more they derogated the victim. With their cherished belief in a just world threatened, these participants became highly sensitive to the concept of “justice” and quick to disparage the innocent victim (Hafer, 2000).

Integration: From Dispositions to Impressions When behavior is attributed to situational factors, we do not make strong infer- ences about the actor. However, personal attributions often lead us to infer that a person has a certain disposition—that the leader of a failing business is incom- petent, for example, or that the enemy who extends an olive branch wants peace. Human beings are not one-dimensional, however, and one trait does not consti- tute a full person. To have a complete picture of someone, social perceivers must assemble the various bits and pieces into a unified impression.

j information integration: The Arithmetic Once personal attributions are made, how are they combined into a single coher- ent picture of a person? How do we approach the process of impression formation? Do we simply add up all of a person’s traits and calculate a mental average, or do we combine the information in more complicated ways? Anyone who has written or received letters of recommendation will surely appreciate the practical implications. Suppose you’re told that an applicant is friendly and intelligent, two highly favor- able qualities. Would you be more or less impressed if you then learned that this applicant was also prudent and even-tempered, two moderately favorable qualities? If you are more impressed, then you are intuitively following a summation model of impression formation: The more positive traits there are, the better. If you are less impressed, then you are using an averaging model: The higher the average value of all the various traits, the better.

To quantify the formation of impressions, Norman Anderson (1968) had partici- pants rate the desirability of 555 traits on a 7-point scale. By calculating the average ratings, he obtained a scale value for each trait (sincere had the highest scale value; liar had the lowest). In an earlier study, Anderson (1965) used similar values and compared the summation and averaging models. Specifically, he asked a group of participants to rate how much they liked a person described by two traits with ex- tremely high scale values (H, H). A second group received a list of four traits that included two that were high and two that were moderately high in their scale values (H, H, M1, M1). In a third group, participants received two extremely low, negative traits (L, L). In a fourth group, they received four traits, including two that were low and two that were moderately low (L, L, M2, M2). What effect did the moderate traits have on impressions? As predicted by an averaging model, the moderate traits diluted from rather than added to the impact of the highly positive and negative traits. The practical advice for those who write letters of recommendation is clear: Applicants are better off if their letters include only the most glowing comments and omit favorable remarks that are somewhat more guarded in nature.

After extensive amounts of research, it appears that although people do tend to combine traits by averaging, the process is more complicated. Consistent with

impression formation The process of integrating information about a person to form a coherent impression.

80220_ch04_ptg01_hr_104-153.indd 134 30/01/16 10:53 AM

135 Integration: From Dispositions to Impressions

Anderson’s (1981) information integration theory, impressions formed of others are based on a combination, or integration, of (1) personal dispositions and the current state of the perceiver and (2) a weighted average, not a simple average, of the target person’s characteristics (Kashima & Kerekes, 1994). Let’s look more closely at these two sets of factors.

j Deviations From the Arithmetic Like other aspects of our social perceptions and attributions, impression forma- tion does not follow the rules of cold logic. Weighted averaging may describe the way most people combine different traits, but the whole process begins with a warm-blooded human perceiver, not a computer. Thus, certain deviations from the “arithmetic” are inevitable.

Perceiver Characteristics To begin with, each of us differs in terms of the kinds of impressions we form of others. Some people seem to measure others with an intellectual yardstick; others look for physical beauty, a warm smile, a good sense of humor, or a firm handshake. Whatever the attribute, each of us is more likely to notice certain traits than others (Bargh et al., 1988; Higgins et al., 1982). Thus, when people are asked to describe a group of target individuals, there’s typically more overlap between the various descriptions provided by the same perceiver than there is between those provided for the same target (Dornbusch et al., 1965; Park, 1986). Part of the reason for differences among perceivers is that we tend to use ourselves as a standard, or frame of reference, when evaluating others. Compared with the inert couch potato, for example, the high-energy athlete is more likely to see others as less active.

A perceiver’s current mood state can also influence the impressions formed of others (Forgas, 2000). In a classic experiment, Joseph Forgas and Gordon Bower (1987) told research participants that they had performed very well or poorly on a test of social adjustment. This positive feedback not only altered their moods but affected their view of other people. When presented with behavioral informa- tion about various characters, participants spent more time attending to positive facts and formed more favorable impressions when they were happy than when they were sad. Follow-up research has shown that people who are induced into a happy mood are also more optimistic, more lenient, and less critical in the attributions they make for others who succeed or fail (Forgas & Locke, 2005). In a positive mood, we are also more likely to interpret another person’s smile as genuine and heartfelt (Forgas & East, 2008a), more trusting and gullible in judging someone who is lying (Forgas & Rebekah, 2008), and more likely to form a quick impression of someone based on the first information we receive (Forgas, 2011).

embodiment effects Current mood is one aspect of our temporary state that can influence how we perceive other people. More and more, social psychologists are also finding that human thought is “embodied”—that the way we view ourselves and others is affected by the physical position, orientation, sensations, and move- ments of our bodies (Lakoff & Johnson, 1999; Niedenthal et al., 2005).

A number of studies illustrate embodiment effects in social perception. Con- sider the physical sensations of warmth and cold. The language we use tells us that people equate physical and social temperature—as when we describe someone as having a “warm personality” or “giving the cold shoulder.” In fact, research has shown that warm–cold is one of the most powerful dimensions in terms of the way we judge others. As we will see shortly, once we see a person as warm versus cold we assume that this person is also trustworthy, friendly,

information integration theory The theory that impressions are based on (1) perceiver dispositions and (2) a weighted average of a target person’s traits.

80220_ch04_ptg01_hr_104-153.indd 135 30/01/16 10:53 AM

136 Chapter 4 Perceiving Persons

caring, and helpful. Of course, warmth is also a physical sensation that because of early life experiences we tend to associate with shelter, safety, and nourish- ment. Would feeling warm, physically, influence our social perceptions? In one study, Lawrence Williams and John Bargh (2008) asked research participants to hold a cup of hot or iced coffee just before they read about and evaluated a fictitious target person. Compared to those who held the iced coffee, those who held the hot drink described the target as warmer, more generous, and more car- ing. In a second study, participants who were seated in a warm room, compared to those in a colder room, reported feeling interpersonally closer to the experi- menter (Ijzerman & Semin, 2009).

These results may seem far-fetched, but other studies have also demonstrated embodiment effects in social perception. In one study, participants made harsher moral judgments about the behavior of other people when they were given a bitter herbal mix rather than a sweet berry punch or water to drink during the experi- ment (Eskine et al., 2011). In a second study, participants on a computer judged people in photographs to be more trustworthy when they were induced to pull the palm of their hand up against the underside of the desk (a motion that we associ- ate with approach) than when they were induced to press that same palm down against the top of the desk (a motion we associate with avoidance) (Slepian et al., 2012). These embodiment effects may be rooted in the brain. Social neuroscience research shows that physical warmth and social closeness activate neural activity in the same region: The structure that regulates body temperature may also regu- late feelings of social warmth (Inagaki & Eisenberger, 2013).

Priming effects The combined effects of stable perceiver differences, fluctuating moods, and bodily sensations point to an important conclusion: To some extent, im- pression formation is in the eye of the beholder. The characteristics we tend to see in other people also change from time to time, depending on recent experiences. Have you ever noticed that once a seldom-used word slips into a conversation or appears on a blog, it is often repeated over and over again? If so, then you have observed priming, the tendency for frequently or recently used concepts to come to mind easily and influence the way we interpret new information.

The effect of priming on person impressions was first demonstrated by E. Tory Higgins and others (1977). Research participants were presented with a list of trait words, ostensibly as part of an experiment on memory. In fact, the task was designed as a priming device to plant certain ideas in their minds. Some participants read words that evoked a positive image: brave, independent, adven- turous. Others read words that evoked a more negative image: reckless, foolish, careless. Later, in what they thought to be an unrelated experiment, participants read about a man named Donald who climbed mountains, drove in a demolition derby, and tried to cross the Atlantic Ocean in a sailboat. As predicted, their im- pressions of Donald were shaped by the trait words they had earlier memorized. Those exposed to positive words later formed more favorable impressions of him than those exposed to negative words. All participants read exactly the same description, yet they formed different impressions depending on what concept was already on their minds to be used as a basis for comparison (Mussweiler & Damisch, 2008). In fact, priming seems to work best when the prime words are presented so rapidly that people are not even aware of the exposure (Bargh & Pietromonaco, 1982).

Our motivations and even our social behaviors are also subject to the automatic effects of priming without awareness. In one provocative study, John Bargh and Tanya Chartrand (1999) gave participants a “word search” puzzle that contained

priming The tendency for recently used or perceived words or ideas to come to mind easily and influence the interpretation of new information.

80220_ch04_ptg01_hr_104-153.indd 136 30/01/16 10:53 AM

137 Integration: From Dispositions to Impressions

either neutral words or words associated with achievement motivation (strive, win, master, compete, succeed). Afterward, the participants were left alone and given 3 minutes to write down as many words as they could form from a set of Scrab- ble letter tiles. When the 3-minute limit was up, they were signaled over an intercom to stop. Did these participants, who were driven to obtain a high score, stop on cue or continue to write? Through the use of hidden cameras, the experimenters observed that 57% of those primed with achievement-related words continued to write after the stop signal, compared to only 22% in the control group.

Looking at priming effects on social behavior, Bargh, Chen, and Burrows (1996) gave people 30 sets of words pre- sented in scrambled order (“he it hides finds instantly”) and told them to use some of the words in each set to form gram- matical sentences. After explaining the test, which would take 5 minutes, the experimenter told participants to locate him down the hall when they were finished so he could ad- minister a second task. So far so good. But when participants found the experimenter, he was in the hallway immersed in conversation, and he stayed in that conversation for 10 full minutes without acknowledging their presence. What’s a per- son to do, wait patiently or interrupt? The participants didn’t know it, but some had worked on a scrambled word test that contained many “politeness” words (yield, respect, consider- ate, courteous), whereas others had been exposed to words related to rudeness (disturb, intrude, bold, bluntly). Would these test words secretly prime participants, a few minutes later, to behave in one way or the other? Yes. Compared with those given the neutral words to unscramble, participants primed for rudeness were more likely—and those primed for politeness were less likely—to break in and interrupt the experimenter (see d Figure 4.10).

What accounts for this effect of priming, not only on our social perceptions but also on our behavior? The link be- tween perception and behavior is automatic; it happens like a mindless reflex. Present scrambled words that prime the “elderly” stereotype (old, bingo) and research participants walk out of the experiment more slowly as if mimicking an elderly person (Dijksterhuis & Bargh, 2001). But why? Joseph Cesario and others (2006) suggest that the automatic priming of behavior is an adaptive social mechanism that helps us to prepare for upcoming encounters with a primed target if we are so motivated. After measuring participants’ attitudes toward the elderly, these researchers predicted and found that those who liked old people walked more slowly after priming (as if synchronizing with a slow friend), while those who disliked old people walked more quickly (as if fleeing from such an interaction).

Target Characteristics Just as not all social perceivers are created equal, nei- ther are all traits created equal. In recent years, personality researchers have dis- covered, across cultures, that individuals can reliably be distinguished from one another along five broad traits, or factors: extroversion, emotional stability, open- ness to experience, agreeableness, and conscientiousness (De Raad, 2000; McCrae

d F i g u r e 4 . 1 0

The Priming of Social Behavior Without Awareness

Would waiting participants interrupt the busy experimenter? Compared with those who had previously been given neutral words to unscramble (center), participants given politeness words were less likely to cut in (left), and those given rudeness words were more likely to cut in (right). These results show that priming can influence not only our social judgments but our behavior as well. Based on Bargh, J. A., Chen, M., & Burrows, L. (1996). Automaticity of social behavior: Direct effects of trait construct and stereotype activation on action. Journal of Personality and Social Psychology, 71, 230–244. Copyright © 1996 by the American Psychological Association.

Pe rc

en ta

ge w

ho in

te rr

up te

Polite Neutral

Priming Condition

Rude

80220_ch04_ptg01_hr_104-153.indd 137 30/01/16 10:53 AM

138 Chapter 4 Perceiving Persons

& Costa, 2003; Wiggins, 1996). Some of these factors are easier to judge than others. Based on a review of 32 studies, David Kenny and others (1994) found that social perceivers are most likely to agree in their judgments of a target’s ex- troversion: that is, the extent to which he or she is sociable, friendly, fun-loving, outgoing, and adventurous. It seems that this characteristic is easy to spot, and different perceivers often agree on it even when rating a target person whom they are seeing for the first time.

The valence of a trait—whether it is considered good or bad—also influ- ences its impact on our final impressions. Over the years, research has shown that people exhibit a trait negativity bias, the tendency for negative information to weigh more heavily on our impressions than positive information (Rozin & Royzman, 2001; Skowronski & Carlston, 1989). This means that we form more extreme impressions of a person who is said to be dishonest than of one who is said to be honest. When you think about it, this makes sense. We tend to view others favorably, so we are quick to take notice and pay careful attention when this expectation is violated (Pratto & John, 1991). For that reason, one bad trait may be enough to tarnish a person’s reputation, regardless of other qualities. In light of this research, Baumeister and others (2001) have concluded that bad is stronger than good in a “disappointingly relentless pattern” (p. 362).

Our positive expectations of others are so strong that the absence of a favor- able evaluation may lead us to assume the worst. If you’ve ever read or written letters of recommendation, you will understand this next study. In evaluating people—whether it’s for school or a job or membership in a social group—it is customary to comment on two characteristics of that person—warmth and com- petence. What if you’re reading a letter that raves about how competent an ap- plicant is but says nothing about how nice he or she is as a person? Or what if a letter comments on how nice the applicant is but says nothing about competence? No negative information is stated, but is it implied by omission? Yes. In a series of studies, Nicolas Kervyn and others (2012) found that when research partici- pants read brief but praiseworthy evaluations of a person that focused only on competence (“seems like a very smart, hardworking, and competent person”) or only on warmth (“seems like a very nice, sociable, and outgoing person”), they drew negative inferences about the dimension that was omitted. Illustrating the expression, “Damned by faint praise,” Kervyn et al. referred to this result as the innuendo effect.

When you think about it, it’s probably adaptive for us to stay alert for nega- tive, potentially threatening information. Recent research suggests that people are quicker to sense their exposure to subliminally presented negative words such as bomb, thief, shark, and cancer than to positive words such as baby, sweet, friend, and beach (Dijksterhuis & Aarts, 2003). This sensitivity to negative information is found in infants less than a year old (Vaish et al., 2008). It can also be “seen” in the brain (N. K. Smith et al., 2003). In one study, Tiffany Ito and oth- ers (1998) exposed research participants to slides that depicted images that were positive (a red Ferrari, people enjoying a roller coaster), negative (a mutilated face, a handgun pointed at the camera), or neutral (a plate, a hair dryer). Using electrodes attached to participants’ scalps, these researchers recorded electrical activity in different areas of the brain during the presentation. Sure enough, certain types of activity were more pronounced when participants saw negative images than when they saw stimuli that were positive or neutral. It appears, as these researchers commented, that “negative information weighs more heavily on the brain” (p. 887).

80220_ch04_ptg01_hr_104-153.indd 138 30/01/16 10:53 AM

139 Integration: From Dispositions to Impressions

The impact of trait information on our impressions of other people depends not only on characteristics of the perceiver and target but on context as well. Think, for example, about a time when you met a person for the first time who was in the company of someone you like or dislike, or find attractive or unattract- ive. Would your impression of the target person be colored by this relationship? In studies of “stigma by association,” John Pryor and his colleagues (2012) found that the answer is yes. Even when a target person’s relationship to the positive or negative other comes about by chance, not as a matter of choice, our first impres- sions are influenced by it. Two other contextual factors are also particularly im- portant: (1) implicit theories of personality and (2) the order in which we receive information about one trait relative to other traits.

implicit Personality Theories Whether we realize it or not, each of us harbors an implicit personality theory—a network of assumptions about the relationships among various types of people, traits, and behaviors. Knowing that someone has one trait leads us to infer that he or she has other traits as well (Bruner & Tagiuri, 1954; Schneider, 1973; Sedikides & Anderson, 1994). For example, you might assume that a person who is unpredictable may also be dangerous or that someone who speaks slowly is also slow-witted. You might also assume that certain traits and behaviors are linked together (Reeder, 1993; Reeder & Brewer, 1979)—that a celebrity with a sweet and beloved persona, for example, could not possibly have skeletons in the closet.

Solomon Asch (1946) was the first to discover that the presence of one trait often implies the presence of other traits. Asch told one group of research partici- pants that an individual was “intelligent, skillful, industrious, warm, determined, practical and cautious.” Another group read an identical list of traits, except that the word warm was replaced by cold. Only the one term was changed, but the two groups formed very different impressions. Participants inferred that the warm person was also happier and more generous, good-natured, and humorous than the cold person. Yet when two other words were varied (polite and blunt), the differences were less pronounced. Why? Asch concluded that warm and cold are central traits, meaning that they imply the presence of certain other traits and exert a powerful influence on final impressions. In fact, when college students in different classes were told ahead of time that a guest lecturer was a warm or cold person, their impressions after the lecture were consistent with these beliefs, even though he gave the same lecture to everyone (Kelley 1950; Widmeyer & Loy 1988).

Is there something magical about the traits warm and cold? To learn more about the structure of implicit personality theories, Seymour Rosenberg and his colleagues (1968) handed research participants 60 cards, each with a trait word written on it, and asked them to sort the cards into piles that represented specific people, perhaps friends, co-workers, acquaintances, or celebrities. The traits were then statistically correlated to determine how often they appeared together in the same pile. The results were plotted to display the psychological distance between the various characteristics. The “map” shown in d Figure 4.11 shows that the traits—positive and negative alike—were best captured by two dimensions: social and intellectual. Based on their review of many years of research, Susan Fiske, Amy Cuddy, and Peter Glick (2007) conclude that people differentiate each other first in terms of their warmth (warm is seen in such traits as friendly, helpful, and sincere), and second in terms of their competence (competent is seen in such traits as smart, skillful, and determined)—and that these are “universal dimensions of social cognition.”

central traits Traits that exert a powerful influence on overall impressions.

80220_ch04_ptg01_hr_104-153.indd 139 30/01/16 10:53 AM

140 Chapter 4 Perceiving Persons

j Perceptions of Moral Character In addition to focusing on warmth and competence, social perceivers all over the world also form strong impressions of how moral people are. When you think about it, accurate perceptions of moral character are essential for social living so that we can identify whether someone we meet, and may need to rely on in the future, is moral and good—and therefore can be trusted.

Geoffrey Goodwin (2015) argues that morality is not just a third facet of social perception, but rather that it is the most important factor in the impressions we form of others. Noting that people sometimes conflate warm and moral, Goodwin and others (2014) conducted a series of studies in which they found that distinctly “moral” traits (such as courageous, fair, principled, just, honest, trustworthy, and loyal) proved more important than distinctly “warm” traits (such as warm, socia- ble, happy, agreeable, enthusiastic, easygoing, fun, and playful) at predicting the positive and negative impressions that people form of others. This result was the

d F i g u r e 4 . 1 1

universal Dimensions of Social Cognition

Rosenberg and others (1968) asked people to sort 60 cards, each with a trait word on it, into piles that depicted specific individuals. Through a statistical procedure used to plot how frequently the various traits appeared together, an implicit personality theory “map” emerged. This map shows that both positive and negative traits can be ordered along two dimensions: social (warmth) and intellectual (competent). Since this study, other research has confirmed that warmth and competence are universal dimensions by which people perceive each other. From Rosenberg, S., Nelson, C., & Vivekananthan, P. S. (1968). A multidimensional approach to the struc- ture of personality impressions. Journal of Personality and Social Psychology, 9, 283–294. Copyright © 1968 by the American Psychological Association. Reprinted by permission.

Bad-Social

Good-Intellectual

Bad-Intellectual

Good-Social

Scientific Persistent

Skillful

Determined Industrious

IntelligentImaginative

Serious Important

Discriminating

Cautious Practical

Artistic

ReliableReserved Daring

Dominating

Moody IrritablePessimistic

Unpopular

Humorless Unsociable

Cold

Unhappy Vain

Finicky

Honest

Modest Tolerant

Helpful Sincere

Sentimental

Humorous Happy

Good Natured Warm

Sociable Popular

Dishonest

Boring Unimaginative

Impulsive Submissive

Naive Clumsy

Frivolous Unintelligent Foolish

Wavering

Irresponsible Wasteful

Unreliable Superficial Insignificant

Squeamish

Critical Stern

Shrewd

Meditative

80220_ch04_ptg01_hr_104-153.indd 140 30/01/16 10:53 AM

141 Integration: From Dispositions to Impressions

same regardless of whether participants judged real people from their own lives, U.S. presidents, fictitious individuals described in the laboratory, or deceased men and women whose New York Times obituaries they were given to read.

There is a growing consensus among researchers that the perception of moral- ity plays a special role in the impressions we form of others (Brambilla & Leach, 2014). In fact, research shows that people make moral judgments about others instantly and intuitively— without intention, thought, or awareness (Haidt, 2001; Van Berkum et al., 2009). If perceptions of moral character are so basic and so adaptive, it would stand to reason that different social perceivers, despite their otherwise unique perspectives, would agree on what constitutes moral character. Does research support this hypothesis? Yes. To measure agreement rates among different social perceivers, Erik Helzer and others (2014) developed a “Moral Char- acter Questionnaire” in which people were asked to rate the extent to which vari- ous brief statements accurately described the self, a friend, an acquaintance, and a family member. Each of the statements described a facet of moral character—such as fairness (“treats everyone in a similar way”), honesty (“tells the truth), compas- sion (“takes time out to help others”), self-control (“is good at resisting tempta- tion”), moral concern (“thinks about how to be a good person”), and general morality (“is an ethical person”). The results showed that there were high levels of agreement both in participants’ self-ratings, their ratings of others, and others’ ratings of the participants themselves.

The Primacy effect The order in which a trait is discovered can also influence its impact. It is often said that first impressions are critical, and social psychologists are quick to agree. Studies show that information often has greater impact when presented early in a sequence rather than late, a common phenomenon known as the primacy effect.

In another of Asch’s (1946) classic experiments, one group of participants learned that a person was “intelligent, industrious, impulsive, critical, stubborn, and envious.” A second group received exactly the same list but in reverse order. Rationally speaking, the two groups should have felt the same way about the per- son. But instead, participants who heard the first list in which the more positive traits came first formed more favorable impressions than did those who heard the second list. Similar findings were obtained among participants who watched a videotape of a woman taking an SAT-like test. In all cases, she correctly answered 15 out of 30 multiple-choice questions. But participants who observed a pattern of initial success followed by failure perceived the woman as more intelligent than did those who observed the opposite pattern of failure followed by success (Jones et al., 1968). There are exceptions, but as a general rule, people tend to be more heavily influenced by the “early returns.”

What accounts for this primacy effect? There are two basic explanations. The first is that once perceivers think they have formed an accurate impression of someone, they tend to pay less attention to subsequent information. Thus, when research participants read a series of statements about a person, the amount of time they spent reading each of the items declined steadily with each succeeding statement (Belmore, 1987).

Does this mean that we are doomed to a life of primacy? Not at all. If you are mentally tired or unstimulated, your attention may wane. But if you are suf- ficiently motivated to avoid tuning out and are not pressured to form a quick first impression, then primacy effects are diminished (Anderson & Hubert, 1963; Kruglanski & Freund, 1983). In one study, college students “leaped to conclu- sions” about a target person on the basis of preliminary information when they were mentally fatigued from having just taken a two-hour exam but not when they

primacy effect The tendency for information presented early in a sequence to have more impact on impressions than information presented later.

80220_ch04_ptg01_hr_104-153.indd 141 30/01/16 10:53 AM

142 Chapter 4 Perceiving Persons

were fresh, alert, and motivated to pay attention (Webster et al., 1996). In addi- tion, Arie Kruglanski and Donna Webster (1996) found that some people are more likely than others to “seize” upon and “freeze” their first impressions. Apparently, individuals differ in their need for closure, the desire to reduce ambiguity. People who are low in this regard are open-minded, deliberate, and perhaps even reluc- tant to draw firm conclusions about others. In contrast, those who are high in the need for closure tend to be impulsive and impatient and to form quick and lasting judgments of others.

More unsettling is the second reason for primacy, known as the change-of-mean- ing hypothesis. Once people have formed an impression, they start to interpret in- consistent information in light of that impression. Asch’s research shows just how malleable the meaning of a trait can be. When people are told that a kind person is calm, they assume that he or she is gentle, peaceful, and serene. When a cruel per- son is said to be calm, however, the same word is interpreted to mean cool, shrewd, and calculating. There are many examples to illustrate the point. Based on your first impression, the word proud can mean self-respecting or conceited, critical can mean astute or picky, and impulsive can mean spontaneous or reckless.

It is remarkable just how creative we are in our efforts to transform a bundle of contradictions into a coherent, integrated impression. For example, the person who is said to be “good” but also “a thief” can be viewed as a Robin Hood–like character (Burnstein & Schul, 1982). Asch and Henri Zukier (1984) presented people with inconsistent trait pairs and found that they used different strategies to reconcile the conflicts. For example, a brilliant-foolish person may be seen as “very bright on abstract matters, but silly about day-to-day practical tasks,” a sociable-lonely person has “many superficial ties but is unable to form deep rela- tions,” and a cheerful-gloomy person may simply be someone who is “moody.”

Confirmation Biases: From Impressions to Reality It is a striking feature of human nature but too often true: Once people make up their minds about something—even if they don’t have all the necessary informa- tion—they become more and more unlikely to change their minds when confronted with new evidence. People can be quite intellectually stubborn. Political leaders often refuse to withdraw support for government programs that don’t work, and scientists often steadfastly defend their pet theories in the face of contradictory research data. These instances are easy to explain. Politicians and scientists have personal stakes in their opinions, as votes, pride, funding, and reputation may be at risk. But what about people who more innocently fail to revise their opinions, often to their own detriment? What about the baseball manager who clings to old strategies that don’t work or the trial lawyer who consistently selects juries ac- cording to false stereotypes? Why are they often slow to face the facts? As we will see, people are subject to various confirmation biases—tendencies to interpret, seek, and create information in ways that verify existing beliefs.

j Perseverance of Beliefs Imagine you are looking at a slide that is completely out of focus. Gradually, it becomes focused enough so that the image is less blurry. At this point, the experi- menter wants to know if you can recognize the picture. The response you’re likely to make is interesting. Participants in experiments of this type have more trouble

need for closure The desire to reduce cognitive uncertainty, which heightens the importance of first impressions.

confirmation bias The tendency to seek, interpret, and create information that verifies existing beliefs.

80220_ch04_ptg01_hr_104-153.indd 142 30/01/16 10:53 AM

143 Confirmation Biases: From Impressions to Reality

making an identification if they watch the gradual focusing procedure than if they simply view the final, blurry image. In the mechanics of the perceptual process, people apparently form early impressions that in- terfere with their subsequent ability to “see straight” once presented with improved evi- dence (Bruner & Potter, 1964). As we will see in this section, social perception is sub- ject to the same kind of interference, which is another reason why first impressions of- ten stick like glue even after we are forced to confront information that discredits them.

Consider what happens when you’re led to expect something that does not mate- rialize. In one study, John Darley and Paget Gross (1983) asked participants to evaluate the academic potential of a 9-year-old girl named Hannah. One group was led to be- lieve that Hannah came from an affluent community in which both parents were well-educated professionals (high expectations). A second group thought that she was from a run-down urban neighborhood and that both parents were uneducated blue-collar workers (low expectations). As shown in d Figure 4.12, participants in the first group were slightly more optimistic in their ratings of Hannah’s potential than were those in the second group. In each of these groups, however, half the partici- pants then watched a videotape of Hannah taking an achievement test. Her perfor- mance on the tape seemed average. She correctly answered some difficult questions but missed others that were relatively easy. Look again at Figure 4.12 and you’ll see that even though all participants saw the same tape, Hannah now received much lower ratings of ability from those who thought she was poor and higher ratings from those who thought she was affluent. Apparently, presenting an identical body of mixed evidence did not extinguish the biasing effects of beliefs; it fueled these effects.

Events that are ambiguous enough to support contrasting interpretations are like inkblots: We see or hear in them what we expect to see or hear. Illustrating the point, one group of researchers had people rate from photographs the extent to which pairs of adults and children resembled each other. Interestingly, the par- ticipants did not see more resemblance in parents and offspring than in random pairs of adults and children. Yet when told that certain pairs were related, they did “see” a resemblance, even when the relatedness information was false (Bressan & Martello, 2002). In a second study, participants listened to noisy, more or less “de- graded” tape recordings of two people talking—not unlike what you hear in 9-1-1 dispatch recordings. Some participants were told that the interviewee was a crime suspect; others were told that he was a job applicant. When the speech recordings were degraded, participants were more likely to “hear” incriminating statements when they thought that the interviewee was a suspect than an applicant (Lange et al., 2011). In a third study, participants acting as mock jurors were asked to determine if the handwriting in a bank robbery note was similar to a handwriting sample taken from a suspect who had confessed or denied involvement in the crime. Consistent with expectations, participants who believed the suspect had confessed were more likely to see the two handwritings as a match—even though they were not (Kukucka & Kassin, 2014; for a discussion of the implications of “forensic confirmation biases,” see Kassin, Dror, & Kukucka, 2013).

“Please your majesty,” said the knave, “I didn’t write it and they can’t prove I did; there’s no name signed at the end.” “If you didn’t sign it,” said the King, “that only makes the matter worse. You must have meant some mischief, or else you’d have signed your name like an honest man.”

This exchange, taken from Lewis Carroll’s Alice’s Adventures in Wonderland, illustrates the power of existing impressions.

A F

ar ch

iv e/

A la

m y

80220_ch04_ptg01_hr_104-153.indd 143 30/01/16 10:53 AM

144 Chapter 4 Perceiving Persons

What about information that plainly disconfirms our beliefs? What then happens to our first im- pressions? Craig Anderson and his colleagues (1980) addressed this question by supplying participants with false information. After they had time to think about it, they were told that it was untrue. In one experiment, half the partici- pants read case studies suggesting that people who take risks make better firefighters than do those who are cautious. The others read cases suggesting the opposite con- clusion. Next, participants were asked to come up with a theory for the suggested correlation. The possibilities are easy to imagine: “He who hesitates is lost” sup- ports risk-taking, whereas “You have to look before you leap” sup- ports caution. Finally, participants were led to believe that the session was over and were told that the information they had received was false, manufactured for the sake of the experiment. Participants, how- ever, did not abandon their theo-

ries about firefighters. Instead, they exhibited belief perseverance, a tendency to retain to one’s initial beliefs even after they had been discredited.

In a brilliant illustration, Tobias Greitemeyer (2014) wondered what happens when a scientific journal retracts an article it has already published because of proof that data were fabricated and could not be trusted. To answer this question, he pre- sented college students with a summary of a research paper “indicating” that people who experience an elevated physical height (as in riding an escalator up) are more likely to offer help to a stranger than those who experience a lowered physical height (as in riding an escalator down). Students who read this summary, compared to those who did not, were later more likely to believe the result when asked for their beliefs about the link between physical height and helping. In a third group, how- ever, the students read after the summary that the data supporting this hypothesis had been fabricated. Did this disclosure and retraction fully erase the effect? No. Stu- dents who read the summary and retraction persevered in their belief in the result.

Why do beliefs often outlive the evidence on which they are supposed to be based? The reason is that once people form a belief, they conjure up explanations that make sense—and those explanations help to perpetuate the belief even after it has been discredited! This is what happened to the students in Greitemeyer’s (2014) study. In fact, research shows that once people form an opinion, that opin- ion becomes strengthened just by thinking about the topic. And therein lies a pos- sible solution. By asking people to consider why an alternative theory might be true, we can reduce or eliminate the belief perseverance effects to which they are vulnerable (Anderson & Sechler, 1986).

belief perseverance The tendency to maintain beliefs even after they have been discredited.

People are slow to change their first impressions on the basis of new information.

True

d F i g u r e 4 . 1 2

Mixed evidence: Does it extinguish or Fuel First impressions?

Participants evaluated the potential of a schoolgirl. Without seeing her test performance, those with high expectations rated her slightly higher than did those with low expectations. Among the participants who watched a tape of the girl taking a test, the expectations effect was even greater. Based on Darley, J. M., & Gross, P. H. (1983). A hypothesis-confirming bias in labeling effects. Journal of Personality and Social Psychology, 44, 20–33. Copyright © 1983 by the American Psychological Association.

Performance not viewed

Performance viewed

pl ac

em en

t o f H

an na

h Pa

rt ic

ip an

t's g

ra de

le ve

Low expectations

Performance not viewed

Performance viewed

MathematicsReading

High expectations

80220_ch04_ptg01_hr_104-153.indd 144 30/01/16 10:53 AM

145 Confirmation Biases: From Impressions to Reality

j Confirmatory Hypothesis Testing Social perceivers are not passive recipients of information. Like detectives, we ask questions and actively search for clues. But do we seek information objectively or are we inclined to confirm the suspicions we already hold?

Mark Snyder and William Swann (1978) addressed this question by having pairs of participants who were strangers to one another take part in a getting- acquainted interview. In each pair, one participant was supposed to inter- view the other. But first, that participant was falsely led to believe that his or her partner was either introverted or extroverted (actually, participants were assigned to these conditions on a random basis) and was then told to select questions from a prepared list. Those who thought they were talking to an in- trovert chose mostly introvert-oriented questions (“Have you ever felt left out of some social group?”), whereas those who thought they were talking to an extrovert asked extrovert-oriented questions (“How do you liven up a party?”). Expecting a certain kind of person, participants unwittingly sought evidence that would confirm their expectations. By asking loaded questions, in fact, the interviewers actually gathered support for their beliefs. Thus, neutral observers who later listened to the tapes were also left with the mistaken impression that the interviewees really were as introverted or extroverted as the interviewers had assumed.

This last part of the study is powerful but in hindsight not all that surprising. Imagine yourself on the receiving end of an interview. Asked about what you do to liven up parties, you would probably talk about organizing group games, play- ing dance music, and telling jokes. On the other hand, if you were asked about difficult social situations, you might talk about being nervous before oral presen- tations or about what it feels like to be the new kid on the block. In other words, simply by going along with the questions that are asked, you supply evidence confirming the interviewer’s beliefs. Thus, perceivers set in motion a vicious cycle: Thinking someone has a certain trait, they engage in a one-sided search for information. In doing so, they create a reality that ultimately supports their beliefs (Zuckerman et al., 1995).

Sadly, the fact that people can be blinded by their existing beliefs is a pervasive phenomenon with consequences. In one study, research participants reviewed a mock police file on a crime investigation into a home invasion and shooting case. The file contained weak circumstantial evidence pointing to a possible suspect. At that point, some participants but not others were asked to form and state an initial hypothesis as to the likely offender. As predicted, those who did went on to search for additional evidence and then interpret that evidence in ways that confirmed their hypothesis. In these ways, a weak suspect became the prime sus- pect (O’Brien, 2009). In a second study, researchers presented psychiatrists and medical students with an experimental patient file and found that 13% of psy- chiatrists and 25% of students who had made a preliminary diagnosis exhibited confirmation biases in searching for new information—and those who did were less likely than others to make the correct diagnosis and prescribed treatment (Mendel et al., 2011).

Let’s stop for a moment and contemplate what all this research means for the broader question of why we often seem to resist changing our negative but mistaken impressions of others more than our positive but mistaken impressions. Jerker Denrell (2005) argued that even when we form a negative first impression of someone on the basis of all available evidence and even when we interpret that evidence accurately, our impression may be misleading. The reason is biased

“It is a capital mistake to theorize before you have all the evidence. It biases the judgment.”

—Arthur Conan Doyle

80220_ch04_ptg01_hr_104-153.indd 145 30/01/16 10:53 AM

146 Chapter 4 Perceiving Persons

experience sampling. Meet someone who seems likable and you may interact with that person again. Then if he or she turns out to be twisted, dishonest, or self-centered, you’ll be in a position to observe these traits and revise your im- pression. But if you meet someone you don’t like, you will try to avoid that person in the future, cutting yourself off from new information and limiting the opportunity to revise your opinion. Attraction breeds interaction, which is why our negative first impressions in particular tend to persist.

self-fulfilling prophecy The process by which one’s expectations about a person eventually lead that person to behave in ways that confirm those expectations.

Pa ul

N ot

h/ N

ew Y

or ke

r C ar

to on

/T he

C ar

to on

B an

j The Self-Fulfilling Prophecy In 1948, sociologist Robert Merton told a story about Cart- wright Millingville, president of the Last National Bank during the Great Depression of the 1930s. The bank was solvent, yet a rumor began to spread that it was flounder- ing. Within hours, hundreds of depositors were lined up to withdraw their savings before no money was left to with- draw. The rumor was false, but the bank eventually failed. Using stories such as this, Merton proposed what seemed like an outrageous hypothesis: that a perceiver’s expecta- tion can actually lead to its own fulfillment, a self-fulfilling prophecy.

Merton’s hypothesis lay dormant within psychology un- til Robert Rosenthal and Lenore Jacobson (1968) published the results of a study in a book entitled Pygmalion in the Classroom. Noticing that teachers had higher expectations for better students, they wondered if teacher expectations influenced student performance rather than the other way around. To address the question, they told teachers in a San Francisco elementary school that certain pupils were on the verge of an intellectual growth spurt. The results of an IQ test were cited, but in fact, the pupils had been randomly selected. Eight months later, when real tests were administered, the “late bloomers” exhibited an increase in their IQ scores compared with children assigned to a control group. They were also evalu- ated more favorably by their classroom teachers.

When the Pygmalion study was first published, it was greeted with chagrin. If positive teacher expectations boost student performance, can negative expecta- tions have the opposite effect? What about the social implications? Could it be that affluent children are destined for success and disadvantaged children are doomed to failure because educators hold different expectations for them? Many researchers were critical of the study and skeptical about the results. Unfortu- nately, these findings could be swept under the rug. In a review of additional studies, Rosenthal (1985) found that teachers’ expectations significantly predicted students’ performance 36% of the time. Fortunately, teachers are less and less able to make these accurate predictions as children graduate from one grade to the next (A. Smith et al., 1999).

How might teacher expectations be transformed into reality? There are two points of view. According to Rosenthal (2002), the process involves covert com- munication. The teacher forms an initial impression of students early in the school year based, perhaps, on their background or reputation, physical appearance, ini- tial classroom performance, and standardized test scores. The teacher then alters his or her behavior in ways that are consistent with that impression. If initial

80220_ch04_ptg01_hr_104-153.indd 146 30/01/16 10:53 AM

147 Confirmation Biases: From Impressions to Reality

expectations are high rather than low, the teacher gives the student more praise, more attention, more challenging homework, and better feedback. In turn, the student adjusts his or her own behavior. If the signals are positive, the student may become energized, work hard, and succeed. If negative, there may be a loss of interest and self-confidence. The cycle is thus complete, and the expectations are confirmed.

While recognizing that this effect can occur, Lee Jussim (2012) has ques- tioned whether teachers in real life are prone in the first place to form errone- ous impressions of their students. It’s true that in many naturalistic studies, in real classrooms, the expectations teachers have at the start of a school year are ultimately confirmed by their students—a result that is consistent with the notion that the teachers had a hand in producing that outcome. But that same result is also consistent with a more innocent possibility—that perhaps the expectations that teachers form of their students are accurate. In other words, sometimes teachers can accurately predict how their students will perform in the future without necessarily influencing that performance (Alvidrez & Weinstein, 1999).

Jussim admits that there are times when teachers stereotype a student and without realizing it behave in ways that create a self-fulfilling prophecy. But his review of the research literature suggests that evidence for bias is not nearly as strong as it appears on the surface. Addressing this question in a longitudinal study of mothers and their children, Stephanie Madon and others (2003) found that underage adolescents are more likely to consume alcohol when their moth- ers had previously expected them to. Statistical analyses revealed that this proph- ecy was fulfilled in part because the mothers influence their sons and daughters, as Rosenthal’s work would suggest, but also in part because the mothers are able to predict their own children’s behavior, as Jussim’s model would suggest. In fact, a follow-up study suggests that the link between a mother’s expectations and her adolescent’s later alcohol consumption did not strengthen or weaken over time, remaining stable as the child moved from the seventh grade through the twelfth (Madon et al., 2006).

It is clear that self-fulfilling prophecies are at work, to a greater or lesser degree, not only in schools but also in a wide range of organizations, including the military (Kierein & Gold, 2000; McNatt, 2000). In a study of 1,000 men as- signed to 29 platoons in the Israel Defense Forces, Dov Eden (1990) led some platoon leaders but not others to expect that groups of trainees they were about to receive had great potential (in fact, these groups were of average ability). After 10 weeks, the trainees assigned to the high-expectation platoons scored higher than the others on written exams and on the ability to operate a weapon. The process may also be found in the criminal justice system when police interrogate suspects. To illustrate, Kassin and others (2003) had some college students but not others commit a mock crime, stealing $100 from a laboratory. All suspects were then questioned by interrogators who were led to believe that their suspect was probably guilty or innocent. Interrogators who presumed guilt asked more incriminating questions, were more coercive, and tried harder to get the suspect to confess. In turn, this more aggressive style made the sus- pects sound defensive and led observers who later listened to the tapes to judge them guilty, even when they were innocent.

On a more personal level, the self-fulfilling prophecy is a process that can have particularly sad and ironic effects on people who are insecure about their social relationships. Studying what they have called the rejection prophecy, Danu

80220_ch04_ptg01_hr_104-153.indd 147 30/01/16 10:53 AM

148 Chapter 4 Perceiving Persons

Stinson, Jessica Cameron, and others have found that (1) people who are insecure are fearful of rejection, which makes them tense and awkward in the social situa- tions; (2) their resulting behavior is off-putting to others, which (3) increases the likelihood of rejection and reinforces their initial insecurity (Stinson et al., 2009, Cameron et al., 2010). Fortunately, there is a way to break this vicious cycle. In a three-phased study that spanned 8 weeks, Stinson and others (2011) found that participants who were provided with an early opportunity to affirm the self—by writing about values that were important to them—became more secure over time and more relaxed in their social interactions.

The self-fulfilling prophecy is a powerful phenomenon (Darley & Fazio, 1980; Harris & Rosenthal, 1985). But how does it work? As social perceivers, how do we transform our expectations of others into reality? Research suggests a three- step process. First, a perceiver forms an impression of a target person, which may be based on interactions with the target or on other information. Second, the perceiver behaves in a manner that is consistent with that first impression. Third, the target person unwittingly adjusts his or her behavior to the perceiver’s actions. The net result is behavioral confirmation of the first impression (see d Figure 4.13).

Now let’s straighten out this picture. It would be a sad commentary on human nature if each of us were so easily molded by others’ perceptions into appearing brilliant or stupid, introverted or extroverted, competitive or cooperative, warm or cold. The effects are well established, but there are limits. By viewing the self- fulfilling prophecy as a three-step process, social psychologists can identify the links in the chain that can be broken to prevent the vicious cycle.

Consider the first step, the link between one’s expectations and one’s behav- ior toward the target person. In the typical study, perceivers try to get to know the target on only a casual basis and are not necessarily driven to form an ac- curate impression. But when perceivers are highly motivated to seek the truth

d F i g u r e 4 . 1 3

The Self-Fulfilling Prophecy as a Three-Step Process

How do people transform their expectations into reality? (1) A perceiver has expectations of a target person; (2) the perceiver then behaves in a manner consistent with those expectations; and (3) the target unwittingly adjusts his or her behavior according to the perceiver’s actions.

en ga

ge L

ea rn

in g®

Step 1

Step 2Step 3

Perceiver's behavior toward the target

Target's behavior toward the perceiver

Perceiver's expectations

80220_ch04_ptg01_hr_104-153.indd 148 30/01/16 10:53 AM

149 Social Perception: The Bottom Line

(as when they need to consider the target as a possible teammate or opponent), they become more objective and often do not confirm prior expectations (Harris & Perkins, 1995; Hilton & Darley, 1991).

The link between expectations and behavior depends in other ways as well on a perceiver’s goals and motivations in the interaction (Snyder & Stukas, 1999). In one study, John Copeland (1994) put either the perceiver or the target into a posi- tion of relative power. In all cases, the perceiver interacted with a target person who was said to be introverted or extroverted. In half the pairs, the perceiver was given the power to accept or reject the target as a teammate for a money-winning game. In the other half, it was the target who was empowered to choose a teammate. The two participants interacted, the interaction was recorded, and neutral observers listened to the tapes and rated the target person. Did perceivers cause the targets to behave as introverted or extroverted, depending on initial expectations? Yes and no. Illustrating what Copeland called “prophecies of power,” the results showed that high-power perceivers triggered the self-fulfilling prophecy, as in past research, but that low-power perceivers did not. In the low-power situation, the perceivers spent less time getting to know the target person and more time trying to be liked.

Now consider the second step, the link between a perceiver’s behavior and the target’s response. In the designs of much of the past research (as in much of life), target persons were not aware of the false impressions held by others. Thus, it is unlikely that Rosenthal and Jacobson’s (1968) “late bloomers” knew of their teachers’ high expectations or that Snyder and Swann’s (1978) “introverts” and “extroverts” knew of their interviewers’ misconceptions. But what if they had known? How would you react if you found yourself being cast in a particular light? When it happened to participants in one experiment, they managed to overcome the effect by behaving in ways that forced the perceivers to abandon their expecta- tions (Hilton & Darley, 1985).

As you may recall from the discussion of self-verification in Chapter 3, this result is most likely to occur when the expectations of perceivers clash with a target person’s own self-concept. When targets who viewed them- selves as extroverted were interviewed by perceivers who believed they were introverted (and vice versa), what changed as a result of the interaction were the perceivers’ beliefs, not the targets’ behavior (Swann & Ely, 1984). Social perception is a two-way street; the persons we judge have their own prophecies to fulfill.

The notion that we can create a “self-fulfilling prophecy” by getting others to behave in ways we expect is a myth.

FALSe

Social Perception: The Bottom Line Trying to understand people—whether they are world leaders, crime suspects, professional athletes, celebrities, trial lawyers, or loved ones closer to home—is no easy task. As you reflect on the material in this chapter, you will notice that there are two radically different views of social perception. One suggests that the process is quick and relatively automatic. At the drop of a hat, without much thought, effort, or awareness, people make rapid-fire snap judgments about others based on physical appearance, preconceptions, cognitive heuris- tics, or just a hint of behavioral evidence. According to a second view, however, the process is far more mindful. People observe others carefully and reserve judgment until their analysis of the target person, behavior, and situation is complete. As suggested by theories of attribution and information integration,

80220_ch04_ptg01_hr_104-153.indd 149 30/01/16 10:53 AM

150 Chapter 4 Perceiving Persons

the process is eminently logical. In light of recent research, it is now safe to con- clude that both accounts of social perception are correct. Sometimes our judg- ments are made instantly; at other times, they are based on a more painstaking analysis of behavior. Either way, we often steer our interactions with others along a path that is narrowed by first impressions, a process that can set in motion a self-fulfilling prophecy. The various aspects of social perception, as described in this chapter, are summarized in d Figure 4.14.

At this point, we must confront an important question: How accurate are our impressions of each other? For years, this question has proved provocative but hard to answer (Cronbach, 1955; Jussim, 2012; Kenny, 1994; Stern et al., 2013; West & Kenny, 2011). Granted, people often depart from the ideals of logic and exhibit occasional bias in their social perceptions. In this chapter alone, we have seen that perceivers typically focus on the wrong cues to judge if someone is lying; use cognitive heuristics without regard for numerical base rates; overlook the situational influences on behavior; disparage victims whose misfortunes threaten their sense of justice; form premature first impressions; and then interpret, seek, and create evidence in ways that support these impressions.

To make matters worse, we often have little awareness of our limitations, leading us to feel overconfident in our judgments. In a series of studies, David Dunning and his colleagues (1990) asked college students to predict how a target person would react in various situations. Some made predictions about a fellow student whom they had just met and interviewed, and others made predictions about their roommates. In both cases, participants reported their confidence in each prediction, and accuracy was determined by the responses of the target persons themselves. The result: Regardless of whether they judged a stranger or a roommate, students consistently overestimated the accuracy of their predictions. In fact, Kruger and Dunning (1999) found that people who scored low on tests of spelling, logic, grammar, and humor appreciation were later the most likely to overestimate their own performance. These results

d F i g u r e 4 . 1 4

The Processes of Social Perception

Summarizing Chapter 4, this diagram depicts the processes of social perception. As shown, it begins with the observation of persons, situations, and behavior. Sometimes we make snap judgments from these cues. At other times, we form impressions only after making attributions and integrating these attributions. Either way, our impressions are subject to confirmation biases and the risk of self- fulfilling prophecy.

en ga

ge L

ea rn

in g®

Confirmation

Snap judgments

IntegrationAttributionObservation Persons Situations Behavior

ImpressionsDispositionsPerceiver

80220_ch04_ptg01_hr_104-153.indd 150 30/01/16 10:53 AM

151 Social Perception: The Bottom Line

suggest that poor performers may be doubly cursed: They don’t know what they don’t know (Dunning et al., 2003)—and they don’t know they are biased (Ehrlinger et al., 2005).

Standing back from the material presented in this chapter, you may find that the list of our shortcomings, punctuated by the problem of overconfi- dence, is long and depressing. How can this list be reconciled with the tri- umphs of civilization? To put it another way: “If we’re so dumb, how come we made it to the moon?” (Nisbett & Ross, 1980, p. 249). Yes, it is true that people fall prey to the biases identified by social psychologists and probably even to some that have not yet been noticed. It is also true that we often get fooled by con artists, misjudge our partners in marriage, vote for the wrong leaders, and hire the wrong job applicants, and that our biases can have real and harmful consequences—sometimes giving rise, as we’ll see in Chapter 5, to stereotypes, prejudice, and discrimination. Despite our imperfections, how- ever, there are reasons to be guardedly optimistic about our competence as social perceivers:

1. The more experience people have with each other, the more accurate they are. For example, although people have a limited ability to assess the personality or emotional states of strangers they meet in a laboratory, they are generally better at judging their own friends and acquaintances (Kenny & Acitelli, 2001; Levesque, 1997; Malloy & Albright, 1990; Sternglanz & DePaulo, 2004).

2. Although we are not good at making global judgments of others (that is, at know- ing what people are like across a range of settings), we are able to make more limited specific predictions of how others will behave in our own presence. You may well misjudge the personality of a roommate or co-worker, but to the extent that you can predict your roommate’s actions at home or your co-worker’s actions on the job, the mistakes may not matter (Swann, 1984).

3. People can form more accurate impressions of others when they are motivated by concerns for accuracy and open-mindedness (Biesanz & Human, 2010). Some of the studies described in this chapter have shown that people show less bias when the accuracy of their social perceptions are personally important—as when participants are asked to judge a prospective teammate’s ability to facilitate suc- cess in a future task (Fiske & Neuberg, 1990) or a future dating partner’s social competence (Goodwin et al., 2002).

4. Not everyone suffers from high levels of error and bias. Some individuals are more accurate than others in their social perceptions—for example, individuals who are psychologically well adjusted (Human & Biesanz, 2011).

To summarize, research on the accuracy of social perceptions offers a valu- able lesson: To the extent that we observe others with whom we’ve had time to interact, make judgments that are reasonably specific, are motivated to form an accurate impression, and are reasonably well adjusted, the problems that plague us can be minimized. Indeed, just being aware of the biases described in this chapter may well be a necessary first step toward a better understanding of others and our perceptions of them.

People are more accurate at judging the personalities of friends and acquaintances than of strangers.

True

80220_ch04_ptg01_hr_104-153.indd 151 30/01/16 10:53 AM

152 Chapter 4 Perceiving Persons

1. People often make snap judgments about other people based on superficial cues such as physical appearance and nonverbal demeanor.

2. Although we often form accurate impressions based on other people’s behavior, research shows that we are not accurate at knowing when someone else is telling the truth or lying.

3. Although we often form first impressions of others quickly, sometimes we are more analytical, observing behavior and making personal or situational attributions for it according to the logic of attribution theory.

4. At least in Western cultures, however, people com- mit the fundamental attribution error when observing others, overestimating the role of a person’s dispositions and underestimating the impact of the situation they are in.

5. In forming impressions of other people, three sets of traits loom as particularly important: how competent the person is, how warm, and how moral.

6. Although we tend to form impressions of others by mentally averaging their various traits, a number of bi- ases come into play—for example, some traits are more central to the overall impressions we form than others.

7. Once we form an impression of someone, we become slow to change that impression when faced with new information that is nonsupportive or even contradictory.

8. Once we form an impression, we engage in confirma- tion biases, tending to seek new information in ways that are likely to confirm what we already believe.

9. Beginning with classic research on the effects of teacher expectations on student performance, research also shows that our perceptions of others can influence our behavior, and in turn their behavior, resulting in a self- fulfilling prophecy.

10. Research promotes two radically different views of social perception. At times we are quick and instinctive social perceivers and at times we are careful and analytic; at times we are accurate, and at times we are biased and in error.

Top 10 Key Points in Chapter 4

Observation: The elements of Social Perception

The impressions we form of others are influenced by superficial aspects of their appearance.

T True. Research shows that first impressions are influenced by height, weight, clothing, facial characteristics, and other aspects of appearance.

Adaptively, people are skilled at knowing when someone is lying rather than telling the truth. F False. People frequently make mistakes in their

judgments of truth and deception, too often accepting what others say at face value.

Attribution: From elements to Dispositions

Like social psychologists, people are sensitive to situational causes when explaining the behavior of others. F False. In explaining the behavior of others, people

overestimate the importance of personal factors and overlook the impact of situations, a bias known as the fundamental attribution error.

Confirmation Biases: From impressions to reality

People are slow to change their first impressions on the basis of new information. T True. Studies have shown that once people form an

impression of someone, they become resistant to change even when faced with contradictory new evidence.

Putting Common Sense to the Test

Review

80220_ch04_ptg01_hr_104-153.indd 152 30/01/16 10:54 AM

153 Review

attribution theory (120) availability heuristic (123) base-rate fallacy (124) belief in a just world (133) belief perseverance (144) central traits (139) confirmation bias (142) counterfactual thinking (125)

covariation principle (122) false-consensus effect (124) fundamental attribution error (126)

impression formation (134) information integration theory (135)

mind perception (112)

need for closure (142) nonverbal behavior (113) personal attribution (120) primacy effect (141) priming (136) self-fulfilling prophecy (146) situational attribution (120) social perception (107)

Key Terms

The notion that we can create a “self-fulfilling prophecy” by getting others to behave in ways we expect is a myth. F False. In the laboratory and in the classroom, a

perceiver’s expectation can actually lead to its own fulfillment.

Social Perception: The Bottom Line

People are more accurate at judging the personalities of friends and acquaintances than of strangers. T True. People often form erroneous impressions

of strangers but tend to be more accurate in their judgments of friends and acquaintances.

80220_ch04_ptg01_hr_104-153.indd 153 30/01/16 10:54 AM

Stereotypes, Prejudice, and Discrimination

This chapter focuses on stereotyping, prejudice, and discrimination. We begin by examining the nature of the problem—how aspects of stereotyping, prejudice, and discrimination have changed dramatically in recent years, as well as how persistent they can be. Next, we examine a variety of causes underlying these problems. We conclude with a number of ways to reduce stereotypes, prejudice, and discrimination.

80220_ch05_ptg01_hr_154-207.indd 154 27/01/16 5:18 PM

Stereotypes, Prejudice, and Discrimination

5 The Nature of the Problem: Persistence and Change | 157

Defining Our Terms

Racism: Current Forms and Challenges

Sexism: Ambivalence, Objectification, and Double Standards

Beyond Racism and Sexism: Age, Weight, Sexuality, and Other Targets

Being Stigmatized

Stereotype Threat: A Threat in the Air

Causes of the Problem: Intergroup, Motivational, Cognitive, and Cultural Factors | 176

Social Categories and Intergroup Conflict

Social Identity Theory

Culture and Social Identity

Culture and Socialization

How Stereotypes Distort Perceptions and Resist Change

Automatic Stereotype Activation

The Shooter Bias

Reducing the Problem: Social Psychological Solutions | 196

Intergroup Contact

Intergroup Friendships and Extended Contact

The Jigsaw Classroom

Shared Identities

Trust, Belonging, and Reducing Stereotype Threat

Exerting Self-Control

Changing Cognitions, Cultures, and Motivations

Review

155Ge tt

y Im

ag es

80220_ch05_ptg01_hr_154-207.indd 155 27/01/16 5:18 PM

156 Chapter 5 Stereotypes, Prejudice, and Discrimination

Heavily-armed police and National Guard forces are called in as racial tensions erupt into protests and riots in multiple American communities in re- sponse to a series of killings of unarmed African American men by the police.

A white man opens fire on worshipers in an historic African-American church in South Carolina, killing nine; investigators report that the shooter told them he wanted to start a race war. Supreme Court justices are split five to four in a ruling about the legality of prac- tices that result in housing discrimination.

If you’re familiar with twentieth-century American history, this should sound like the volatile 1960s, as bat- tles over Civil Rights boiled over in many parts of the nation. But what we’ve just described occurred a full half century later, at the end of 2014 and in the first half of 2015. The sins of the past seemed to be repeating, demonstrating that although much has changed, much also remains the same.

The problems are by no means limited to black– white relations, or to the United States. Around the same time period, violence and hatred directed toward Jews throughout Europe and Australia were on the rise (see d Figure 5.1). Laws against pro-LGBT (lesbian, gay, bisexual, and transgender) “propaganda” were passed in Russia. Anti-Muslim sentiment increased in many parts of the West, while anti-West sentiment con- tinued to flourish among large numbers of radical Mus- lims. The prime minister of India decried the continuing (though officially banned) practice of parents selectively aborting girls. As billionaire Donald Trump announced his candidacy for president of the United States, he denounced Mexican immigrants as criminals and rap- ists and leaped to the top of the polls (Ahmed, 2015; Fletcher, 2014; Gjelten, 2015; Luke, 2015; Mahr, 2015).

1965 or 2015? Although there certainly has been great progress in civil rights and interracial relations in a half century, these two photos of African American men being arrested during protests about institutional racism in 1965 (left) and 2015 (right) illustrate the tragic persistence of some deep-rooted problems.

Circle Your Answer

T F Children do not tend to show biases based on race; it is only after they become adolescents that they learn to respond to people differently based on race.

T F Interracial interactions tend to go better and to reduce the perceptions of racism if a colorblind mentality is used, which denies or minimizes any acknowledgment of racial differences.

T F An African American student is likely to perform worse on an athletic task if the task is described as one reflecting sports intelligence than if it is described as reflecting natural athletic ability.

T F Being reminded of one’s own mortality makes people put things into greater perspective, thereby tending to reduce ingroup–outgroup distinctions and hostilities.

T F People’s very quick judgments are not influenced by a stereotype unless they actually believe the stereotype to be true.

Putting Common SenSe to the Test

Be tt

m an

n/ Co

rb is

A P

Im ag

es /J

ef f R

ob er

so n

80220_ch05_ptg01_hr_154-207.indd 156 27/01/16 5:18 PM

157 The Nature of the Problem: Persistence and Change

In this section, we discuss some of the progress that has been made concerning stereotypes, prejudice, and discrimination, along with the persistence of more subtle forms of these biases. To provide a focus and to reflect the topics that have most dominated the research literature, we will concentrate in this sec- tion primarily on racism and sexism in particular—even though many of the points hold true across a wide variety of targets of stereotypes, prejudice, and discrimination.

j Defining Our Terms Given the complexity of these issues, defining concepts such as prejudice or rac- ism is no simple matter. Debates persist about how best to define the terms—how broad or specific they should be, whether they should focus on individual or institutional levels, and so on. For example, one way to define racism is as preju- dice and discrimination based on a person’s racial background. It is important to realize, however, that racism exists at several different levels.

At the individual level, as this definition reflects, any of us can be racist to- ward anyone else. At the institutional and cultural levels, in contrast, some people

Faced with these headlines, it’s easy to lose sight of the fact that progress, in some cases tremendous prog- ress, has been made. The United States had elected, and then re-elected, its first African American president. The Supreme Court ruled that same-sex marriage was now legal throughout the United States. Today more people than ever rush to defend the targets and denounce the perpetrators of prejudice and discrimination. The march toward progress is real, but its rhythm is frustratingly unsteady, at its best a “two-steps forward and one-step back” motion.

To better understand and improve our diverse world, to help the march toward progress acceler- ate in the right direction, it is critically important to understand the complexity and causes of stereo- types, prejudice, and discrimination. That is the pri- mary goal of this chapter. We begin by taking a close look at the nature of the problem of intergroup bias in contemporary life. Later in the chapter we address some of the key causes and important consequences of intergroup biases, and we close by discussing some of the most promising directions in efforts to reduce these problems.

The Nature of the Problem: Persistence and Change

racism Prejudice and discrimination based on a person’s racial background, or institutional and cultural practices that promote the domination of one racial group over another.

d F i g u r e 5 . 1

Harassment of Jews reaches Seven-Year High

This is one of many indicators of the increase in prejudice and discrimination against Jews in recent years. These numbers are higher than for any other religious group, and are especially high in Europe, where Jews faced harassment by the government or social groups in 76% of the countries. Pew Research Center (2015a).

June 2007

Dec 2013

Dec 2012

Dec 2011

June 2010

June 2009

June 2008

% of Countries Where Jews Were Harassed, Either by Government or Social Groups

80220_ch05_ptg01_hr_154-207.indd 157 27/01/16 5:18 PM

158 Chapter 5 Stereotypes, Prejudice, and Discrimination

are privileged while others are disadvantaged. Aspects of various institutions and the culture more generally may perpetuate this inequality, even if unintention- ally. For example, institutions may unwittingly perpetuate racism by tending to accept or hire individuals similar or connected to the people who already are in the institution, and popular culture may signal what kinds of people are most and least valued. Therefore, another way to define racism is as institutional and cultural practices that promote the domination of one racial group over another (Jones, 1997b). The incidents described at the beginning of this chapter sparked much discussion about systemic racism, which concerns these institutional and cultural levels of racial discrimination. Similarly, sexism may be defined as preju- dice and discrimination based on a person’s gender or as institutional and cultural practices that promote the domination of one gender (typically men) over another (typically women).

For the purposes of this chapter, we define stereotypes as beliefs or asso- ciations that link whole groups of people with certain traits or characteristics. Prejudice consists of negative feelings about others because of their connec- tion to a social group. Whereas stereotypes concern associations or beliefs and prejudice concerns feelings, discrimination concerns behaviors—specifically, negative behaviors directed against persons because of their membership in a particular group. Stereotypes, prejudice, and discrimination can operate some- what independently, but they often influence and reinforce each other.

j racism: Current Forms and Challenges A close examination of legislation, opinion polls, sociological data, and social psy- chological research indicates that racial prejudice and discrimination have been decreasing in the United States and in many other countries over the last 70 years, although elements of it may once again be on the rise, particularly in Western Europe. In a classic study of ethnic stereotypes published in 1933, Daniel Katz and Kenneth Braly found that white college students viewed the average white American as smart, industrious, and ambitious, and they saw the average African American as superstitious, ignorant, lazy, and happy-go-lucky. In multiple follow- up surveys with demographically similar samples of white students conducted from 1951 through 2001, these negative images of blacks largely faded and were replaced by more favorable images (Dovidio et al., 1996; Madon et al., 2001). For example, 75% of white participants chose “lazy” as a trait to describe black Americans in 1933, whereas only 5% did so 60 years later. Similarly, public opin- ion polls have indicated that racial prejudice in the United States has dropped sharply since World War II. d Figure 5.2 depicts one dramatic example of this trend, concerning attitudes toward inter-racial marriage (Newport, 2015).

The election of Barack Obama as the first African American president of the United States in 2008 was a significant sign of racial progress, as was his re- election in 2012. As Obama pointed out when he was inaugurated, his own father would not have been served in many restaurants in the nation’s capital 60 years before, and now that father’s son was being sworn in to the highest office in the land. That amount of progress in a person’s lifetime is staggering. However, when those of us who study stereotypes, prejudice, and discrimination saw story after story in the popular media at the time of Obama’s election heralding the dawn of a “postracial America,” we knew how naive and wrong such notions were.

While serving as president, for example, Obama—along his wife, Michelle— was the target with shocking regularity of hateful racial epithets and blatant ste- reotypes hurled by politicians, police officials, journalists, political cartoonists,

sexism Prejudice and discrimination based on a person’s gender, or institutional and cultural practices that promote the domination of one gender over another.

stereotype A belief or association that links a whole group of people with certain traits or characteristics.

prejudice Negative feelings toward persons based on their membership in certain groups.

discrimination Behavior directed against persons because of their membership in a particular group.

80220_ch05_ptg01_hr_154-207.indd 158 27/01/16 5:18 PM

159 The Nature of the Problem: Persistence and Change

celebrities, and ordinary citizens. During this time the spate of killings of unarmed African American men by police in several cities led to investigations that revealed persistent systemic racial discrimination. Data from archival analyses, surveys, and even experiments in which other variables are held constant continue to show how African American and Hispanic individuals suffer in comparison to white Americans in housing, employment, salaries, incarceration rates, and a host of other important quality-of-life variables (Gabrielson et al., 2014; Horwitz, 2015; Pager & Shepherd, 2008; Reskin, 2012).

In sum, then, there are legitimate reasons both to celebrate racial progress and to acknowledge that racism remains a fact of life and is by no means limited merely to the actions of some fringe individuals or groups. And as we will see in the following section, it exists in ways that escape the recognition of most people.

Modern, Aversive, and implicit racism Consider two stories from the world of sports:

1. As the first half ended during a high school basketball game near Pittsburgh, fans of a predominately white high school ran onto the court in full body banana suits, sur- rounded the players from the predominately African American opposing team, and “allegedly began making monkey noises and hurling racial epithets” at the players (C. Smith, 2012). A few months later in Poland and Croatia, black members of the Dutch and Italian national soccer teams were the targets of racial abuse—including being on the receiving end of monkey chants and a flung banana—during the Euro 2012 championships. Indeed, the 2012 tournament, and the World Cup tournament in 2014 were marred by a variety of incidents of racist taunting by players and fans from mul- tiple European and South American nations (Barnes, 2014; Brown, 2012; Cue, 2012).

2. Christopher Parsons and others (2009) analyzed every pitch from four Major League Baseball seasons—more than 3.5 million pitches in all—and found a

d F i g u r e 5 . 2

Changes in Attitudes Toward interracial Marriage

Source: Newport (2015).

‘59 ‘62 ‘10‘07‘04‘01‘98‘95‘92‘89‘86‘83‘80‘77‘74‘71‘68‘65 ‘13

29 36

43 48

76 79

77 86

87 73

Do You Approve or Disapprove of Marriage Between Blacks and Whites?

1958 wording: “... marriages between white and colored people” 1968-1978 wording: “... marriages between whites and nonwhites”

% Approve

80220_ch05_ptg01_hr_154-207.indd 159 27/01/16 5:18 PM

160 Chapter 5 Stereotypes, Prejudice, and Discrimination

fascinating set of results. Umpires were more likely to call strikes for pitchers who were of the same race/ethnicity as they were. Even more interesting is the fact that this bias emerged only under three conditions: (1) if the game was played in the subset of ballparks that did not have a computerized monitoring system the league was using to review umpires’ performance in calling balls and strikes; (2) if the number of people attending the game was relatively low; and (3) if the call would not be the final ball or strike for the player at bat. In other words, the racial/ethnic bias was evident only under the conditions when there would be the least accountability or public outcry.

The first of these examples—concerning racist taunting—illustrates what some call old-fashioned racism. It is blatant, explicit, and unmistakable. The second— concerning the bias in umpiring—is what some call modern racism, a subtle form of prejudice that tends to surface when it is safe, socially acceptable, or easy to rationalize. Modern racism is far more subtle and most likely to be present under the cloud of ambiguity. Like germs lurking beneath a seemingly clean countertop, stereotypes, prejudice, and discrimination in contemporary life live under the sur- face to a much, much greater extent than most people realize. And like germs, their existence can have a profound effect on us, despite how hidden they may be.

According to theories of modern racism, many people are racially ambivalent. They want to see themselves as fair, but they still harbor feelings of anxiety and discomfort about other racial groups (Hass et al., 1992). There are several spe- cific theories of modern racism, but they all emphasize contradictions and ten- sions that lead to subtle, often unconscious forms of prejudice and discrimination (Gawronski et al., 2012; Levy et al., 2013; Nier & Gaertner, 2012). For example, Samuel Gaertner and John Dovidio (1986; Hodson et al., 2010) proposed the related concept of aversive racism, which concerns the ambivalence between individu- als’ sincerely fair-minded attitudes and beliefs, on the one hand, and their largely unconscious and unrecognized prejudicial feelings and beliefs, on the other hand. In addition, some scholars today use the term microaggression to characterize the everyday, typically subtle but hurtful forms of discrimination that are experienced quite frequently by members of targeted groups (Forrest-Bank et al., 2015).

An opponent tries to prevent Barcelona’s Samuel Eto from leaving the game after he was targeted by racist taunts from Spanish League fans.

A P

Im ag

es /M

an u

Fe rn

an de

modern racism A form of prejudice that surfaces in subtle ways when it is safe, socially acceptable, and easy to rationalize.

aversive racism Racism that concerns the ambivalence between fair-minded attitudes and beliefs, on the one hand, and unconscious and unrecognized prejudicial feelings and beliefs, on the other hand.

80220_ch05_ptg01_hr_154-207.indd 160 27/01/16 5:18 PM

161 The Nature of the Problem: Persistence and Change

In these forms of racism, prejudice and discrimination surface in very subtle ways, often clouded by ambiguity. For example, several studies have found that white participants playing the role of jurors may be more likely to convict a black than a white defendant for a crime when the evidence is rather ambiguous, and thus one can justify either a guilty or not-guilty verdict for reasons having nothing to do with race. If, on the other hand, race seems to be an important aspect of the case, then this bias may be eliminated or even reversed (Bucolo & Cohn, 2010; Fein et al., 1997; Sommers & Ellsworth, 2009).

These modern, ambivalent forms of racism are often evident in the “but some of my best friends are . . .” excuse. That is, people establish their moral creden- tials of not being racist by demonstrating—to others or even to themselves—that they have good friends from the racial or ethnic group in question or they have behaved in ways that were quite fair to members of this group (Cascio & Plant, 2015; Merritt et al., 2010). Having such good behavior to their credit gives people the license to take actions that might otherwise put them at risk for seeming preju- diced. Indeed, Anna Merritt and others (2012) found that people sometimes go out of their way to try to establish such credits, such as by rating an African American job candidate more positively, if they anticipate being in a situation later in which they might be judged as racist.

In just about any setting, examples of subtle but impactful discrimination can be found. Here is just a sample from recent research: In the workplace, businesses were less likely to contact job candidates based on their resumes if the name of the candidate was stereotypically African American than European American; in the classroom, students expected lower competence from a pro- fessor with a stereotypically African American rather than European American name, and professors were less likely to reply to e-mail requests about research opportunities if the name of the sender seemed African American than European American; in the marketplace, car salespeople charged black customers higher prices than white customers for the same cars, and iPods sold on the Internet received lower offers when the ads showed the iPods being held by black hands rather than by white hands (Bavishi et al., 2010; Brewster & Lynn, 2014; Doleac & Stein, 2013; Eberhardt et al., 2006; E. B. King et al., 2006; Milkman et al., 2015; Morgan et al., 2013).

Let’s take a closer look at one recent finding along these lines. Jason Okonofua and Jennifer Eberhardt (2015) gave schoolteachers information about an incident in which a middle-school child misbehaved (such as by sleeping in class and then not listening to the teacher when the teacher wakes him up). The teachers rated how troubled they would be by the behavior if they were the teacher and how se- vere the discipline of the child should be. Then they read about a second incident of misbehavior by this same child. The teachers then made the same judgments a second time.

For half of the teachers, the name of the boy was, according to previous re- search, stereotypically black (Darnell or DeShawn), and for the others it was stereotypically white (Jake or Greg). The researchers found no bias in teachers’ judgments after the first infraction. That is, they reported being equally troubled and recommended equal punishments whether the boy seemed to be black or white. This would suggest no implicit racism here, right? But after the second infraction, a different story emerges. The teachers reported being more troubled, and recommended more severe discipline, after the second round of misbehavior if he apparently was black than white (see d Figure 5.3).

All these examples suggest two key points. First, although these biases are often very difficult to see, they are present in abundance, across a multitude

80220_ch05_ptg01_hr_154-207.indd 161 27/01/16 5:18 PM

162 Chapter 5 Stereotypes, Prejudice, and Discrimination

of settings, ranging from the everyday to the profound. On the more profound side of things, consider a pair of findings in the legal system. Jennifer Eberhardt and others (2006) examined more than 600 death-penalty-eligible trials from Philadelphia and found that in cases involving a white victim and a black male defendant, the more the defendant’s physical appearance was stereotypically black, the more likely he would be sentenced to death. Jill Viglione and others (2011) found in a sample of more than 12,000 adult black women imprisoned in North Carolina that black women with lighter skin color received on average about 12% less prison time than black women with darker skin (see d Figure 5.4).

The second key point is that in each of the cases described in the previous several paragraphs, it would be impossible to be sure that racism was behind any one person’s specific behavior. After all, there may be many reasons why any one person might receive a job offer or not, or a severe punishment or not. There’s just too much ambiguity or subtlety. This is even more true when looking at our own behaviors than when looking at those of others, as we tend to think that we’re less biased than others (Scopelliti, 2015). But by looking at trends and by running experiments with careful controls, the underlying bias can be revealed.

Measuring implicit racism The modern forms of racism sometime operate con- sciously but more frequently operate outside people’s conscious awareness. Schol- ars call racism that operates unconsciously and unintentionally implicit racism. This raises a question: How can we measure how implicitly biased someone is? Because of

d F i g u r e 5 . 3

Two Strikes: race and Teachers’ reactions to Children’s Misbehavior

Teachers read about an instance in which a middle-school child misbehaved in class. The teachers reported being no more troubled by, nor did they suggest more severe discipline, in response to the misbehavior if they were led to believe the child was black than white. However, after the child misbehaved a second time, a significant racial bias emerged, with teachers being more troubled and recommending more severe discipline if the child was black than white. Based on Okonofua & Eberhardt (2015)

implicit racism Racism that operates unconsciously and unintentionally.

Ra tin

Feeling Troubled

First infraction Second infraction

Disciplinary Action

First infraction Second infraction 1

White Students Black Students

80220_ch05_ptg01_hr_154-207.indd 162 27/01/16 5:18 PM

163 The Nature of the Problem: Persistence and Change

its implicit nature, it can’t be assessed by simply asking people to answer some ques- tions about their attitudes. Rather, much more subtle, indirect measures typically are used. By far the most well known of these measures is the Implicit Association Test (IAT), first developed and tested by Anthony Greenwald and others (1998). The IAT measures the extent to which two concepts are associated. It measures implicit rac- ism toward African Americans, for instance, by comparing how quickly participants associate African American cues (such as a black face) with negative or positive con- cepts compared to how quickly they associate European American cues with these same concepts. If someone is consistently slower identifying something good after seeing a black face than a white face, for example, this would indicate a degree of implicit racism.

Other IATs focus on associations concerning older versus younger people, men versus women, and so on. The IAT is discussed in more detail in Chapter 6 on Attitudes. It has sparked an explosion of research, with many hundreds of scien- tific publications already, along with some controversy about its validity and uses. The IAT has been so popular that between October 1998 and October 2015 more than 17 million IATs were completed by visitors to the IAT website (Nosek, 2015).

Implicit racial bias as measured by the IAT has been found between groups around the world and even among children as young as 3 and 4 years old (Baron & Banaji, 2006; Dunham et al., 2008, 2013, 2015; Newheiser & Olson, 2012). One interesting point is that compared to 6-year-olds, 10-year-olds begin to show less bias on explicit measures, and adults exhibit even less than the 10-year-olds; how- ever, levels of implicit racism as measured by the IAT remain consistent across children and adults.

Implicit racism correlates with a variety of attitudes and behaviors. For example, higher implicit racism by white participants in several studies predicted negative,

d F i g u r e 5 . 4

Facial Features and Prison Sentences

According to a study by Jill Viglione and others (2011) of more than 12,000 adult black women imprisoned in North Carolina, the chances are good that if the two women depicted here were each found guilty of a crime, the woman on the left (whose face would be considered more “stereotypically black”) would receive a longer prison sentence than the woman on the right.

Children do not tend to show biases based on race; it is only after they become adolescents that they learn to respond to people differently based on race.

FAlSe

Lu m

in ai

m ag

es /S

hu tt

er st

oc k.

co m

Fa nc

y Ph

ot og

ra ph

y/ Ve

80220_ch05_ptg01_hr_154-207.indd 163 27/01/16 5:19 PM

164 Chapter 5 Stereotypes, Prejudice, and Discrimination

unfriendly nonverbal behaviors in interracial interactions, such as physical distanc- ing or lack of eye contact—actions that can make the other person feel disliked and that can lead to, for instance, poor performance in a job interview (Greenwald et al., 2009). A particularly disturbing, and growing, set of findings concerns the link be- tween doctors’ and other health care providers’ implicit racism and their treatment of patients from racial and ethnic minority groups (Blair et al., 2015; Penner et al., 2013; Schaa et al., 2015). For example, Janice Sabin and Anthony Greenwald (2012) found that doctors with stronger pro-white bias on the IAT were more likely to recommend prescribing pain-relieving medication after surgery for white patients, and less likely to do so for black patients, compared to physicians with less of a pro-white IAT score.

interracial interactions The divides between racial and ethnic groups tend to be more vast and may promote stronger feelings of hostility, fear, and distrust than the divides based on other social categories, such as those based on gender, appearance, and age. This can make interracial interaction particularly challenging and fraught with emotion and tension. When engaging in interracial interactions, whites may be concerned about a number of things, including not wanting to be, or to appear to be, racist. They may therefore try to regulate their behaviors, be on the lookout for signs of distrust or dislike from their interaction partners, and so on. Because of these concerns, what should ideally be a smooth-flowing normal interaction can become awkward and even exhausting. This, in turn, can affect their partner’s perceptions of them, possibly leading to the ironic outcome of well-intentioned individuals

appearing to be racist precisely because they were trying not to be. Because of these concerns, engaging in interracial interactions can be so stressful as to leave the individuals cognitively exhausted, less able immediately after the interaction to complete mental tasks (Shelton & Richeson, 2015).

According to Jacquie Vorauer (2003; Vorauer & Sasaki, 2014), in- dividuals engaging in intergroup interactions often activate metaste- reotypes, or thoughts about the outgroup’s stereotypes about them, and worry about being seen as consistent with these stereotypes. In- deed, this kind of concern about interracial interactions can lead to unhealthy cardiovascular reactions associated with feelings of threat (Mendes et al., 2002; Trawalter et al., 2012). It should not be surpris- ing, then, that people sometimes try to avoid interracial interaction for fear of appearing racist or being treated in a racist way, and this avoidant behavior can have the ironic effect of making things all the worse. Ashby Plant and David Butz (2006), for example, found that when nonblack participants with this avoidant concern interacted with a black confederate, they had shorter and less pleasant interac- tions. Philip Goff, Claude Steele, and Paul Davies (2008) found that white male students sat farther away from black students than white students in an interaction—but only if they thought they would be engaged in a conversation about a racially sensitive topic.

In a clever demonstration of the kind of anxiety whites sometimes feel about race, Michael Norton and others (2006) paired white par- ticipants with either a white or black confederate in a game (similar to the children’s board game Guess Who?) that required the partici- pants to ask the confederate questions so they could guess which of a series of photographs the confederate had been given. As can be seen in d Figure 5.5, participants were significantly less likely to ask about the race of the person in the photograph when playing the game with

d F i g u r e 5 . 5

Colorblind?

When white participants played a face-matching game in which they had to ask questions of a confederate to guess which of a series of photographs the confederate had, they were much less likely to ask about the race of the people in the photographs if they were interacting with a black confederate than a white confederate, even though this hurt their performance in the game. Based on Norton et al., 2006.

Mention Race?

White Black Confederate’s race

100

% o

f Tr

ia ls

in w

hi ch

Ra ce

W as

M en

tio ne

80220_ch05_ptg01_hr_154-207.indd 164 27/01/16 5:19 PM

165 The Nature of the Problem: Persistence and Change

a black confederate than a white confederate, even though this hurt their ability to win the game. It seemed that the white participants would rather lose the game than run the risk of appearing racist by paying any attention to the race of the people in the photographs.

An interesting follow-up to this study examined the performance of children in this task (Apfelbaum et al., 2008). On a race-neutral version of the game, older children (10 and 11 years old), not surprisingly, outperformed younger children (8 and 9 years old). However, when race was a relevant category, the older chil- dren were much more likely to avoid asking about race, presumably because they were more aware of the sensitivities surrounding race. The result of this was that in the race-relevant version of the game, the younger children significantly outper- formed the older children!

White adults in an interracial interaction often try to adopt a “colorblind” mentality and demeanor, acting—or trying to act—as if race is so unimportant to them that they don’t even notice and certainly don’t care about their interaction partner’s race. Often this attempt is sincere and with the best of intentions, but a growing body of research findings indicates that this approach often backfires and makes members of racial minority groups more, rather than less, uncomfort- able. A multicultural approach that acknowledges and positively values racial and ethnic differences is often more effective in promoting better intergroup attitudes and behaviors (Plaut et al., 2015; Rattan & Ambady, 2013). Lisa Rosenthal and Sheri Levy (2013) similarly find that thinking in a more polycultural way—that is, focusing on the ways that racial and ethnic groups have interacted and influenced each other’s cultures throughout history—can be very effective in promoting positive intergroup relations.

Anxieties and challenges associated with intergroup interactions are not lim- ited to those between whites and blacks, of course. Research has demonstrated that feelings of threat and anxiety affect interactions involving a large number of groups, such as gay and lesbian individuals, overweight people, and a variety of other social categories (Blair et al., 2003; Everly et al., 2012; Madera & Hebl, 2012; Newheiser et al., 2015; Stephan, 2014).

j Sexism: Ambivalence, Objectification, and Double Standards

As with racism, old-fashioned blatant displays of sexism are less socially accepted than in years past, although they continue to exist at a frequency and with an intensity that would surprise many. As with racism, researchers have been docu- menting and studying modern and implicit forms of sexism that tend to escape the notice of most people but that can exert powerful discriminatory effects (Girvan et al., 2015; Swim & Hyers, 2009).

There are some ways that sexism is different, however. Gender stereotypes are distinct from virtually all other stereotypes in that they often are prescriptive rather than merely descriptive. In other words, they indicate what many people in a given culture believe men and women should be like, not merely what people think they actually are like. Few people, for example, think that gays should be artistic and sensitive or that old people should be forgetful and conservative, but many think that women should be nurturing and that men should be unemotional. Therefore, women who exhibit traits that are valued in society but that defy gender stereo- types, such as by being ambitious or assertive, often are viewed in especially harsh terms, contributing to the double standards that are a hallmark of sexism

interracial interactions tend to go better and to reduce the perceptions of racism if a colorblind mentality is used, which denies or minimizes any acknowledgment of racial differences.

FAlSe

80220_ch05_ptg01_hr_154-207.indd 165 27/01/16 5:19 PM

166 Chapter 5 Stereotypes, Prejudice, and Discrimination

(Brescoll et al., 2010; Prentice & Carranza, 2002; Rudman et al., 2012; Rudman, Fetterolf, & Sanchez, 2013).

Another way that sexism is unique concerns the degree to which the ingroup and outgroup members interact. Men and women are intimately familiar with each other. Girls and boys often grow up together, women and men often live together. In contrast to the effects of contact in reducing many other intergroup biases, however, all this contact between women and men often does little to re- duce sexist beliefs, attitudes, and behaviors.

Ambivalent Sexism It may surprise you to learn that, overall, stereotypes of women tend to be more positive than those of men (Eagly et al., 1994). However, the positive traits associated with women are less valued in important domains such as the business world than the positive traits associated with men.

These contradictions are reflected in Peter Glick and Susan Fiske’s (2001, 2012) concept of ambivalent sexism. Ambivalent sexism consists of two elements: hostile sexism, characterized by negative, resentful feelings about women’s abili- ties, value, and challenge to men’s power (e.g., “Women seek special favors under the guise of equality”), and benevolent sexism, characterized by affectionate, chiv- alrous feelings founded on the potentially patronizing belief that women need and deserve protection (e.g., “Women should be cherished and protected by men”). Benevolent sexism, on the surface, does not strike many women or men as ter- ribly troubling, but the two forms of sexism are positively correlated. Benevolent sexism is associated in particular with negative reactions toward women who defy traditional gender roles and stereotypes.

Both types of sexism are associated with supporting gender inequality in a variety of ways, and both predict many kinds of discriminatory behaviors and negative consequences (Durán et al., 2011; Masser et al., 2010; Rudman & Fetterolf, 2014). For example, Allison Skinner and others (2015) found that, depending on the context, hostile and benevolent sexism each predicted more negative judgments of the driver in an accident if the driver was said to be a woman rather than a man. Kristen Salomon and colleagues (2015) found that being the target of either type of sexism triggered negative cardiovascular responses in the women in their study.

Glick, Fiske, and others (2000) conducted an ambitious study of 15,000 men and women in 19 nations across 6 continents and found that ambivalent sexism was prevalent around the world. Among their most intriguing findings was that people from countries with the greatest degree of economic and political inequal- ity between the sexes tended to exhibit the most hostile and benevolent sexism.

Objectification Women are all too often treated in objectifying ways. That is, they are viewed or treated more as mere bodies or objects and less as fully func- tioning human beings. The advertising industry specifically, and the popular media more generally, are filled with imagery of women represented as sexual objects or just parts of a body (Kilbourne, 2003). For example, Julie Stankiewicz and Fran- cine Rosselli (2008) examined almost 2,000 advertisements depicting women from 58 popular magazines in the United States and found that half of them featured women as sex objects. Women also experience being treated and seen as objects in numerous interactions in their real lives. In one study involving in-depth interviews with 600 women in and around Paris, France, an astonishing 100% reporting having been sexually harassed while using public transportation (Palet, 2015)!

Although men are objectified in the media as well (and this is a growing trend), and are sometimes objectified in real interactions, it is still the case that

ambivalent sexism A form of sexism characterized by attitudes about women that reflect both negative, resentful beliefs and feelings and affectionate and chivalrous but potentially patronizing beliefs and feelings.

80220_ch05_ptg01_hr_154-207.indd 166 27/01/16 5:19 PM

167 The Nature of the Problem: Persistence and Change

women experience this much more frequently, and a good deal of research documents a variety of negative effects of this objectification on women, including on their mental and physical health, their academic performance, and their social interactions (Calogero et al., 2011; Fredrickson et al., 1998; Saguy et al., 2010; Tiggemann & Williams, 2012).

Sex Discrimination: Double Standards and Perva- sive Stereotypes Many years ago, Philip Goldberg (1968) asked students at a small women’s college to evaluate the con- tent and writing style of some articles. When the material was supposedly written by John McKay rather than Joan McKay it received higher ratings, a result that led Goldberg to wonder if even women were prejudiced against women. Certain other studies showed that people often devalue the performance of women who take on tasks usually reserved for men (Lott, 1985) and attribute women’s achievements to luck rather than ability (Deaux & Emswiller, 1974; Nieva & Gutek, 1981). These studies generated a lot of attention, but it now appears that this kind of devaluation of women is not commonly found in similar studies. More than 100 studies modeled after Gold- berg’s indicate that people are not generally biased by gender in the evaluation of performance (Swim & Sanna, 1996; Top, 1991). More recently, two different sets of studies involved having professors from science-related fields evaluate the ma- terials of candidates for research or faculty positions in their fields, and these studies found completely contradictory re- sults: One found a bias in favor candidates if they were said to be male (Moss-Racusin et al., 2012), and the other found a bias in favor candidates if they were said to be female (Williams & Ceci, 2015). The point here is that in studies involving the evaluation of materials that are identical except for the gender of the person who supposedly produced the materials, the findings have been inconsistent.

What is quite clear, however, is that sex discrimination continues to exist in numerous other ways and instances. As with forms of modern and implicit rac- ism, subtle but impactful examples of sexism abound. For example, Juan Madera and others (2015) looked at real letters of recommendation that professors wrote for candidates for academic jobs. A quick look at these letters probably would not reveal any obvious differences based on the gender of the candidates. A more thorough analysis, however, indicated that both male and female professors tended to include more pieces of information raising slight doubts (e.g., “she has a somewhat challenging personality,” “she might make a good colleague”) for female than for male candidates. Moreover, these seemingly minor doubts made a significant difference in the evaluations of people reading these letters.

Sexism today is by no means limited to subtle biases. In many parts of the world blatant sexism not only is still quite evident, but it is the law of the land. A law in Morocco that enables rapists to escape prosecution by marrying their victim came under fire when a 16-year-old girl committed suicide after a court ordered her to marry the man who raped her (Hirsch, 2012). Malala Yousafzai, a Pakistani teenage girl, was shot in the head in 2012 by the Taliban when she de- fied their bans against girls attending school. She survived the attack, became an inspiring activist dedicated to female education, and in October 2014 became the youngest person to win a Nobel Peace Prize.

After surviving being shot in the head by the Taliban for defying their ban against girls attending school in Pakistan, Malala Yousafzai has worked tirelessly in her quest to help girls around the world receive education. She became the youngest person to win the Nobel Peace Price in 2014.

Th e

A sa

hi S

hi m

bu n/

G et

ty Im

ag es

80220_ch05_ptg01_hr_154-207.indd 167 27/01/16 5:19 PM

168 Chapter 5 Stereotypes, Prejudice, and Discrimination

Other examples are far less extreme but still important. Look at d Figure 5.6 and Table 5.1, for instance, and you’ll notice some striking sex differences in occupational choice. How many female airline pilots have you met lately? What about male dental hygienists? (It’s interesting to note that 98% of the dental hy- gienists in the United States in 2011 were women, but only 22% of the dentists were.) The question is, of course, what explains these differences? Decades of social science research point to sexist attitudes and discrimination as a key part of the equation. Sex discrimination during the early school years may pave the way for diverging career paths in adulthood. Then, when equally qualified men and women compete for a job, gender considerations enter in once again, as business professionals and others favor men for so-called masculine jobs (such as a manager for a machinery company) and women for so-called feminine jobs (such as a receptionist) (Barbulescu & Bidwell, 2013; Caleo & Heilman, 2014; Koch et al., 2015).

d F i g u r e 5 . 6

Percentage of Women in Specific Occupations in the united States, 2011

Recent labor statistics reveal that men and women occupy very different positions in the U.S. workforce. U.S. Bureau of Labor Statistics, 2012.

40 1000 20 8060

Speech-language pathologists

Social workers

Tailors, dressmakers, and sewers

Elementary and middle school teachers

Meeting, convention, and event planners

Psychologists

Hotel, motel, and resort desk clerks

Retail salespersons

Postal service mail carriers

Accountants and auditors

Total, 16 years and over

New analysts, reporters, and correspondents

Physicians and surgeons

Computer systems analysts

Lawyers

Chief executives

Chefs and head cooks

Firefighters

Security guards and gaming surveillance officers

Architects and engineers

Aircraft pilots and flight engineers

Percent women

80220_ch05_ptg01_hr_154-207.indd 168 27/01/16 5:19 PM

169 The Nature of the Problem: Persistence and Change

Even when women and men have comparable jobs, the odds are good that the women will be paid less than their male counterparts and will be confronted with a so-called glass ceiling that makes it harder for women to rise to the highest positions of power in a business or organization (Evers & Sieverding, 2014; Gorman & Kmec, 2009). A 2011 study by Anthony Lo Sasso and others found that the starting salaries of new female physicians trained in New York State averaged almost $17,000 less than that of their male counterparts! (This difference could not be accounted for by things such as different areas of specialization, work hours, or setting.) Reshma Jasi and others (2013) found a similar result with a different sample of newly hired physicians.

Women vying for jobs, career advancement, and political office often face a virtually impossible dilemma: They are seen as more competent if they present themselves with stereotypically masculine rather than feminine traits, yet when they do this, they are also perceived as less socially skilled and attractive—a per- ception that may ultimately cost them the position or career advancement they were seeking (Phelan et al., 2008; Rudman et al., 2012; Schneider & Bos, 2014).

For both women and men, being in a job that is traditionally seen as more typical of the other gender can be especially challenging. Consistent with this idea are the results of a study by Victoria Brescoll and others (2010). These researchers found that men and women were judged more harshly for a mistake made on a job traditionally held by the other sex than for the same kind of mistake on a job in

According to a 2011 national survey by Gallup, 40% of Ameri- cans would prefer to have a boy compared to 28% preferring a girl, if they could have only one child (the rest indicated no pref- erence). These percentages are almost identical to those from a similar survey in 1941.

Data from International Labour Office, 2008.

m T a b l e 5 . 1

Women in Work Settings in Selected Countries around the World

Country Total Workforce Clerks Craft and Trade Workers

legislators, Senior Officials, Managers

Sales and Service Workers

Australia 45% 67% 5% 37% 68%

Canada 47% 77% 8% 37% 63%

Columbia 39% 58% 19% 46% 61%

Costa Rica 37% 57% 14% 27% 54%

Egypt 21% 29% 3% 11% 10%

Iran 18% 24% 23% 13% 11%

Israel 46% 74% 5% 30% 60%

Italy 39% 59% 14% 33% 57%

Republic of Korea 42% 52% 15% 9% 63%

Mexico 37% 61% 24% 31% 54%

Morocco 27% 25% 19% 12% 6%

Netherlands 45% 69% 5% 28% 69%

United Kingdom 46% 78% 8% 34% 76%

These international labor statistics show the percentage of workers in each category who are women in each of several countries. These data show some of the differences between nations, as well as some similarities concerning which types of jobs women are more or less likely to occupy around the world.

80220_ch05_ptg01_hr_154-207.indd 169 27/01/16 5:19 PM

170 Chapter 5 Stereotypes, Prejudice, and Discrimination

which their gender was the majority. In another study Brescoll and others (2012) found that men were perceived as less masculine if they worked for a female super- visor in a traditionally masculine occupation. Laurie Rudman, Kris Mescher, and Corinne Moss-Racusin (2013) found in a series of experiments that women liked men described as caring about issues of gender equality and women’s rights, but both these women and other men judged these supportive men in more feminized ways compared to men who were less supportive of women, perceiving them to be relatively weak and questioning their sexuality. Examples like these show that defying traditional gender roles can still be daunting.

j Beyond racism and Sexism: Age, Weight, Sexuality, and Other Targets

We have focused in this section on racism and sexism not only because of their historic significance but also because they have been dominant in social psycho- logical research. It is important to note that other forms of bias and discrimina- tion are, of course, quite important and are the subject of contemporary social psychological research. In fact, social psychologists today are studying a wider variety of types of stereotypes, prejudices, and discrimination than ever. It is prob- ably no coincidence, for example, that as our population ages and people tend to live longer, more researchers are studying ageism—prejudice and discrimina- tion targeting the elderly (Nelson, 2011; North & Fiske, 2015). Other forms of discrimination that are getting recent attention include those targeting people’s physical disabilities or disfigurements, mental health, political ideology, economic class, being unmarried, or religion (or the lack of religious beliefs) (DePaulo, 2011; Gervais, 2013; Madera & Hebl, 2013; West et al., 2014).

Russ Espinoza and Cynthia Willis-Esqueda (2015), for example, studied biases based on individuals’ socioeconomic status. They gave almost 600 individuals

who were called for jury duty information about a case and asked them to make a verdict and, if they thought the defendant was guilty, rec- ommend a sentence. European American par- ticipants who thought the defendant was guilty were significantly more likely to recommend the death penalty if the defendant was described as having little education or money than if he was highly educated and had a good deal of money.

Although it is much less socially acceptable to display obvious prejudice and discrimination of many kinds today, some forms of these bi- ases appear to be considered more acceptable by many people. Among these are prejudice based on weight and based on sexuality. Although what is considered ideal body types can vary dramati-

cally across times and cultures, it is clear that in some cultures today, jokes, negative attitudes, and insensitive behaviors directed toward those who are perceived to be overweight can be pervasive and hurtful (Brochu & Esses, 2011; Hebl et al., 2009; Ruggs et al., 2015; Schafer & Ferraro, 2011). Lisa Rosenthal and others (2015), for example, found that fifth and sixth grade children who experienced weight-based bullying were more likely to show negative health outcomes two years later, even after statistically controlling for any physical and mental health differences that ex- isted at the beginning of the study.

When former Olympic champion Bruce Jenner announced to the world in 2015 that he was transitioning to a female identity with the name Caitlyn, the response was much more understanding and supportive than could have been expected just a few years before.

A P

Im ag

es /C

hr is

P iz

ze llo

80220_ch05_ptg01_hr_154-207.indd 170 27/01/16 5:19 PM

171 The Nature of the Problem: Persistence and Change

Prejudice and discrimination based on sexuality have gone through some dramatic changes very recently. When the former Olympic champion Bruce Jenner revealed to the public in 2015 that he was transitioning to a female identity, with the name Caitlyn, the support that Cait- lyn received from the general public was enormous (al- though certainly not universal) and probably would have been unthinkable just a few years before. Americans’ at- titudes toward same-sex marriage also have shifted dra- matically in just a few years (see d Figure 5.7).

On the other hand, acts of overt and blatant preju- dice, discrimination, and violence toward members of the LGBT community continue to occur in disturbing numbers, and discriminatory attitudes and even laws remain in many parts of the world. In addition to these overt examples, there are many more ways in which these prejudices and discrimination can operate more subtly. For example, András Tilcsik (2011) sent pairs of resumes in response to almost 1,800 job postings in seven states across the United States. The resumes in each pair were virtually identical except that in one of them, the job candidate indicated having volunteered time for a gay campus organization. Would these (fictitious) job appli- cants get invited to interview for the jobs? Even though the job-relevant skills and experiences were identical, the applicant with the experience in the gay campus organization was about 40% less likely to be invited for an interview than the other applicant.

j Being Stigmatized We are all targets of other people’s stereotypes and prejudices. These may be based on how we look, how we talk, how we dress, where we come from, and so on. None of us is immune from having our work evaluated in a biased way, our motives questioned, or our attempts at making new friends rejected because of stereotypes and prejudices. But for the targets of some stereotypes and prejudices, these concerns are relentless and profound. For them, there seem to be few safe havens. Social psychologists often refer to these targets as stigmatized—individu- als who are targets of negative stereotypes, perceived as deviant, and devalued in society because they are members of a particular social group or because they have a particular characteristic (Major & Crocker, 1993). What are some of the effects of being stigmatized by stereotypes and prejudice?

In Color-Blind, writer Ellis Cose (1997), who is African American, tells a story about how he was treated in a job interview 20 years earlier. He was an award-winning newspaper reporter at the time and was hoping to land a job with a national magazine. The editor he met with was pleasant and gracious, but he said that the magazine didn’t have many black readers. “All the editor saw was a young black guy, and since Esquire was not in need of a young black guy, they were not in need of me. . . . He had been so busy focusing on my race that he was incapable of seeing me or my work” (p. 150). Then, a few years later, and in light of affirmative action policies, Cose was asked if he was interested in a position in a firm as corporate director of equal opportunity. “I was stunned, for the question made no sense. I was an expert neither on personnel nor on equal

stigmatized Being persistently stereotyped, perceived as deviant, and devalued in society because of membership in a particular social group or because of a particular characteristic.

d F i g u r e 5 . 7

Changes in Attitudes Toward Same-Sex Marriage

Opinion polls from 2005 to 2015 have shown that responses to the question, “Do you think marriages between same-sex couples should or should not be recognized by law as valid, with the same rights as traditional marriages?” have become much more favorable in recent years, showing a complete reversal in percentages. Based on Gallup (2015)

0 ‘14‘13‘12‘11‘10‘09‘08‘07‘06‘05 ‘15

% Should be valid % Should not be valid

59 60

37 37

80220_ch05_ptg01_hr_154-207.indd 171 27/01/16 5:19 PM

172 Chapter 5 Stereotypes, Prejudice, and Discrimination

employment law; I was, however, black, which seemed to be the most important qualification” (p. 156).

The targets of stigmatizing stereotypes frequently wonder whether and to what extent others’ impressions of them are distorted through the warped lens of social categorization. On some occasions these suspicions can actually serve a self-protective function. For example, in a study by Jennifer Crocker and her col- leagues (1991), black students who received negative interpersonal feedback from a white student suffered less of a blow to their self-esteem if they could easily attribute the white student’s negative reaction to racism than if they could not. On the other hand, the self-esteem of black students was actually reduced by positive feedback from a white student if they could suspect that this feedback may have been due to their race.

Although attributing negative feedback to discrimination can protect one’s overall self-esteem, it can also make people feel as if they have less personal con- trol over their lives and thus feel worse about themselves, especially when they have reason to think that the discrimination against them could persist over time. Indeed, a recent meta-analysis by Michael Schmitt and others (2014) of hundreds of correlational and experimental studies involving nearly 150,000 participants supports this point, that perceiving that one has been discriminated against and that the discrimination is likely to be experienced frequently is associated with negative effects on one’s self-esteem, depression, anxiety, and life satisfaction. Even simply anticipating interacting with someone who one thinks is prejudiced can trigger cardiovascular stress responses (Sawyer et al., 2012). It is understand- able, therefore, that when people who have a stigmatized status that they can hide during an interaction with someone, such as a history of mental illness, they often prefer to do so, but even the act of hiding it during the interaction can itself be stressful and lead to negative effects and outcomes (Newheiser & Barreto, 2014; Newheiser et al., 2015).

More generally, there has been a surge of research findings in recent years concerning how stigmatized targets are at increased risk for serious and long-term physical and psychological problems, including increased blood pressure, depres- sion, breast cancer rates, diabetes, stroke, respiratory problems, chronic pain, sub- stance abuse, and impaired relationships (e.g., Earnshaw et al., 2015; Lukachko et al., 2014; Murphy et al., 2015; Taylor, 2015).

j Stereotype Threat: A Threat in the Air One of the more tragic effects of stereotyping in contemporary life is its effects on the intellectual performance and identity of its targets. An enormous wave of re- search on this issue was generated when social psychologist Claude Steele began writing about this problem in the 1990s. Steele proposed that in situations where a negative stereotype can apply to certain groups, members of these groups can fear being seen “through the lens of diminishing stereotypes and low expecta- tions” (1999, p. 44). Steele (1997) called this predicament stereotype threat, for it hangs like “a threat in the air” when the individual is in the stereotype-relevant situation. The predicament can be particularly threatening for individuals whose identity and self-esteem are invested in domains where the stereotype is relevant. Steele argued that stereotype threat plays a crucial role in influencing the intel- lectual performance and identity of stereotyped group members. Steele and his colleagues (2002) later broadened the scope of their analysis to include social identity threats more generally. These threats are not necessarily tied to specific stereotypes but instead reflect a more general devaluing of a person’s social group.

stereotype threat The experience of concern about being evaluated based on negative stereotypes about one’s group.

80220_ch05_ptg01_hr_154-207.indd 172 27/01/16 5:19 PM

173 The Nature of the Problem: Persistence and Change

According to Steele’s theory, stereotype threat can hamper achievement in aca- demic domains in two ways. First, reactions to the “threat in the air” can directly interfere with performance—for example, by increasing anxiety and triggering distracting thoughts. Second, if this stereotype threat is chronic in the academic domain, it can cause individuals to disidentify from that domain—to dismiss the domain as no longer relevant to their self-esteem and identity.

To illustrate, imagine a black student and a white student who enter high school equally qualified in academic performance. Imagine that while taking a particularly difficult test at the beginning of the school year, each student strug- gles on the first few problems. Both students may begin to worry about failing, but the black student may have a whole set of additional worries about appearing to confirm a negative stereotype. Even if the black student doesn’t believe the stereotype at all, the threat of being reduced to a stereotype in the eyes of those around her can trigger anxiety and distraction, impairing her performance. And if she experiences this threat in school often—perhaps because she stands out as one of only a few black students in the school or because she is treated by others in a particular way—the situation may become too threatening to her self-esteem. To buffer herself against the threat, she may disidentify with school. If she does this, her academic performance will become less relevant to her identity and self- esteem, and she will therefore work less hard and perform worse.

The Original experiments Steele and others con- ducted a series of experiments in which they manipu- lated factors likely to increase or decrease stereotype threat as students took academic tests. For example, Steele and Joshua Aronson (1995) had black and white students from a highly selective university take a very difficult standardized verbal test. To some participants, it was introduced as a test of intellectual ability; to others, it was introduced as a problem-solving task unrelated to ability. Steele and Aronson reasoned that because of the difficulty of the test, all the students would struggle with it. If the test was said to be related to intellectual ability, however, the black students would feel the threat of a negative stereotype in addition to the stress of struggling with the test. In contrast, if the test was described simply as a research task and not a real test of intelligence, then negative stereotypes would be less applicable and the stereotype threat would be reduced. In that case, black students would be less impaired while taking the test. As shown in d Figure 5.8, the results supported these predictions.

Thus, a seemingly minor change in the setting—a few words about the meaning of a test—had a pow- erful effect on the black students’ performance. In a second study, the researchers used an even more subtle manipulation of stereotype threat: whether or not the students were asked to report their race just before tak- ing the test. Making them think about race for a few seconds just before taking the test impaired the per- formance of black students but had no effect on white students. Think about the implications of such findings

"Test of Intelligence"

"Task Unrelated to Intelligence"

M ea

n ite

m s

so lv

ed (

ad ju

st ed

b y

SA T*

s co

re )

White participants Black participants

*SAT = Scholastic Assessment Test

d F i g u r e 5 . 8

Stereotype Threat and Academic Performance

Black students did worse on a difficult verbal test than white students (even after the students’ scores were adjusted based on their performance on standardized college entrance verbal examinations) if they had been told the test was of intellectual ability (left). In contrast, there was no difference in adjusted scores between black and white students on the same test if they had been told the test was unrelated to ability (right). Source: Based on Steele & Aronson, 1995.

80220_ch05_ptg01_hr_154-207.indd 173 27/01/16 5:19 PM

174 Chapter 5 Stereotypes, Prejudice, and Discrimination

for important real-world contexts. (After years of conducting this research, social psychologists were finally able to convince the College Board and Educational Testing Service in 2012 to agree to discontinue the practice of asking students de- mographic questions immediately before they begin some high-stakes testing such as Advanced Placement exams.)

Because negative stereotypes concerning women’s advanced math skills are prevalent, women may often experience stereotype threat in settings relevant to these skills. Reducing stereotype threat in these settings, therefore, should reduce the underperformance that women tend to exhibit in these areas. To test this idea, Steven Spencer and others (1999) recruited male and female students who were good at math and felt that math was important to their identities. The researchers gave these students a very difficult standardized math test, one on which all of them would perform poorly. Before taking the test, some students were told that the test generally showed no gender differences, thereby implying that the nega- tive stereotype of women’s ability in math was not relevant to this particular test. Other students were told that the test did generally show gender differences. As Steele’s theory predicted, women performed worse than men when they were told that the test typically produced gender differences, but they performed as well as men when they were told that the test typically did not produce gender differences.

The Prevalence and Diversity of Threats Since these original studies, re- search inspired by the theory of stereotype threat grew at a stunningly fast pace. The evidence for underperformance due to stereotype threat is quite strong and broad (Inzlicht & Schmader, 2012; Schmader et al., 2015). It has been found both in the lab- oratory and in real-world settings, including schools and businesses. The examples of these threats run far and wide. For instance, many white athletes feel stereotype threat whenever they step onto a court or playing field where they constitute the minority. Will the white athlete feel the added weight of this threat while struggling against the other athletes in a game? To address this question, Jeff Stone and others (1999) had black and white students play miniature golf. When the experimenters characterized the game as diagnostic of “natural athletic ability,” the white students did worse. But when they characterized it as diagnostic of “sports intelligence,” the black students did worse.

It should be pointed out that a person can be affected by stereotype threat even if he or she does not believe in the negative stereotype. Just knowing about the stereotype seems to be enough, particularly if the individual identifies strongly with the targeted group and cares about performing well. This last point is particu- larly poignant—the performance of people who have had success at something and who care the most may be most affected by stereotype threat effects. (An in- teresting side note: The first experiment to show that students who most identify with and care about success in school may be the most vulnerable to stereotype threat effects was conducted in an inner-city Los Angeles high school by a Stan- ford University undergraduate student named Mikel Jollet, working with Claude Steele. Jollet soon after formed and became lead singer and songwriter of the very successful alt-rock group, The Airborne Toxic Event.)

Here is a small sample of groups whose performance in various domains was hurt by stereotype threat, as demonstrated in experiments around the world:

d Low-socioeconomic-status students in France and the United States on a verbal test when the test was said to be diagnostic of intellectual ability (Croizet & Claire, 1998)

d European American men on a math test when compared with Asians (Aronson et al., 1999)

An African American student is likely to perform worse on an athletic task if the task is described as one reflecting sports intelligence than if it is described as reflecting natural athletic ability.

True

80220_ch05_ptg01_hr_154-207.indd 174 27/01/16 5:19 PM

175 The Nature of the Problem: Persistence and Change

d Older adults on memory or other cognitive tests when presented, even subliminally (that is, with- out their conscious awareness), with negative stereotypes about the elderly (Krendl et al., 2015; Lamont et al., 2015)

d First-generation students taking a biology class (Harackiewicz et al., 2014)

d Women on a math test in a co-ed rather than an all-female setting (Ben-Zeev et al., 2005; Sekaqua- ptewa & Thompson, 2003)

d Women playing chess on the computer when they were told that their opponent was male (Maass et al., 2008)

d White participants taking an IAT when they thought the test was diagnostic of racism (Frantz et al., 2004)

d Individuals with a history of mental illness on a test of reasoning ability when asked about their illness before taking the test (Quinn et al., 2004)

d Women performing soccer drills when the task was described as one in which men tend to outperform women (Heidrich & Chiviacowsky, 2015)

d Student athletes primed to think about their identity as athletes before taking a difficult math test (Yopyk & Prentice, 2005)

d Overweight individuals primed to think about weight-related stereotypes (Major et al., 2014; Seacat & Mickelson, 2009)

d Women driving after being reminded of demeaning stereotypes about female driv- ers (causing women in a driving simulator to crash into jaywalkers!) (Yeung & von Hippel, 2008)

It is important to note that in each of these cases, the underperformance is significantly or completely eliminated when the stereotype threat is reduced. We will discuss how this occurs in the final section of the chapter (see, for example, Table 5.4).

Causes of Stereotype Threat effects Stereotype threat exerts its effects in multiple ways (Forbes & Leitner, 2014; Rydell et al., 2014; Schmader et al., 2015). Stereotype threat has been shown to do each of the following to people: trigger physi- ological arousal and stress; drain cognitive resources; cause a loss of focus to the task at hand because of attempts to suppress thoughts about the relevant stereotype; impair working memory; activate negative thoughts, worry, feelings of dejection, and concerns about trying to avoid failure rather than trying to achieve success; elicit neural activity biased toward negative, stereotype-confirming feedback. Think about trying to do your best on a difficult test that is important to you while all of these things are happening to you—you’d get some sense of how stereotype threat can undermine people’s performance and ambitions.

Even though stereotype threat effects are widespread, the growing body of research on this subject also gives us reason to hope. Social psychologists have been uncovering ways that people can be better protected against these threats. We will focus on these promising ways in the final section of this chapter. Before

Three scholars who have done research on stereotype threat reunite backstage after an Airborne Toxic Event concert. Can you guess which one became the rock star?

Ca rin

P er

ill ou

80220_ch05_ptg01_hr_154-207.indd 175 27/01/16 5:19 PM

176 Chapter 5 Stereotypes, Prejudice, and Discrimination

we get to the solutions to stereotype threat and the other problems we have been discussing thus far in this chapter, we turn next to the social psychological causes at the root of these problems.

Causes of the Problem: Intergroup, Motivational, Cognitive, and Cultural Factors

One of the reasons that stereotypes, prejudice, and discrimination persist is because they are caused by multiple factors. There are many sources fueling these problems, and they operate both independently and in tandem. Some stem from the ways that humans cognitively process and remember information. Others can be traced to motivations and goals that drive us to see or react to our social worlds in particular ways. Still others concern how groups of people are represented or valued in one’s culture. In this section we turn to look at some of the most impor- tant of these causes.

j Social Categories and intergroup Conflict At the root of stereotyping, prejudice, and discrimination is the fact that we divide our social world into groups. As perceivers, we routinely sort each other into groups on the basis of gender, race, age, and other common attributes in a process called social categorization. In some ways, social categorization is natural and adaptive. It allows us to form impressions quickly and use experience to guide new interactions. With so many things to pay attention to in our social worlds, we can save time and effort by using people’s group memberships to make inferences about them. The time and energy saved through social categorization does come at a cost, however. Categorizing people leads us to overestimate the differences between groups and to underestimate the differences within groups (Krueger & DiDonato, 2008; Wyer et al., 2002).

Indeed, even basic perception is affected by categorization. For example, studies have shown that people see racially ambiguous faces as darker, and they trigger more negative implicit association, if the faces are labeled racially black than white (Levin & Banaji, 2006; Willadsen-Jensen & Ito, 2015). In addi- tion, prejudice can heighten this kind of bias. Markus Kemmelmeier and Lysette Chavez (2014) found that white Americans perceived Barack Obama’s face as darker if they were relatively high in racism. Even young children show these kinds of effects: Yarrow Dunham and others (2013) found that children were more likely to categorize a racially ambiguous face as black than white if it is expressing anger.

Each of us is a member of multiple social categories, but some categorizations— particularly involving race, gender, and age—are more likely to quickly dominate our perceptions than others (Amodio et al., 2014; Kaul et al., 2014; Ito, 2011). The distinctions between some of these social categories may be seen as more rigid, even more biologically rooted, than they actually are. Many people assume, for ex- ample, that there is a clear genetic basis for classifying people by race. The fact is, however, that numerous biologists, anthropologists, and psychologists note that there is more genetic variation within races than between them and emphasize that race is more of a social conception than a genetic reality (Marks, 2011; Markus, 2008; Plaks et al., 2012). Indeed, how societies make distinctions between races

“In the 2010 census of the popu- lation of the United States, more than 21.7 million Americans did not believe the government’s traditional categories of race fit them.”

—Yen (2012)

social categorization The classification of persons into groups on the basis of common attributes.

80220_ch05_ptg01_hr_154-207.indd 176 27/01/16 5:19 PM

177 Causes of the Problem: Intergroup, Motivational, Cognitive, and Cultural Factors

can change dramatically as a function of historical contexts. For instance, it was fairly common for Americans in the early part of the twentieth century to consider Irish Americans as a racial group distinct from whites, but today such thinking is quite rare. As people today increasingly identify themselves in multiracial ways, or in ways that defy the traditional binary distinction between men and women, a greater recognition of the role of factors beyond biology in social categorization becomes all the more relevant. Along these lines, Diana Sanchez and others (2015) found that after a brief interaction with a racially ambiguous other student, white students became less likely to see race as a fixed, biological entity, an effect that endured when the students’ attitudes were assessed again two weeks later.

Whether individuals think of various social categories as fixed and biologically rooted or not can be important (Andreychik & Gill, 2015; Cimpian & Salomon, 2014; D. Sanchez & Garcia, 2009). For example, Melissa Williams and Jennifer Eberhardt (2008) found that people who tend to think of race as a stable, biologically deter- mined entity are less likely to interact with racial outgroup members and are more likely to accept racial inequalities than are people who see race as more socially de- termined. Other research has found that biracial individuals are more vulnerable to some effects of stereotypes if they think of race as stable and biological (D. Sanchez & Garcia, 2009; Shih et al., 2007).

Just as race is a blurrier category than many people realize, so too are nation- alities. A series of studies by Thierry Devos and others demonstrated how various ethnic minority groups such as Latino Americans, Asian Americans, and African Americans are not seen as truly American, but some situational factors—such as presenting individuals with examples of positive stereotypic traits of one of these groups—can reduce this tendency (Devos et al., 2010; Huynh et al., 2011; Rydell et al., 2010). As immigration battles intensify throughout much of the world, a variety of social, historical, economic, and political factors all play a role in who is categorized as “foreign.”

Diana Sanchez and Julie Garcia (2012) reported a number of ways in which racial categorization is also affected by people’s social and economic status. For example, perceivers are more likely to categorize others as racially black if they are of lower socioeconomic status, if they are incarcerated, or if they are unemployed. These same factors may have similar effects on how people categorize themselves racially.

ingroups Versus Outgroups Although categorizing humans is much like cat- egorizing objects, there is a key difference. When it comes to social categorization, perceivers themselves are members or nonmembers of the categories they use. Groups that we identify with—our country, religion, political party, even our hometown sports team—are called ingroups, whereas groups other than our own are called outgroups. We see people in fundamentally different ways if we consider them to be part of our ingroup or as part of an outgroup.

One consequence is that we exaggerate the differences between our ingroup and other outgroups, and this exaggeration of differences helps to form and reinforce stereotypes. Another consequence is a phenomenon known as the outgroup homogeneity effect, whereby perceivers assume that there is a greater similarity among members of outgroups than among members of one’s own group. In other words, there may be many and subtle differences among “us,” but “they” are all alike (Linville & Jones, 1980). It is easy to think of real-life examples. People from China, Korea, Taiwan, and Japan see them- selves as quite distinct from one another, of course, but to many Westerners they are seen simply as Asian. English majors see themselves as dissimilar to

“The percentage of babies clas- sified by the U.S. Census as ‘multiracial’ has risen from 1% in 1970 to 10% in 2013.”

—Pew Research Center, 2015b

ingroups Groups with which an individual feels a sense of membership, belonging, and identity.

outgroups Groups with which an individual does not feel a sense of membership, belonging, or identity.

outgroup homogeneity effect The tendency to assume that there is greater similarity among members of outgroups than among members of ingroups.

80220_ch05_ptg01_hr_154-207.indd 177 27/01/16 5:19 PM

178 Chapter 5 Stereotypes, Prejudice, and Discrimination

history majors, but science majors often lump them together as “humanities types.” Californians proclaim their tremendous cultural, ethnic, and economic diversity, whereas outsiders talk of the “typical Californian.” To people outside the group, outgroup members even look alike: People are less accurate in distinguishing and recognizing the faces of members of racial outgroups than of ingroups (Horry et al., 2015; McDonnell et al., 2014; Meissner et al., 2013).

Why do people tend to perceive outgroups as homogeneous? One reason is that people tend to have less personal contact and familiarity with individual members of outgroups. Indeed, the more familiar people are with an outgroup, the less likely they are to perceive it as homogeneous. Second, people often do not en- counter a representative sample of outgroup members. A student from one school who encounters students from a rival school only when they cruise into town for a Saturday football game, scream- ing at the top of their lungs, sees only the most avid rival fans act- ing in their most rowdy, competitive ways—hardly a diverse lot.

Lack of familiarity and lack of diversity of experiences with outgroup members are two reasons why “they all look alike,” but there’s more to the story than that. Research using brain imaging or cognitive methods has found that as soon as we categorize an unfamiliar person as a member of our ingroup or an outgroup, we immediately process information about them differently at even the most basic levels. For example, student participants in experi-

ments by Kurt Hugenberg and Olivier Corneille (2009) were exposed to unfamil- iar faces of people who were the same race as the participants. These faces were categorized as ingroup members (from the same university as the participants) or outgroup members (from a rival university). The students processed faces more holistically (that is, they integrated the features of the faces into a global repre- sentation of the overall face) when they had been categorized as being from their ingroup than they did when they had been categorized as members of the outgroup. Jay Van Bavel, William Cunningham, and their colleagues (2008; 2011; Van Bavel et al., 2014) have found related results in a series of studies, revealing greater activa- tion in particular areas of perceivers’ brains, such as the fusiform face area and the orbitofrontal cortex, upon exposure to unfamiliar faces labeled as ingroup members than outgroup members (see d Figure 5.9). The effects of ingroup/outgroup label- ing can even override the effects of racial biases on these measures.

Dehumanizing Outgroups Perceivers may not only process outgroup faces more superficially but also sometimes process them more like objects and lower- order animals than like fellow humans. Dehumanization has played a role in atrocities throughout history, such as in the Nazi propaganda that characterized the Jews in Germany as disease-spreading rats and blacks as half-apes. The con- tinued presence of some of this kind of imagery in contemporary life is chilling, as in the examples discussed in the introduction of this chapter of black athletes in numerous countries being taunted with monkey chants. When former rock star Ted Nugent called Barack Obama “a chimpanzee” and “sub-human mon- grel” in a 2014 interview (Haraldsson, 2014), this marked only one of countless examples of white Americans using such imagery to describe Obama.

These blatant examples are shockingly plentiful deep into the second decade of the twenty-first century, but they are more the exception than the rule. What

Whether people are likely to immediately categorize this person by her race, gender, or occupation depends on a combination of cognitive, cultural, and motivational factors.

Kr is

T im

ke n/

Bl en

d Im

ag es

/C or

bi s

80220_ch05_ptg01_hr_154-207.indd 178 27/01/16 5:19 PM

179 Causes of the Problem: Intergroup, Motivational, Cognitive, and Cultural Factors

d F i g u r e 5 . 9

The Neuroscience of ingroups and Outgroups

Highlighted in color in these anatomical images of the brain are some key brain regions associated with making ingroup/outgroup distinctions and related intergroup evaluations. “OFC” is the orbitofrontal cortex, and “mPFC” is the medial prefrontal cortex. Greater activity in the OFC, for example, has been associated with stronger preference for ingroup faces. Based on Cikara & Van Bavel, 2014

OFC

Ventral striatum fusiform gyn

Anterior cingulate

mPFC

may be much more common, however, is how people often implicitly dehumanize members of particular outgroups. Just as measures like the IAT are used to capture implicit racism, similar measures have been used to assess automatic, implicit dehumanization. Using these tech- niques, researchers around the world have found strong evidence for automatic dehumaniza- tion of various outgroups (Haslam, 2015; Kteily et al., 2015). Studies have shown, for example, that many individuals (even as young as 6 years old) automatically but nonconsciously associate black men, people of low socioeconomic status, and people from various countries with animals such as apes, rats, and dogs (Costello & Hodson, 2014; Goff, Eberhardt et al., 2008; Loughnan et al., 2014; Wilde et al., 2014). Indeed, this ten- dency may be so deeply rooted that we can see it play out in the brain. Lasana Harris and others have found that when people perceive or think about members of particular stigmatized out- groups, their patterns of brain activity suggest that they are responding to these outgroup mem- bers more as they would to objects than to fully human individuals who are capable of their own agency and mental states (Harris & Fiske, 2011; Lee & Harris, 2014).

These implicit processes of dehumanization may be subtle, but the con- sequences can be profound. Phillip Goff and his colleagues (2014) found that police officers who more strongly associated black men with apes were more likely to use force against black children. Laurie Rudman and Kris Mescher (2012) found that men who automatically associated women with animals or objects showed stronger inclination to sexually harass or rape women. Luca Andrighetto and others (2014) found that Italian students who tended to as- sociate Japanese and Haitians in dehumanizing ways (as machines or animals, respectively) were less willing than other students to help them after natural disasters in Japan and Haiti.

Fundamental Motives Between groups The roots of dividing into ingroups and outgroups run quite deep in our evolutionary history, as early humans’ sur- vival depended on forming relatively small groups of similar others. A fundamental motive to protect one’s ingroup and be suspicious of outgroups is therefore likely to have evolved. Consistent with this idea are the results of experiments that demonstrate that when people’s basic motivations of self-protection are activated—such as in response to a threatening situation, economic scarcity, a scary movie, concerns about the flu, or even being in a completely dark room— people are more prone to exhibit prejudice toward outgroups or to be especially hesitant to see possible outgroup members as part of one’s ingroup (Makhanova et al., 2015; Maner et al., 2012; Schaller & Neuberg, 2012).

The flip side to the distrust of outgroups is the positive feelings we have toward being part of an ingroup. The feeling of connection and solidarity we have with our own groups enhances our sense of control and meaning, and it is associated with numerous psychological as well as physical health

80220_ch05_ptg01_hr_154-207.indd 179 27/01/16 5:19 PM

180 Chapter 5 Stereotypes, Prejudice, and Discrimination

benefits (Greenaway et al., 2015; Paéz et al., 2015). William Swann and Michael Buhrmester (2015) use the term identity fusion to describe the sense of “oneness” that people may feel with a group. This feeling can motivate helpful behavior toward the group, even at the risk of personal sacrifice. In general, when we are feeling threatened or uncertain, we become especially motivated to reaffirm our identification and closeness with an ingroup, which can make us feel more safe and secure (Cikara & Van Bavel, 2014; Hogg, 2014; Knapton et al., 2015).

From time to time our fundamental motivation for self-protection and preservation runs smack into the ultimate obstacle: thoughts about death and mortality. According to Terror Management Theory, which was discussed in Chapter 3, people cope with the fear of their own death by constructing world- views that help preserve their self-esteem and important values. According to this perspective, favoring ingroups over outgroups is one important way that people preserve their cultural worldviews and, by doing so, try to attain a kind of immortality. This theory has been supported by numerous studies that demonstrate that when individuals are made to think about mortality—such as by presenting them with images of cemeteries or making them think about decomposing bodies—they become more likely to exhibit various ingroup bi- ases, including through stereotypes and prejudice toward a variety of outgroups (Greenberg & Arndt, 2012; Greenberg et al., 2009). Male students in an experi- ment by Russell Webster and Donald Saucier (2011), for example, expressed significantly more discrimination toward gay men if they had just written about what they thought would happen to them when they die than if they had writ- ten about the pain of going to the dentist.

Motives Concerning intergroup Dominance and Status Some people are especially motivated to preserve inequities between groups of people in soci- ety. For example, people with a social dominance orientation have a desire to see their ingroups as dominant over other groups and tend to support cultural values that contribute to the oppression of other groups. Individuals with this orientation tend to endorse sentiments such as “If certain groups stayed in their place, we would have fewer problems” and to disagree with statements such as “Group equality should be our ideal.” Research in numerous coun- tries throughout the world has found that ingroup identification and outgroup derogation and dehumanization can be especially strong among people with a social dominance orientation (Kteily et al., 2015; Levin et al., 2013; Prati et al., 2015; Pratto et al., 2013).

Social dominance orientations promote self-interest. But some ideolo- gies support a social structure that may actually oppose one’s self-interest, depending on the status of one’s groups. John Jost and his colleagues (Jost et al., 2015; van der Toorn et al., 2015) have focused on what they call system justification theory, which proposes that people are motivated (at least in part) to defend and justify the existing social, political, and economic conditions. Sys- tem-justifying beliefs protect the status quo. Groups with power, of course, may promote the status quo to preserve their own advantaged position. But although some disadvantaged groups might be able to improve their circumstances if they were to challenge an economic or political system, members of disadvantaged groups with a system justification orientation think that the system is fair and just, and they may admire and even show outgroup favoritism to outgroups that thrive in this system.

Being reminded of one’s own mortality makes people put things into greater perspective, thereby tending to reduce ingroup– outgroup distinctions and hostilities.

FAlSe

social dominance orientation A desire to see one’s ingroup as dominant over other groups and a willingness to adopt cultural values that facilitate oppression over other groups.

system justification theory A theory that proposes that people are motivated (at least in part) to defend and justify the existing social, political, and economic conditions.

80220_ch05_ptg01_hr_154-207.indd 180 27/01/16 5:19 PM

181 Causes of the Problem: Intergroup, Motivational, Cognitive, and Cultural Factors

Stereotype Content Model According to the stereotype content model (Kervyn et al., 2015; North & Fiske, 2014), many group stereotypes vary along two dimensions: warmth and competence. Groups may be considered high on both dimensions, low on both, or high on one dimension but low on the other. For example, the elderly may be stereotyped as high on warmth but low on competence.

The stereotype content model proposes that stereotypes about the com- petence of a group are influenced by the relative status of that group in society—higher relative status is associated with higher competence. Stereo- types about the warmth of a group are influenced by perceived competition with the group—greater perceived competition is associated with lower warmth. For example, groups that are of low status but that remain compliant and do not try to upset the status quo are likely to be stereotyped as low in competence but high in warmth. On the other hand, a wave of immigrants who enter a coun- try with low status but are seen as competing for jobs and resources may be perceived as low in both competence and warmth. For groups that are seen as high on one dimension but low on the other, there may be a perceived trade-off between competence and warmth. A woman climbing up the corporate lad- der by demonstrating strong competence, for example, may be seen as much less warm. If she tries to demonstrate warmth, however, she may be seen as less competent.

robbers Cave: A Field Study in intergroup Conflict Robbers Cave State Park in Oklahoma was the unlikely setting for one of the most classic field experi- ments in social psychology. In the summer of 1954, a small group of 11-year-old boys—all white middle-class youngsters, all strangers to one another—arrived at a 200-acre camp located in a densely wooded area of the park. The boys spent the first week or so hiking, swimming, boating, and camping out. After a while, they gave themselves a group name and printed it on their caps and T-shirts. At first, the boys thought they were the only ones at the camp. Soon, however, they discovered that there was a second group and that tournaments had been arranged between the two groups.

What these boys didn’t know was that they were participants in an elabo- rate study conducted by Muzafer Sherif and his colleagues (1961). Parents had given permission for their sons to take part in an experiment for a study of com- petitiveness and cooperation. The two groups were brought in separately, and only after each had formed its own culture was the other’s presence revealed. When the “Rattlers” and the “Eagles” finally met, they did so under tense circumstances, competing against each other in football, a treasure hunt, tug of war, and other events. For each event, the winning team was awarded points; the tournament winner was promised a trophy, medals, and other prizes. Almost overnight, the groups turned into hostile antagonists, and their rivalry escalated into a full-scale war. Group flags were burned, cabins were ransacked, and a food fight that resembled a riot exploded in the mess hall. Keep in mind that the participants in this study were well-adjusted boys. Yet as Sherif (1966) noted, a naive observer would have thought the boys were “wicked, disturbed, and vicious” (p. 85).

Creating a monster through competition was easy. Restoring the peace, how- ever, was not. First the experimenters tried saying nice things to the Rattlers about the Eagles and vice versa, but the propaganda campaign did not work. Then the two groups were brought together under noncompetitive circumstances,

stereotype content model A model proposing that the relative status and competition between groups influence group stereotypes along the dimensions of competence and warmth.

80220_ch05_ptg01_hr_154-207.indd 181 27/01/16 5:19 PM

182 Chapter 5 Stereotypes, Prejudice, and Discrimination

but that didn’t help either. What did eventually work was the introduction of superordinate goals, mutual goals that could be achieved only through coop- eration between the groups. For example, the experimenters arranged for the camp truck to break down, and both groups were needed to pull it up a steep hill. This strategy worked like a charm. By the end of camp, the two groups were so friendly that they insisted on traveling home on the same bus. In just three weeks, the Rattlers and Eagles experienced the kinds of changes that often take generations to unfold: They formed close-knit groups, went to war, and made peace.

The events of Robbers Cave mimicked the kinds of conflict that plague people all over the world. The simplest explanation for this conflict is competition. As- sign strangers to groups, throw the groups into contention, stir the pot, and soon there’s conflict. Similarly, the intergroup benefits of reducing the focus on compe- tition by activating superordinate goals are also evident around the world. When the disparate and contentious groups of a country or university find themselves pulling together in conflict or competition with another country or college, for ex- ample, the prejudices and discrimination that kept the groups apart often are cast aside in the pursuit of a common goal.

realistic Conflict Theory The view that direct competition for valuable but lim- ited resources breeds hostility between groups is called realistic conflict theory (Levine & Campbell, 1972). As a simple matter of economics, one group may fare better in the struggle for land, jobs, or power than another group. The losing group becomes frustrated and resentful, the winning group feels threatened and protective—and, before long, conflict heats to a rapid boil. It is likely that a good deal of prejudice in the world is driven by the realities of competi- tion (Duckitt & Mphuthing, 1998; Filindra & Pearson-Merkowitz, 2013; Stephan et al., 2005; Zárate et al., 2004). For example, Marcel Coenders and others (2008) found that support for discrimination against ethnic minority groups tended to increase in the Netherlands when the unemployment level had re- cently risen. David Butz and Kuma Yogeeswaran (2011) found that students in the United States indicated more prejudice against Asian Americans if they had just read information about serious economic problems and growing competi- tion for scarce resources.

But there is much more to prejudice than real competition. “Realistic” com- petition for resources may in fact be imagined—a perception in the mind of an individual who is not engaged in any real conflict. In addition, people may become resentful of other groups not because of their conviction that their own security or resources are threatened by these groups but because of their sense of relative deprivation, the belief that they fare poorly compared with others. What matters to the proverbial Smiths is not the size of their house per se but whether it is larger than the Jones’s house next door (Moscatelli et al., 2014; H. J. Smith & Pettigrew, 2015).

j Social identity Theory People all over the world believe that their own nation, culture, language, and religion are better and more deserving than others. Part of the rea- son for that is even more basic than real or perceived competition for finite resources. Rather, it stems from something more subtle and psychological.

superordinate goal A shared goal that can be achieved only through cooperation among individuals or groups.

realistic conflict theory The theory that hostility between groups is caused by direct competition for limited resources.

relative deprivation Feelings of discontent aroused by the belief that one fares poorly compared with others.

80220_ch05_ptg01_hr_154-207.indd 182 27/01/16 5:19 PM

183 Causes of the Problem: Intergroup, Motivational, Cognitive, and Cultural Factors

American fans bask in the glory of their team’s success at a World Cup soccer game.

A classic study of high school boys in Bristol, England, conducted by Henri Tajfel and his colleagues (1971) begins to reveal this point. The boys in this study were shown a series of dotted slides, and their task was to estimate the number of dots on each. The slides were presented in rapid-fire succession so the dots could not be counted. Later, the experimenter told the participants that some people are chronic “overestimators” and that others are “underestimators.” As part of a sec- ond, entirely separate task, participants were divided into two groups—one that was said to consist of overestimators and the other of underestimators. (In fact, they were divided randomly.) Partici- pants were then told to allocate points to other participants that could be cashed in for money.

This procedure was designed to create minimal groups in which people are categorized on the basis of trivial, minimally important similarities. Tajfel’s overestimators and underestimators were not long-term rivals, did not have a history of antagonism, were not frustrated, did not compete for a limited resource, and were not even acquainted with each other. Still, participants con- sistently allocated more points to members of their own group than to members of the other group. This pattern of discrimination, called ingroup favoritism, has been found repeatedly in studies in many countries and using a variety of different measures (Capozza & Brown, 2000; Pinter & Greenwald, 2011; Scheepers et al., 2006).

To explain ingroup favoritism, Tajfel (1982) and John Turner (1987) pro- posed social identity theory. According to this theory, which is illustrated in d Figure 5.10, each of us strives to enhance our self-esteem, which has two components: (1) a personal identity and (2) various collective or social identities that are based on the groups to which we belong. In other words, people can boost their self-esteem through their own personal achievements or through affiliation with successful groups. What’s nice about the need for social identity is that it leads us to derive pride from our connections with oth- ers even if we don’t receive any direct benefits from these others. What’s sad, however, is that we often feel the need to belittle “them” in order to feel secure about “us.” Religious fervor, racial and ethnic conceit, and aggressive national- ism may all fulfill this more negative side of our social identity. Even gossiping can play this role; Jennifer Bosson and others (2006; Weaver & Bosson, 2011) found that when people shared negative attitudes about a third party, they felt closer to each other.

Shadenfreude is an intimidating-looking word for a familiar feeling—the ex- perience of pleasure at other people’s misfortunes, particularly for celebrities or others we don’t feel empathy for. Mina Cikara (2015; Cikara et al., 2014) has found that people who identify strongly with their social groups frequently experience this pleasure toward outgroup misfortunes, along with a lack of empathy.

Two basic predictions arose from social identity theory: (1) Threats to one’s self-esteem heighten the need for ingroup favoritism, and (2) expressions of

ingroup favoritism The tendency to discriminate in favor of ingroups over outgroups.

social identity theory The theory that people favor ingroups over outgroups in order to enhance their self-esteem.

Jo e

Ra ed

le /G

et ty

Im ag

80220_ch05_ptg01_hr_154-207.indd 183 27/01/16 5:19 PM

184 Chapter 5 Stereotypes, Prejudice, and Discrimination

ingroup favoritism enhance one’s self-esteem. Research generally supports these predictions (Fiske & Tablante, 2015; Mackie & Smith, 2015).

Steven Fein and Steven Spencer (1997) proposed that threats to one’s self- esteem can lead individuals to use available negative stereotypes to derogate mem- bers of stereotyped groups, and that by derogating others they can feel better about themselves. In one study, for example, Fein and Spencer gave participants posi- tive or negative feedback about their performance on a test of social and verbal skills—feedback that temporarily bolstered or threatened their self-esteem. These participants then took part in what was supposedly a second experiment in which they evaluated a job applicant. All participants received a photograph of a young woman, her résumé, and a videotape of a job interview. Half the participants were given information that suggested that the woman (named Julie Goldberg) was Jew- ish. The other half was given information that suggested that the woman (named Maria D’Agostino) was not Jewish. On the campus where the study was held, there was a popular negative stereotype of the “Jewish American Princess” that often targeted upper-middle-class Jewish women from the New York area.

As predicted, there were two important results (see d Figure 5.11). First, among participants whose self-esteem had been lowered by negative feedback, they rated the woman more negatively if she seemed to be Jewish than if she did not, even though the videotaped job interview and credentials of the two women were the same. Second, participants who had received negative feedback and were given an opportunity to belittle the Jewish woman later exhibited a post-experiment increase in self-esteem—the more negatively they evaluated the Jewish woman, the better

d F i g u r e 5 . 1 0

Social identity Theory

According to social identity theory people strive to enhance self-esteem, which has two components: a personal identity and various social identities that derive from the groups to which we belong. Thus, people may boost their self-esteem by viewing their ingroups more favorably than outgroups. © Cengage Learning

Favoritism toward ingroup and derogation of outgroups

Self-esteem

Personal achievements

Group achievements

Social identities

Personal identity

Need for self-esteem

80220_ch05_ptg01_hr_154-207.indd 184 27/01/16 5:19 PM

185 Causes of the Problem: Intergroup, Motivational, Cognitive, and Cultural Factors

these participants felt about themselves. In sum, the results of this experiment sug- gest that a blow to one’s self-image evokes prejudice and the expression of prejudice helps restore self-image.

Intergroup discrimination is achieved not only through negative reactions and behaviors toward outgroups but also through being especially favorable and help- ful toward one’s ingroups. In their aptly titled paper, “With Malice Toward None and Charity For Some,” Anthony Greenwald and Thomas Pettigrew (2014) propose that this is the more common type of intergroup bias, at least in the United States today. Malice toward outgroups tends to be frowned upon, whereas providing important advantages to one’s ingroups may be more subtle and acceptable.

j Culture and Social identity Individuals’ social identities are clearly important to people across cultures. Collectivists are more likely than individualists to value their connectedness and interdependence with the people and groups around them, and their personal identities are tied closely with their social identities. Collectivists do show some biases favoring their ingroups—indeed, being oriented strongly toward one’s in- group can be highly valued in their cultures—and may draw sharper distinctions

d F i g u r e 5 . 1 1

Self-esteem and Prejudice

Participants in a study by Fein and Spencer received positive or negative feedback and then evaluated a female job applicant who was believed to be either Jewish or not Jewish. This study had two key results: (1) Participants whose self-esteem had been lowered by negative feedback evaluated the woman more negatively if they thought she was Jewish than if they thought she was not (left); and (2) negative-feedback participants given the opportunity to belittle the Jewish woman showed a post-experiment increase in self-esteem (right). Fein and Spencer, 1997.

Positive

Non-Jewish Jewish

Ra tin

g of

ta rg

Negative

Type of Feedback

In cr

ea se

in s

el f-

es te

Positive Negative

Type of Feedback

80220_ch05_ptg01_hr_154-207.indd 185 27/01/16 5:19 PM

186 Chapter 5 Stereotypes, Prejudice, and Discrimination

Being part of a small, close-knit group can be an important, rewarding part of one’s personal identity.

between ingroup and outgroup members than individualists do (Chen et al., 2002; Gudykunst & Bond, 1997; Owe et al., 2013; Ruffle & Sosis, 2006). Along these lines, in a study of nearly 45,000 individuals from 36 countries André van Hoorn (2015) found that collectivists tend to have a narrower circle of people they trust than do individualists.

However, a number of researchers have found that people from collectivist cultures are less likely to enhance their ingroups in order to boost their own self-esteem (Heine, 2005; Lehman et al., 2004; Snibbe et al., 2003; Yuki, 2003). In addition, people from the collectiv- ist cultures of East Asia tend to have higher tolerance for what Westerners would consider contradictions (such as that something can be both good and bad at the same time), and this may explain why East Asianers are more likely to see their ingroups as having both positive and negative qualities compared to Westerners, who tend to emphasize the positive aspects of

their ingroups much more exclusively (Spencer-Rogers et al., 2012).

j Culture and Socialization The list of familiar stereotypes is quite long. Athletes are dumb, math majors are geeks, Americans are loud, Italians are emotional, Californians are laid back, white men can’t jump, car salesmen can’t be trusted. And on and on it goes. Dividing people into social categories, including ingroups and outgroups, cer- tainly is a key factor in the formation of stereotypes and prejudices. But with so many well-known stereotypes and prejudices, many of which are shared around the world, it is clear that we are somehow taught these stereotypes from our culture. We turn now to examine those processes.

Socialization refers to the processes by which people learn the norms, rules, and information of a culture or group. We learn a tremendous amount of informa- tion (often without even realizing it) by absorbing what we see around us in our culture, groups, and families. These lessons include what various stereotypes are, how valued or devalued various groups are, and which prejudices are acceptable to have.

Consider the story of something that happened to one of the authors of this book. When he was about 8 years old, his two best friends one day turned on him and derisively called him a “Jew ball.” They had never thought of him as differ- ent from them or categorized him as Jewish before, and yet on this day, suddenly Jewishness was relevant—and negative to them. But why then, and how did they come up with “Jew ball”?! Only much later did it become clear that they had mis- heard their father say “Jew boy.” Trying to model their father’s values, they used an approximation of this expression against their friend, and thereafter they saw him in a different way. The biased lens through which the father saw people was passed down to the next generation.

Although it certainly isn’t always the case, the stereotypes and prejudices of a parent can shape the stereotypes and prejudices of a child, often in implicit

A P

Im ag

es /M

ar k

G ra

ha m

80220_ch05_ptg01_hr_154-207.indd 186 27/01/16 5:19 PM

187 Causes of the Problem: Intergroup, Motivational, Cognitive, and Cultural Factors

ways (Castelli et al., 2009). More generally, and more pervasively, the stereo- types and prejudices exhibited by peers, the popular media, and one’s culture are part of the air each of us breathes as we develop, and these influences can be profound. These effects typically are gradual, as they develop over the course of a lifetime of exposure. But they can also be immediate, as demonstrated in a recent set of studies by Greg Willard and others (2015). White students who were exposed to a fellow white student’s negative nonverbal behavior to- ward a black student became more likely to exhibit a variety of negative ra- cial attitudes and behaviors than if they saw the same behavior targeting a white student or if they saw positive nonverbal behavior toward a black student. Willard and his colleagues characterized this effect as a contagion of racial bias, as if the students “caught” the disease from exposure to the subtly prejudiced behavior of a fellow student.

To narrow our discussion we will focus in this section on gender stereotypes and sexism, but it is important to recognize that these processes are relevant to all kinds and targets of stereotypes, prejudice, and discrimination.

gender Stereotypes: Blue for Boys, Pink for girls Our traditional story be- gins with what are often the first words uttered when a baby is born: “It’s a boy!” or “It’s a girl!” In many hospitals, the newborn boy is immediately given a blue hat and the newborn girl a pink hat. The infant receives a gender-appropriate name and is showered with gender-appropriate gifts. Over the next few years, the typical boy is sup- plied with toy trucks, baseballs, pretend tools, toy guns, and chemistry sets; the typi- cal girl is furnished with dolls, stuffed animals, pretend makeup kits, kitchen and tea sets, and dress-up clothes. As they enter school, many expect the boy to earn money by mowing lawns and to enjoy violent superhero movies, while they expect the girl to earn money by babysitting and to enjoy sweet stories about friendship or love.

These distinctions persist in college, as more male students major in econom- ics and the sciences and more female students major in the arts, languages, and humanities. In the workforce, more men become doctors, construction workers, auto mechanics, airplane pilots, investment bankers, and engineers. In contrast, more women become secretaries, schoolteachers, nurses, flight attendants, bank tellers, and housewives. Back on the home front, the life cycle begins again when a couple has their first baby and hear the words, “It’s a girl!” or “It’s a boy!”

The Blues and the Pinks. Even a very quick look at a toy store illustrates dramatic differences in how boys and girls are socialized. For example, boys are encouraged to play active, loud, and violent games (left), whereas girls are encouraged to engage in quieter, nurturing role play (right).

U rb

an Zo

ne /

A la

m y

Sp en

ce r G

ra nt

/A rt

D ire

ct or

s &

T rip

/ A

la m

80220_ch05_ptg01_hr_154-207.indd 187 27/01/16 5:19 PM

188 Chapter 5 Stereotypes, Prejudice, and Discrimination

The traditional pinks and blues are not as distinct today as they used to be. Many gender barriers of the past have broken down, and the colors have some- what blended together. Indeed, there has been a dramatic increase in awareness of gender as something other than a binary category and that the gender one is assigned to at birth does not always reflect the gender identity of the individual later in life. In addition, people are more likely to defy and confront gender-based stereotyping and discrimination today than in years past. Nevertheless, the stereo- types—and, as we discussed earlier, sexism—persist.

What do people say when asked to describe the typical man and woman? Around the world males are said to be more adventurous, assertive, aggressive, independent, and task-oriented; females are thought to be more sensitive, gentle, dependent, agreeable, emotional, and people-oriented. Young children distinguish men from women well before their first birthday, often identify themselves and others as boys or girls by 3 years of age, form gender-stereotypic beliefs and pref- erences about stories, toys, and other objects soon after that, and then use their stereotypes in judging others and favoring their own gender in intergroup situa- tions (Golombok & Hines, 2002; Knobloch et al., 2005; Löckenhoff et al., 2014; Martin & Ruble, 2010; Wood & Eagly, 2010).

Gerianne Alexander (2003) proposed that children’s preferences about sex- based toys, although partly due to gender socialization, also have neurobiological and evolutionary roots. She based this conclusion from data about children ex- posed prenatally to atypical levels of sex hormones and from data about sex differ- ences in toy preferences among nonhuman primates. For example, one intriguing study reported that vervet monkeys showed sex differences in toy preferences similar to those seen in human children (Alexander & Hines, 2002).

Although biological and evolutionary factors may play a role in some of these preferences, it is clear that children have ample opportunity to learn gender stereo- types and roles from their parents and other role models (Montañés et al., 2012; Tenenbaum & Leaper, 2002). Beliefs about males and females are so deeply in- grained that they influence the behavior of adults literally the moment a baby is born. In one fascinating study, the first-time parents of 15 girls and 15 boys were interviewed within 24 hours of the babies’ births. There were no differences be- tween the male and female newborns in height, weight, or other aspects of physical appearance. Yet the parents of girls rated their babies as softer, smaller, and more finely featured. The fathers of boys saw their sons as stronger, larger, more alert, and better coordinated (Rubin et al., 1974). Could it be there really were differences that only the parents were able to discern? Doubtful. In another study, Emily Mond- schein and others (2000) found that mothers of 11-month-olds underestimated their infants’ crawling ability if they were girls but overestimated it if they were boys.

As they develop, boys and girls receive many divergent messages in many dif- ferent settings. Barbara Morrongiello and Tess Dawber (2000) conducted a study that reflects this point. They showed mothers videotapes of children engaging in somewhat risky activities on a playground and asked them to stop the tape and indicate whatever they would ordinarily say to their own child in the situation shown. Mothers of daughters intervened more frequently and more quickly than did mothers of sons. As shown in Table 5.2, mothers of daughters were more likely to caution the child about getting hurt, whereas mothers of sons were more likely to encourage the child’s risky playing. In a more recent study by Morrongiello and others (2010), mothers and fathers were more likely to react to a toddler’s risk-taking behavior with anger and a focus on discipline if it was a boy, and with safety and the sense that the child could learn not to engage in such behavior if it was a girl.

80220_ch05_ptg01_hr_154-207.indd 188 27/01/16 5:19 PM

189 Causes of the Problem: Intergroup, Motivational, Cognitive, and Cultural Factors

Social role Theory As children develop, they begin to look at the larger culture around them and see who occupies what roles in society as well as how these roles are valued. According to Alice Eagly’s (Eagly & Wood, 2012; Koenig & Eagly, 2014) social role theory, although the perception of sex differences may be based on some real differences, it is magnified by the unequal social roles men and women occupy.

The process involves three steps. First, through a combination of biological and social factors, a divi- sion of labor between the sexes has emerged over time, both at home and in the work setting. Men are more likely to work in construction or business; women are more likely to care for children and to take lower-paying jobs. Second, since people behave in ways that fit the roles they play, men are more likely than women to wield physical, social, and economic power. Third, these behavioral differences provide a continuing basis for social perception, lead- ing us to perceive men as dominant and women as domestic “by nature,” when in fact the differences may reflect the roles they play. In short, sex stereo- types are shaped by—and often confused with—the unequal distribution of men and women into different social roles. According to this theory, perceived differences between men and women are based on real be- havioral differences that are mistakenly assumed to arise from gender rather than from social roles. Social role theory and socialization processes more generally can, of course, be extended beyond gender stereotypes and sexism. Seeing what groups of people tend to occupy what roles in society can fuel numerous stereo- types and prejudices.

Media effects More than ever, children, adolescents, and adults seem to be im- mersed in popular culture transmitted via the mass media. Watching TV shows on our phones or iPads while on the stationary bike at the gym, checking out the latest viral videos or celebrity Instagram posts while taking a break at work or the coffee shop, seeing advertisements popping up on our computer screens like weeds, glanc- ing at the tabloid cover shots of the latest starlet hounded by relentless paparazzi— there often seems no escape. Through the ever-present media, we are fed a steady diet of images of people. These images have the potential to perpetuate stereotypes and discrimination.

Fortunately, the days when the media portrayed women and people of color in almost exclusively stereotypical, powerless roles are gone. Still, research in numerous countries around the world examining such varied content as music videos, commercials and print ads, TV programs, and children’s books reveals that stereotyping does persist (Kay & Furnham, 2013; Kirsch & Murnen, 2015; Michelle, 2012; Wallis, 2011).

More to the point is the fact that media depictions can influence viewers, often without the viewers realizing it. Studies have shown, for example, that female college students who had just watched a set of commercials in which the female characters were portrayed in stereotypical fashion tended to express lower self- confidence, less independence, and fewer career aspirations, and even performed more poorly on a math test, than did those who viewed stereotype-irrelevant or

social role theory The theory that small gender differences are magnified in perception by the contrasting social roles occupied by men and women.

Adapted from Morrongiello & Dawber, 2000.

m T a b l e 5 . 2

What mothers Would Say

Frequency of Statement by

Context of Statement Mothers of girls Mothers of Boys

Caution 3.9 0.7

Worry about injury 9.2 0.2

Directive to stop 9.3 0.6

Encouragement 0.5 3.0

Mothers of young boys or girls watched a videotape of another child engaging in somewhat risky behavior on a playground. The mothers were instructed to stop the videotape whenever they would say something to the child if the child were theirs and to indicate what they would say. Mothers of daughters stopped the tape much more often than mothers of sons to express caution (“Be careful!”), worry about injury (“You could fall!”), and directives to stop (“Stop that this instant!”). In contrast, mothers of sons were more likely to indicate encouragement (“Good job! Let me see you go higher!”).

80220_ch05_ptg01_hr_154-207.indd 189 27/01/16 5:19 PM

190 Chapter 5 Stereotypes, Prejudice, and Discrimination

Although images of attractive people sell magazines and many consumers enjoy looking at them, they do raise the question of whether exposure to so many of these kinds of images also produce negative consequences. For example, does repeated exposure to such images perpetuate stereotypes or cause some people to engage in dangerous behaviors to try to achieve what are often impossible and unhealthy standards of masculinity and femininity?

counter-stereotypical ads (Davies et al., 2002; Geis et al., 1984; Jennings et al., 1980; Ward & Friedman, 2006). These effects are by no means limited to gender. Playing violent video games or listening to particular types of music, for example, can heighten aggression and discrimination against outgroup members (Greite- meyer, 2014; LaMarre et al., 2012).

Media images of impossibly thin or proportioned female models are implicated in the near-epidemic incidence of eating disorders and debilitating anxiety over physical appearance, particularly among young European American women (Henderson-King et al., 2001; Moradi et al., 2005; Ward & Friedman, 2006). The media’s impact may be especially negative among individuals who already have concerns about their appearance or are particularly concerned with other people’s opinions (Henderson-King & Henderson-King, 1997; McCabe et al., 2010). Men’s body images are also affected by the media (Ricciardelli & McCabe, 2011; Sebag-Montefiore, 2015). Indeed, graphic images of muscu- lar and lean male models have become increasingly prevalent of late. More and more cases come to light every year of boys and young men copying star athletes by taking steroids and other drugs that can make them look more like their role models but that can seriously threaten their health (Martin & Govender, 2011; McCreary, 2011; Melki et al., 2015).

j How Stereotypes Distort Perceptions and resist Change

Once stereotypes and prejudices are in place due to the factors we’ve been discussing, why are they often so resistant to change? Although some stereo- types may be accurate, some certainly are false, and many are at least over- simplifications (Jussim, 2012; Madon et al., 1998; McCrae et al., 2013; Scherer et al., 2015). Why, then, do inaccurate stereotypes persist despite evidence that should discredit them? In this section we turn to some of the mechanisms that help perpetuate stereotypes.

Confirmation Biases and Self-Fulfilling Prophecies Imagine learning that a mother yelled at a 16-year-old girl, a lawyer behaved aggressively, and a Boy Scout grabbed the arm of an elderly woman crossing the street. Now imagine that a construction worker yelled at a 16-year-old girl, a homeless man behaved aggressively, and an ex-con grabbed the arm of an elderly woman crossing the street. Do very different images of these actions come to mind? This is a funda- mental effect of stereotyping: Stereotypes of groups influence people’s percep-

tions and interpretations of the behaviors of group members. This is especially likely when a target of a stereotype behaves in an ambiguous way; perceivers reduce the ambiguity by interpreting the behavior as consistent with the stereotype (Dunning & Sherman, 1997; Kunda et al., 1997).

The effect of stereotypes on individuals’ perceptions is a type of confirmation bias, which, as we saw in Chapter 4, involves people’s tendencies to interpret, seek, and create information that seems to confirm their expectations. In one classic study, for example, black and white sixth-grade boys saw pictures and descriptions of ambiguously aggressive behaviors (such as one child bumping into another). Both the black and the white boys judged the behaviors as more mean and threatening if the behaviors were performed by black boys than if they were done by white boys, thereby seeming to confirm racial stereotypes (Sagar & Schofield, 1980).

Bi ll

A ro

n/ Ph

ot oE

di t

M ik

e M

ar go

l/ Ph

ot oE

di t

80220_ch05_ptg01_hr_154-207.indd 190 27/01/16 5:19 PM

191 Causes of the Problem: Intergroup, Motivational, Cognitive, and Cultural Factors

Stereotypes can be reinforced also through the illusory correlation, a tendency for people to overestimate the link between variables that are only slightly or not at all correlated (Hamilton & Rose, 1980; Kutzner & Fiedler, 2015; Sherman et al., 2009; Van Rooy et al., 2013). One kind of illusory correlation occurs when people overestimate the association between variables that are relatively rare. For example, if people read about a variety of criminal acts, most of which are com- mitted by members of a majority group and some of which are committed by members of a particular minority group, they may overestimate the association between minority group status (a relatively rare group) and criminal behavior (a relatively rare behavior). This tendency can create or perpetuate negative stereo- types. Illusory correlations may also be produced through people’s tendency to overestimate the association between variables they already expect to go together. For example, if perceivers who hold stereotypes about women being poor drivers witness 100 men and 100 women driving, and 10% of each group get into an accident, they may overestimate the number of women and underestimate the number of men who had accidents. In other words, they see an association between gender and accidents that is not supported by the data.

Stereotypes typically are held not just by individuals but by many people within a culture, and they are often perpet- uated through repeated communications. In a classic dem- onstration, Gordon Allport and Leo Postman (1947) showed participants a picture of a subway train filled with passengers. In the picture were a black man dressed in a suit and a white man holding a razor. One participant viewed the scene briefly and then described it to a second participant who had not seen it. The second participant communicated the description to a third participant and so on, through six rounds of com- munication. The result: In more than half the sessions, the final participant’s report indicated that the black man, not the white man, held the razor.

In a more recent demonstration of a similar point, Yoshi- hisa Kashima and others (2013) had Australian students read a story about an Australian Rules Football player. The stu- dents were put in groups of four. One person read the story, and after a delay of a few minutes transmitted the story to the next student, and so on down the four-person chain. The students were supposed to relay the story as accurately as possible. Some of the information in the story was consis- tent with stereotypes about Australian Rules Football players (e.g., “On the way, Gary and his mate drank several beers in the car”), and some of it was inconsistent with the stereo- type (e.g., “He switched on some classical music”). Although the first student in the chain was likely to communicate both stereotype-consistent and stereotype-inconsistent informa- tion, as the story went from person to person the stereotype-inconsistent infor- mation was progressively screened out (see d Figure 5.12). By the time the third person told the story to the final person in the chain, the football player seemed more clearly stereotypical than he had seemed in the original story.

Confirmation biases are bad enough. But even more disturbing are situations in which stereotyped group members are led to actually behave in stereotype- confirming ways. In other words, stereotypes can create self-fulfilling prophecies

illusory correlation An overestimate of the association between variables that are only slightly or not at all correlated.

2.5

0.5

1.5

0 1st 2nd 3rd

Stereotype-Inconsistent Information

Conversation

Stereotype-Consistent Information

d F i g u r e 5 . 1 2

Stereotype Confirmation via Communication

Australian students read a story about an Australian Rules Football player who behaved in ways both consistent and inconsistent with stereotypes about such players. Each student who read the story then had to orally tell another student as much as he or she could remember about the story, who then in turn told a third student, who in turn told a fourth. Although the first student told both stereotype-consistent and -inconsistent details, by the time of the final telling of the story, more of the stereotype-inconsistent details dropped out, making the player sound much more stereotypic than he was. Kashima et al., 2013

80220_ch05_ptg01_hr_154-207.indd 191 27/01/16 5:19 PM

192 Chapter 5 Stereotypes, Prejudice, and Discrimination

(Madon et al., 2011; Rosenthal, 2002). As noted in Chapter 4, a self-fulfilling prophecy occurs when a perceiver’s false expectations about a person cause the person to be- have in ways that confirm those expectations. Stereotypes can trigger such behavioral confirmation. Consider a classic experiment by Carl Word and others (1974) involv- ing a situation of great importance in people’s lives: the job interview. White partici- pants, without realizing it, sat farther away, made more speech errors, and held shorter interviews when interviewing black applicants than they did when interviewing white applicants. This colder interpersonal style, in turn, caused the black applicants to behave in a nervous and awkward manner. In short, the whites’ racial stereotypes and prejudice actually hurt the interview performance of the black candidates. Since the black candidates’ interview performance tended to be objectively worse than that of the white candidates, it seemed to confirm the interviewers’ negative stereotypes— but this poor performance was caused by the interviewers, not the interviewees.

Attributions and Subtyping People also maintain their stereotypes through how they explain the behaviors of others. Chapter 4 discusses how perceivers make attri- butions, or explanations, about the causes of other people’s behaviors and how these attributions can sometimes be flawed. One key finding discussed in that chapter is that people often don’t take into account the context that someone was in when they try to explain his or her behavior. This can help perpetuate negative stereotypes by, for example, people failing to recognize how a poor performance by a member of a stereotyped group may be due to the effects of these stereotypes rather than a lack of

actual ability. On the other hand, when people

see others acting in ways that seem to contradict a stereotype, they may be more likely to think about situational factors in order to explain the sur- prising behavior. Rather than accept a stereotype-disconfirming behavior at face value, such as a woman de- feating a man in an athletic contest, perceivers imagine the situational factors that might explain away this apparent exception to the rule, such as random luck, ulterior motives, or other special circumstances. In this way, perceivers can more easily main- tain their stereotypes of these groups (Espinoza et al., 2014; Sekaquaptewa et al., 2003; Sherman et al., 2005).

If we encounter someone’s be- havior that clearly contradicts our ste- reotypes and we can’t easily explain it away as due to some situational

factor, we may unwittingly pull out another trick to preserve our stereotypes: We consider the action or the person a mere exception to the rule. Confronted with a woman who does not seem modest and nurturing, for example, people can either develop a more diversified image of females or toss the mismatch into a special subtype—say, “cutthroat women” or “divas.” To the extent that people create this subtype, their existing image of women in general may remain relatively intact (Riek et al., 2013; Schneider & Bos, 2014; Wilder et al., 1996).

“Not everyone’s life is what they make it. Some people’s life is what other people make it.”

—A character in Alice Walker’s short story “You Can’t Keep a Good Woman Down”

People who believe that women are not tough and athletic may maintain this stereotype even in the face of examples like these women who defy it. One way they do this is by subtyping such individuals as exceptions who don’t truly represent the general category of women.

A P

Im ag

es /T

or u

Ta ka

ha sh

80220_ch05_ptg01_hr_154-207.indd 192 27/01/16 5:19 PM

193 Causes of the Problem: Intergroup, Motivational, Cognitive, and Cultural Factors

j Automatic Stereotype Activation Part of the power of stereotypes is they can bias our perceptions and responses even if we don’t personally agree with them. In other words, we don’t have to believe a stereotype for it to trigger illusory correlations and self-fulfilling prophe- cies or to influence how we think, feel, and behave toward group members. Some- times just being aware of stereotypes in one’s culture is enough to cause these effects. Moreover, stereotypes can be activated without our awareness.

In a very influential line of research, Patricia Devine (1989) distinguished be- tween automatic and controlled processes in stereotyping. She argued that people have become highly aware of the content of many stereotypes through socializa- tion from their culture. Because of this high awareness people may automatically activate stereotypes whenever they are exposed to members of groups for which popular stereotypes exist. Thus, just as after hearing bacon and many of us are automatically primed to think eggs, when we think of a stereotyped group we are also primed to think of concepts relevant to the stereotype.

To demonstrate this point Devine exposed white participants in one study to subliminal presentations on a computer monitor. Subliminally presented informa- tion is presented so quickly that perceivers do not even realize that they have been exposed to it. In Devine’s study these presentations consisted of words relevant to stereotypes about black people, such as Africa, ghetto, welfare, and basket- ball. Because the words were presented subliminally, the participants were not consciously aware that they had seen these words. Those who were subliminally primed with many of these words activated the African American stereotype and saw another person’s behavior in a more negative, hostile light. Especially note- worthy is the fact that these effects occurred even among participants who did not consciously believe in the stereotypes in question.

Devine’s theory sparked an explosion of interest in these issues. The conclu- sions from the ensuing research is that stereotypes can be activated implicitly and automatically, influencing subsequent thoughts, feelings, and behaviors even among perceivers who are relatively low in prejudice. But it also is clear that sev- eral factors can make such activation more or less likely to happen. For example, some stereotypes are much more prevalent than others in a particular culture, and with more exposure to a stereotype comes a greater likelihood of automatic activation. Another factor is how prejudiced the perceiver is. Although stereotypes can be activated automatically even among perceivers very low in prejudice, the threshold for what triggers stereotype activation may be lower for those relatively high in prejudice (Lepore & Brown, 2002; Wittenbrink et al., 1997).

Motivation can also play an important role. For example, when people’s self- esteem is threatened, they may become motivated to stereotype others so that they will feel better about themselves, and this can make them more likely to activate stereotypes automatically (Spencer et al., 1998). What about being moti- vated to control oneself from activating or applying stereotypes? Is this effective? People who are motivated for intrinsic reasons (that is, they really don’t want to be prejudiced, rather than just not wanting to be seen by others as prejudiced) tend to be somewhat more successful at this kind of self-regulation. We will return to this issue in more detail later in the chapter.

j The Shooter Bias The issue of automatic activation of stereotypes and its effects can be seen in concrete terms by focusing on one particular type of tragedy that has occurred far

People’s very quick judgments are not influenced by a stereotype unless they actually believe the stereotype to be true.

FAlSe

subliminal presentation A method of presenting stimuli so faintly or rapidly that people do not have any conscious awareness of having been exposed to them.

80220_ch05_ptg01_hr_154-207.indd 193 27/01/16 5:19 PM

194 Chapter 5 Stereotypes, Prejudice, and Discrimination

too often: the shooting of an unarmed African American man by the police. The opening of this chapter mentioned protests stemming from a series of incidents in late 2014 and 2015 in which police shot or killed unarmed African American men in response to what the police said was threatening behavior. One such incident that occurred more than a decade before inspired a wave of social psychologi- cal research relevant to this issue. In New York City on February 4, 1999, a West African immigrant named Amadou Diallo was spotted by four white police of- ficers who thought he was acting suspiciously and that he matched the general description of a suspected rapist they were searching for. As they approached him, Diallo reached into his pocket and began pulling out his wallet. One of the officers yelled, “Gun!” The police fired at Diallo 41 times, hitting him with 19 of the bul- lets. Diallo lay dead in the vestibule; he did not have a weapon.

Protesters held rallies in the days that followed, chanting “41 shots” and hold- ing up wallets. Others defended the police officers, noting how difficult it is to make life-or-death decisions in the blink of an eye. The officers were found not guilty of any criminal charges. A central question, of course, was whether ste- reotypes associated with the color of Diallo’s skin made the officers more likely to misperceive the wallet as a gun. Although none of us can ever know whether this was the case in the Diallo tragedy, it is possible to apply social psychology re- search to answer the more general question of whether an unarmed man is more likely to be misperceived as holding a gun if he is black than if he is white.

Keith Payne (2001) was the first to publish a study directly inspired by this question. The participants in his study were undergraduate students, not police officers, but their task was to try to make the kind of decision the police had to make: very quickly identify an object as a weapon or not. Pictures of these objects (such as guns or tools) were presented on a computer screen, immediately pre- ceded by a quick presentation of a black or a white male face. The pictures were presented for fractions of a second. Payne found that the participants were more likely to mistake a harmless object for a weapon if it was preceded by a black face than if the object was preceded by a white face. In other words, a quick glimpse of a black male face primed the participants to see a threatening object more than seeing a white male face did.

Joshua Correll and others (2002) designed a video game for their experiment in which they had participants decide whether or not to “shoot” a target person who appeared on their computer screen (see d Figure 5.13). Some of these tar- gets were white men and others were black men. Some of them held guns and others held harmless objects (such as a black cell phone or a wallet). If the target held a gun, the participants were supposed to hit a “shoot” key as quickly as pos- sible. If he held a harmless object, they were to hit a “don’t shoot” key as quickly as they could.

As in the Payne study, these participants showed a bias consistent with racial stereotypes. If the target held a gun, they were quicker to press the “shoot” key if he was black than if he was white. If the target held a harmless object, they took longer to press the “don’t shoot” key if he was black than if he was white. In ad- dition, participants were more likely to mistakenly “shoot” an unarmed target if he was black than if he was white.

Since these first experiments, a large number of others followed to examine the processes involved more closely. For example, Debbie Ma and Joshua Correll (2011) found that this racial bias in decisions to shoot was significantly stronger if the targets looked more stereotypic of their respective races (as rated by samples of participants) than if they did not. In a study examining a different stereotype, Christian Unkelbach and others (2008) found that Australian participants were

According to a 2015 analysis by The Guardian, black Americans are more than twice as likely to be unarmed when killed during encounters with police as white Americans.

Swaine et al., 2015

80220_ch05_ptg01_hr_154-207.indd 194 27/01/16 5:19 PM

195 Causes of the Problem: Intergroup, Motivational, Cognitive, and Cultural Factors

more likely to shoot at targets wearing Muslim headgear than bareheaded targets who did not appear Muslim.

Taken together the results of these studies suggest that when the decision must be made very quickly, members of some groups are more likely to be mis- takenly perceived as holding a gun than members of other groups. It is important to note, though, that the participants in the initial studies were not police officers; they were undergraduate students or individuals from a community sample. Po- lice officers receive extensive training in these kinds of tasks. But as the quote in the margin by social psychologist Anthony Greenwald indicates, the police may not be trained to avoid activating racial stereotypes, and given the prevalence and power of these stereotypes in our society, there is good reason to suspect that they would be influenced by them in split-second decisions.

Fortunately, several researchers have conducted these experiments with police officers as participants (Correll et al., 2007b; Peruche & Plant, 2006). In some of this research, police officers have shown a similar bias to mistakenly “shoot” a black than white target, and in some they have shown little or no racial bias in the decision to shoot. Even in the studies in which the officers’ decisions were not as racially biased, however, racial bas was evident in the officers’ re- sponse times, indicating that they were quicker to decide to shoot armed black than white targets and slower to decide to not shoot unarmed black than white targets. In addition, several studies have also found that for both civilians and police officers, training designed specifically to curtail these biases can be effec- tive, at least to some degree.

Reviewing all these studies, Joshua Correll and others (2014) concluded that stereotypes certainly can alter perceptions about the presence of weapons and the decision to shoot. In addition, even though police officers are able to avoid these biases under testing conditions, when officers must make these decisions under conditions of fatigue, high stress, and distraction—the conditions officers often face when having to make real shoot-or-not decisions—their ability to overcome stereotype-based biases are compromised, thus increasing the likelihood of the mistakes seen in the tragic incidents that sparked all this research.

d F i g u r e 5 . 1 3

Shoot or Not?

These are examples of scenes from the video game that Joshua Correll and his colleagues created to investigate whether perceivers, playing the role of police officers, would be biased by the target’s race when trying to determine very quickly whether they should shoot him because he is holding a weapon or refrain from shooting because he is holding a harmless object. Correll et al., 2002.

“Police receive training to make them more sensitive to weapons, but they don’t get training to undo unconscious race stereo- types or biases.”

—Anthony Greenwald

80220_ch05_ptg01_hr_154-207.indd 195 27/01/16 5:19 PM

196 Chapter 5 Stereotypes, Prejudice, and Discrimination

If individuals are racially biased in these perceptions and decisions, does that mean that they have racist at- titudes and beliefs? The evidence thus far suggests that this may not be the case. For example, Correll and oth- ers (2002) found that the the racial bias in the decision to shoot was not related to participants’ levels of racial prejudice. In addition, these researchers also found that African American participants showed the same bias against black targets as white participants did, again sug- gesting that racial prejudice is not necessarily reflected in this bias. Consistent with much of the research we’ve re- ported in this chapter, awareness of the stereotype was a necessary factor, but believing it to be true was not. Indeed, Correll and others (2007a) found that by manipulating the accessibility of stereotypes that associate blacks with dan- ger in perceivers’ minds (such as by having them first read newspaper articles about black or white criminals), they could strengthen or weaken this bias.

D av

id S

ip re

ss /T

he N

ew Y

or ke

r C ol

le ct

io n

20 00

/C ar

to on

ba nk

.c om

Reducing the Problem: Social Psychological Solutions We have discussed the persistence of stereotypes, prejudice, and discrimination today, as well as numerous factors at the root of these problems. In this final sec- tion, we focus on some of the approaches that have been suggested for combating stereotypes, prejudice, and discrimination.

j intergroup Contact One of the many enduring ideas in Gordon Allport’s (1954) classic book, The Nature of Prejudice, was the contact hypothesis, which states that under certain conditions, direct contact between members of rival groups will reduce intergroup prejudice. Around the time of the publication of this book, the U.S. Supreme Court ruled in the historic 1954 case of Brown v. Board of Education of Topeka that racially separate schools were inherently unequal and violated the U.S. Constitu- tion. In part, the decision was informed by empirical evidence supplied by 32 emi- nent social scientists on the harmful effects of segregation on both race relations and the self-esteem and academic achievement of black students (Allport et al., 1953). The Supreme Court’s decision propelled the nation into a large-scale social experiment. What would be the effect?

Despite the Court’s ruling, desegregation proceeded slowly. There were stalling tactics, lawsuits, and vocal and even violent opposition. Many schools remained untouched until the early 1970s. Then, as the dust began to settle, research brought the grave realization that contact between black and white schoolchildren was not having the intended effect on intergroup attitudes. Walter Stephan (1986) reviewed studies conducted during and after desegregation and found that although 13% of the studies reported a decrease in prejudice among whites, 34% reported no change, and 53% reported an increase. These findings forced social psychologists to challenge the wisdom of their testimony to the Supreme Court and to reexamine the contact hypothesis that had guided that advice in the first place.

contact hypothesis The theory that direct contact between hostile groups will reduce intergroup prejudice under certain conditions.

“‘see that man over there?’ ‘Yes’ ‘Well, I hate him’ ‘But you don’t know him.’ ‘That’s why I hate him.’

—Gordon Allport

You look like this sketch of someone who is thinking about committing a crime.

80220_ch05_ptg01_hr_154-207.indd 196 27/01/16 5:19 PM

197 Reducing the Problem: Social Psychological Solutions

Is the original contact hypothesis wrong? No. Although desegregation did not immediately produce the desired changes, it’s important to realize that the ideal conditions for successful intergroup contact did not exist in the public schools that desegregated. Nobody ever said that deeply rooted prejudices could be erased just by throwing groups together. According to the contact hypothesis, four conditions are ideal for contact to succeed. These conditions are summarized in Table 5.3.

Thomas Pettigrew and Linda Tropp have extensively reviewed the research relevant to the contact hypothesis, and their conclusions have been very encour- aging (Pettigrew et al., 2011; Tropp, 2013; Tropp & Page-Gould, 2015). In a series of meta-analyses involving more than 500 studies and a quarter of a million partici- pants in 38 nations Pettigrew and Tropp (2000, 2006, 2008) found reliable support for the benefits of intergroup contact in reducing prejudice, particularly when the contact satisfies at least some of the conditions in Table 5.3. These researchers propose that contact reduces prejudice by (1) enhancing knowledge about the outgroup; (2) reducing anxiety about intergroup contact; and (3) increasing empathy and perspective taking.

One of the most successful demonstrations of desegrega- tion took place on the baseball diamond. On April 15, 1947, Jackie Robinson played for baseball’s Brooklyn Dodgers and became the first black man to break “the color barrier” in a major American sport. Robinson’s opportunity came through Dodgers owner Branch Rickey, who felt that integrating base- ball was both moral and good for the game (Pratkanis & Turner, 1994). Rickey knew all about the contact hypothesis and was assured by a social scientist friend that a team could furnish the conditions needed for it to work: equal status among team- mates, personal interactions, dedication to a common goal, and a positive climate from the owner, managers, and coaches. The rest is history. Rickey signed Robinson and tried to cre- ate the situation necessary for success. Although Robinson did face a great deal of racism, he endured, and baseball was integrated. At the end of his first year, Jackie Robinson was named rookie of the year, and in 1962 he was elected to the

m T a b l e 5 . 3

The Contact Hypothesis: Conditions

1. Equal status. The contact should occur in circumstances that give the two groups equal status.

2. Personal Interaction. The contact should involve one-on-one interactions among individual members of the two groups.

3. Cooperative activities. Members of the two groups should join together in an effort to achieve superordinate goals.

4. Social norms. The social norms, defined in part by relevant authorities, should favor intergroup contact.

Four conditions are deemed ideal for intergroup contact to serve as a treatment for racism.

(Left) Students at Central High School in Little Rock, Arkansas, in September 1957 shout insults at 16-year-old Elizabeth Eckford as she walks toward the school entrance. National Guardsmen blocked the entrance and would not let her enter. (Right) Jackie Robinson and Branch Rickey discuss Robinson’s contract with the Brooklyn Dodgers. In 1947, Robinson became the first African American to cross “the color line” and play Major League Baseball, thereby beginning the integration of major American sports.

A P

Im ag

es /W

ill C

ou nt

A P

Im ag

es /W

id e

W or

ld P

ho to

80220_ch05_ptg01_hr_154-207.indd 197 27/01/16 5:19 PM

198 Chapter 5 Stereotypes, Prejudice, and Discrimination

Baseball Hall of Fame. At his induction ceremony, Robinson asked three people to stand beside him: his mother, his wife, and his friend Branch Rickey.

Another potential cause for optimism, although perhaps tinged with some regret, is the finding by Nicole Shelton and Jennifer Richeson (2005) that both whites and blacks would like to have more contact with each other but believe that the other group does not want to have contact with them! This may be a case in which education about the problem can be an important tool in correcting it.

j intergroup Friendships and extended Contact Developing friendships across groups is one of the best ways to experience many of the optimal conditions for contact that are listed in Table 5.3. Friendships typi- cally involve equal status, meaningful one-on-one interactions that extend across time and settings, and cooperation toward shared goals. It therefore makes sense that one of the most encouraging lines of research on improving intergroup rela- tions has focused on cross-group friendships. A recent meta-analysis by Kristin Davies and others (2011) on 135 studies supports the idea that cross-group friend- ships are associated with more positive attitudes and behaviors toward outgroup members.

The evidence for this comes not only from correlational studies, such as surveys that show the strong as- sociation between cross-group friend- ships and less intergroup anxiety and prejudice, but also from experiments in which the researchers create such friendships among some of their par- ticipants. A series of creative experi- ments along these lines by Elizabeth Page-Gould, Rodolfo Mendoza-Denton, and their colleagues has found strong evidence for the causal role that such friendships can play in reducing inter- group prejudice (Mendoza-Denton & Page-Gould, 2008; Page-Gould et al., 2008; 2010).

Of course there’s often even greater contact—psychologically as well as physically—between romantic part-

ners than friends, and some research has examined whether cross-group dating is associated with more positive intergroup attitudes. For example, a longitudinal study of dating in college by Shana Levin and others (2007) revealed that white, Asian American, and Latino students who dated outside their group more during college showed less ingroup bias and intergroup anxiety at the end of college than students who did not date outside their own racial group.

Even people who do not have a friend from an outgroup can benefit from hav- ing ingroup friends who do. A number of studies have found evidence for what is known as the extended contact effect, or the indirect contact effect—knowing that an ingroup friend has a good and close relationship with a member of an out- group can produce positive intergroup benefits in ways similar to direct contact. Why would this happen? Several causes are involved, such as reducing ignorance and anxiety about outgroup members and providing individuals with positive

“A friendship is bonding capital, but when that friendship is multi- racial, it is also bridging capital.”

—Emerson et al., 2002, p. 70.

Although Charleston High School in Mississippi has been integrated for decades, it was not until 2008 that the school board allowed a racially integrated senior prom, although some at the school organized a “white prom” as an alternative. Pictured here are two of the students who attended the prom standing in front of a poster for the movie Prom Night in Mississippi, an award-winning documentary about this story. Some integrated schools in the United States still hold racially separate proms.

Ja so

n La

Ve ris

/W ire

Im ag

e/ G

et ty

Im ag

80220_ch05_ptg01_hr_154-207.indd 198 27/01/16 5:19 PM

199 Reducing the Problem: Social Psychological Solutions

examples of outgroup members (Cameron et al., 2011; Dovidio et al., 2011; Eller et al., 2011; Wright et al., 2008).

j The Jigsaw Classroom As the third condition in Table 5.3 indicates, cooperation and shared goals are ideal for intergroup contact to be successful. Yet the typical classroom is filled with competition—exactly the wrong ingredient. Picture the scene. The teacher stands in front of the class and asks a question. Many children wave their hands, each straining to catch the teacher’s eye. Then, as soon as one student is called on, the others groan in frustration. In the competition for the teacher’s approval, they are losers—hardly a scenario suited to positive intergroup contact. To combat this problem in the classroom, Elliot Aronson and his colleagues (1978) developed a cooperative learning method called the jigsaw classroom. In newly desegregated public schools in Texas and California, they assigned fifth graders to small racially and academically mixed groups. The material to be learned within each group was divided into subtopics, much the way a jigsaw puzzle is broken into pieces. Each student was responsible for learning one piece of the puzzle, after which all members took turns teaching their material to one another. In this system, everyone—regardless of race, ability, or self-confidence—needs everyone else if the group as a whole is to succeed.

The method produced impressive results (Aronson, 2004, 2011). Compared with children in traditional classes, those in jigsaw classrooms grew to like each other more, liked school more, were less prejudiced, and had higher self-esteem. What’s more, academic test scores improved for minority students and remained the same for white students.

j Shared identities One important consequence of the jigsaw classroom technique is that individuals became more likely to classify outgroup members as part of their own ingroup. Instead of seeing racial or ethnic “others” within the classroom, the students now see fellow classmates. Students feel that they are all in the same boat together. More generally, intergroup contact that emphasizes shared goals and fates can effectively reduce prejudice and discrimination—specifically by changing how group members categorize each other (Bettencourt et al., 2007; Van Bavel & Cunningham, 2009).

The Common Ingroup Identity Model developed by Samuel Gaertner and John Dovidio (2010; 2012) proposes that if members of different groups recategorize themselves as members of a more inclusive superordinate group, intergroup at- titudes and relations can improve. By recognizing their shared categorization, just as the Rattlers and Eagles did when they changed from competitors to collabora- tors in Robbers Cave State Park, “they” become “we,” and a common ingroup identity can be forged. Researchers have induced a common ingroup identity among participants in their studies in a variety of ways, such as by having dif- ferent groups share a common label, dispersing them around a table so that they are not separated according to group, or having them wear similar t-shirts. For example, Blake Riek and others (2010) put Democrats and Republicans together into two-person groups. When their different political party memberships were highlighted, the participants rated the other group more negatively. When their shared identity as Americans was highlighted, however, such as by having them wear shirts with the American flag printed on the front and calling their group the America Group, their intergroup ratings were significantly more positive.

jigsaw classroom A cooperative learning method used to reduce racial prejudice through interaction in group efforts.

80220_ch05_ptg01_hr_154-207.indd 199 27/01/16 5:19 PM

200 Chapter 5 Stereotypes, Prejudice, and Discrimination

It is worth noting that at least in some contexts, individuals from minority groups or groups that have less power in a society tend to not feel as positively as majority group members do about recategorizing their groups into one common ingroup. A group with smaller numbers or less power may feel overwhelmed and a sense of lost identity if they merge completely with a larger or more power- ful group. Instead, members of these groups sometimes prefer or benefit more from dual-identity categorizations, in which their distinctiveness as a member of their specific group is preserved but in which they recognize their connection and potential for cooperation with the majority or more powerful group (Dovidio et al., 2012). A key point remains, though, that seeing connections between the groups and ways in which their identities are shared is essential.

j Trust, Belonging, and reducing Stereotype Threat Claude Steele’s theory of stereotype threat discussed earlier in the chapter re- ceived a great deal of attention because it not only helped explain causes underly- ing some profound social problems—such as the underperformance or reduced interests of large groups of people in various academic and career pursuits—but also offered encouragement rather than pessimism. It illustrated that making even small changes in the situational factors that give rise to stereotype threat can reduce the tremendous weight of negative stereotypes, allowing the targets of stereotypes to perform to their potential.

Dozens and dozens of experiments on stereotype threat were published soon after the theory was introduced in the mid-1990s, and in virtually each of them the nega- tive effects of stereotype threat were reduced significantly or eliminated completely in some conditions of the study. Often this was achieved with seemingly very minor—but important—changes in the setting. Table 5.4 lists a sample of these ways.

A theme running through many of the successful in- terventions against stereotype threat effects is that the individuals feel a sense of trust and safety in the situation. That is, in this particular setting they feel that they are not the target of others’ low expectations and they do not have to be concerned with unfairness or other obstacles that would otherwise distract, worry, or discourage them. The importance of establishing this trust is evident in the results of an experiment by Geoffrey Cohen and others (1999). Students in this study performed a writing task and then received criticism of their writing from a reviewer who pre- sumably was white. Compared to white students, black stu- dents tended to respond less constructively and were more likely to dismiss the criticism as biased against them. These negative reactions to the criticism were eliminated, how- ever, if the reviewer who gave the criticism added two ele- ments: (1) he made it clear that he had high standards, and (2) he assured the students that he was confident that they had the capacity to achieve those standards. This combina- tion of factors made the students trust the criticism and gave them the sense that they had a fair shot at succeeding at the task.

“The many factors which divide us are actually much more superficial than those we share. Despite all of the things that differentiate us—race, language, religion, gen- der, wealth and so on—we are all equal concerning our fundamental humanity.”

—Dalai Lama

m T a b l e 5 . 4

Combating Stereotype Threat effects

d Describing the task as not indicative of individuals’ intellectual capabilities (Steele & Aronson, 1995)

d Informing individuals that their group typically does not perform worse than other groups on the task (Spencer et al., 1999)

d Giving individuals reason to attribute their anxiety while taking a test to irrelevant factors (Ben-Zeev et al., 2005)

d Getting individuals to think of intelligence as not a fixed trait but instead as something that is malleable and can be improved (Good et al., 2003)

d Exposing individuals to a member of their group who is said to be an expert in the domain in question (Stout et al., 2011)

d Having individuals think about values and interests that are very important to them that are not under threat (Taylor & Walton, 2011; C. von Hippel et al., 2011)

d Highlighting other aspects of individuals’ identities that are associated with positive performance on the task at hand (Shih et al., 2012; Yopyk & Prentice, 2005)

d Excluding the presence of members of outgroups (Sekaquaptewa & Thompson, 2003)

Researchers have reduced or eliminated the negative effects of stereotype threat in particular settings in a variety of ways. Here is a sample of these successful interventions.

80220_ch05_ptg01_hr_154-207.indd 200 27/01/16 5:19 PM

201 Reducing the Problem: Social Psychological Solutions

In addition to causing distrust, one of the most powerful ways in which stereo- type threat undermines students is that it reduces their sense of belonging. That is, targeted students are likely to feel that people like me do not belong here, such as in a particular major, school, career, or some other domain (Cook et al., 2012; Walton & Cohen, 2007). Geoffrey Cohen, Gregory Walton, and their colleagues have applied social psychological techniques to create interventions outside the lab that have had truly remarkable success with students in a number of schools (Walton, 2014). For example, Walton and Cohen (2011) gave some first-semester African American students at a predominately European American college infor- mation designed to reduce feelings of uncertainty about their sense of belonging at the university. They read that it is quite typical of most students—regardless of their gender, race, or ethnicity—to go through periods of social stress and un- certainty during their freshman year, and that these struggles tend to go away soon after their first year. Walton and Cohen found that giving this information to African American students during their first semester at school raised their grade-point averages significantly higher relative to other African American stu- dents who were not given this information. Especially impressive was that this effect lasted through their entire college experience. As discussed in Chapter 1 of this book, Walton and others (2015) used a similar type of intervention that dramatically improved the performance of female students in majors dominated by male students at an engineering school. (You can see this effect illustrated in d Figure 1.1 on p. 9 of Chapter 1.)

Nilanjana Dasgupta and others (2015) also conducted a recent study with female engineers. They reasoned that women’s uncertainty about be- longing in engineering is heightened by typically being so outnumbered by men. To examine this issue, the researchers had female engineers work on engineering problems in small groups that were either mostly female, mostly male, or equally male and female. The results indicated that women who were in the majority were more motivated and in- terested in the group task, demonstrated more con- fidence, and gave more helpful input than women who were in the minority or even women who were in groups with equal numbers of males and females (d see Figure 5.14 for an illustration of one of these findings).

j exerting Self-Control A key point in this chapter is that people often ste- reotype and show prejudice toward others even when they would rather not, sometimes by merely being aware of the stereotype. Can we learn to con- trol and rise above these impulses?

One of the challenges is that trying to suppress stereotyping or to control prejudiced actions can take mental effort, and people often don’t have the time, energy, or awareness to dedicate to this effort (Ito et al., 2015). Eric Wesselmann and colleagues (2012), for example, found that after watching a

D eg

re e

of w

or ry

Sex composition of groups 0

0.2

0.6

0.4

0.8

1.6

1.4

1.2

75% Women 50% Women 25% Women

G ro

up p

ar tic

ip at

io n

Sex composition of groups 3

3.5

4.5

5.5

d F i g u r e 5 . 1 4

Safety in Numbers

Female engineers worked on engineering problems in small groups that were either mostly (75%) female, equally male and female, or mostly (75%) male. Women in the male-majority groups felt significantly more worried about working on the group task than did the women in the female-majority groups (left). Women in the female-majority groups actively participated on the task much more than did women in the other two groups (right). From Dasgupta et al., 2015

80220_ch05_ptg01_hr_154-207.indd 201 27/01/16 5:19 PM

202 Chapter 5 Stereotypes, Prejudice, and Discrimination

video of a young man’s behavior, participants judged him to be more threaten- ing if they had been led to believe that he had been suffering from mental health problems than from dental pain. Interestingly, this effect occurred only when par- ticipants had to make their judgments quickly, suggesting that they relied on their stereotypes more when they didn’t have time to try to overcome their biases.

Some factors make people less likely to have sufficient cognitive resources for successful control. For example, older people have a harder time suppressing stereotypes than younger people, which may explain in part why older people often appear more prejudiced than younger people (von Hippel & Henry, 2011). Some research suggests that being low in blood sugar, such as from being hun- gry, can weaken people’s ability to control stereotyping and prejudice, and that a sugar-heavy drink can correct this problem (Gailliot, 2013; Gailliot et al., 2009). However, some researchers are more skeptical about these blood-sugar findings (Lange & Eggert, 2014; Molden et al., 2012), and so more research will certainly be done to clarify the issue. What is more clear is that being intoxicated makes people have a difficult time suppressing thoughts or inhibiting impulses (for example, consider the dreaded drunk dialing or texting). It may come as little surprise, therefore, that intoxication impairs people’s ability to control stereotype activation and application (Bartholow et al., 2012; Loersch et al., 2015; Parrott & Lisco, 2015).

Being physically tired or being affected by strong emotion or arousal can sap per- ceivers of the cognitive resources necessary to avoid stereotyping. In an intriguing demonstration of this, Galen Bodenhausen (1990) ran an experiment with two types of people: “morning people” (who are most alert early in the morning) and “night people” (who peak much later, in the evening). By random assignment, participants took part in an experiment that was scheduled at either 9 p.m. or 8 p.m. The result? Morning people were more likely to use stereotypes when tested at night; night owls

were more likely to do so early in the morning. More recently, Sonia Ghumman and Chris- topher Barnes (2013) found that the more sleepy the stu- dents in their study were, the more stereotyping and preju- dice they exhibited. (Sleep also plays a role in a particu- larly unusual recent finding reported by Xiaoqing Hu and others (2015), in which partic- ipants who had received train- ing to reduce implicit racial and gender bias showed lon- ger reductions in these biases if they were exposed to cues associated with that training while they slept!)

Just as a good night’s sleep might improve people’s

cognitive control, so too might the right kind of brain stimulation. That is the implication of recent work by Roberta Sellaro and others (2015). Through noninvasive methods they stimulated a part of the brain (the medial prefrontal cortex) that is thought to be associated

Being intoxicated robs people of cognitive resources, making them less able to inhibit impulses and control expressions of stereotyping and prejudice.

M ar

tin T

ho m

as P

ho to

gr ap

hy /A

la m

80220_ch05_ptg01_hr_154-207.indd 202 27/01/16 5:19 PM

203 Reducing the Problem: Social Psychological Solutions

with cognitive control. As the researchers predicted, this led to a reduction in participants’ implicit biases toward outgroup members.

In addition to people’s ability to control prejudice, another issue is what motivates people to do so. Researchers have distinguished between two kinds of motivation to control prejudiced responses and behaviors. One kind is externally driven—not wanting to appear to others to be prejudiced. A second type is inter- nally driven—not wanting to be prejudiced, regardless of whether or not others would find out (Dunton & Fazio, 1997; Plant & Devine, 1998, 2009). Internally motivated individuals are likely to be more successful at controlling stereotyping and prejudice, even on implicit measures, but even they are vulnerable to the strong power of automatic stereotyping and implicit biases.

According to the self-regulation of prejudiced responses model proposed by Margo Monteith and others (2002; Monteith & Mark, 2009), internally motivated individuals in particular may learn to control their prejudices more effectively over time. According to this model, people who are truly motivated to be fair and unprejudiced are often confronted with the sad reality that they have failed to live up to that goal. These realizations lead to unpleasant emotions such as guilt. As individuals experience such feelings of guilt repeatedly, they begin to develop ex- pertise at recognizing the situations and stimuli that tend to trigger these failures, and therefore they can exert more control over them. In so doing, they begin to interrupt what had been automatic stereotype activation.

A great example of this process is reflected in the quotes of a student in one of Monteith’s studies (Monteith et al., 2010). The student described how close he once came to describing a classmate in a racist way, and how terrible he felt about it. “I don’t know why I was going to say it. It was kind of scary how it had been programmed into my head by those people around me.” Later, this partici- pant indicated how often he is reminded of that experience: “When I see people say stuff like that I think about it a lot. . . . It kind of quiets me down and makes me think about what I’m doing and makes me think about what I’m thinking, what I’m saying. It makes me feel that I should be more careful about what I say” (pp. 193–194).

Just as people who are internally motivated to control prejudice tend to be- come more successful at self-regulation than people who are motivated for exter- nal reasons, antiprejudice messages that are designed to appeal more to people’s internal motivations may be more effective than messages that seem more exter- nally focused. To study this issue, Lisa Legault and others (2011) developed a pair of prejudice-reduction brochures. One of these brochures was designed to appeal to internal motivations. This brochure emphasized the personal as well as societal benefits of tolerance and diversity. For example, one line read, “When we let go of prejudice, the rich diversity of society is ours to enjoy.” The other brochure ap- pealed to external motivations, emphasizing the obligations that compel individu- als and organizations to act in nonprejudiced ways. For example, an excerpt from this brochure included, “Teachers and students caught displaying racist attitudes and behavior can face serious consequences, such as termination and expulsion.”

Students at a Canadian university were randomly assigned to read either of the two brochures, or they received no brochure at all. Later these stu- dents completed a measure of their prejudice toward black people (no black students were included in the study). The researchers found that these stu- dents exhibited significantly less prejudice if they received the brochure that appealed to internal motivations to control prejudice than if they received no brochure. On the other hand, the students who received the brochure appealing to external factors exhibited significantly more prejudice than

80220_ch05_ptg01_hr_154-207.indd 203 27/01/16 5:19 PM

204 Chapter 5 Stereotypes, Prejudice, and Discrimination

the students who received no brochure (see d Figure 5.15). The brochure that emphasized the external rather than internal reasons to con- trol prejudice likely came across to the students as very controlling and made students rebel against it somewhat. People don’t like to be told how they must think and not think, and so an- tiprejudice messages that are perceived in that way can be counterproductive.

4.0

3.8

4.4

4.2

3.6

4.6

4.8

Internal Motives

No Brochure

External Motives

Pr ej

ud ic

e d F i g u r e 5 . 1 5

Productive and Counterproductive Antiprejudice Messages

Compared to students who did not receive any brochure, those who read an antiprejudice brochure that emphasized internally driven motivations to not be prejudiced (such as why one would benefit by promoting diversity) later exhibited significantly less prejudice toward black people. In contrast, students who received the brochure that appealed to external factors (such as how one could get in trouble if caught doing or saying something racist) exhibited significantly more prejudice. From Legault et al., 2012

j Changing Cognitions, Cultures, and Motivations

We have just discussed some of the challenges involved in trying not to think about stereotypes or act in a prejudiced way. There are several ways of thinking that can be more productive. Social– cognitive factors that research has shown can re- duce stereotyping and prejudice include:

d Being exposed to and thinking about examples of group members that are inconsistent with the ste- reotype (Columb & Plant, 2011).

d Learning about the variability that exists among the people in a group (Brauer & Er-rafiy, 2011; Brauer et al., 2012).

d Being induced to take the perspective of a person from a stereotyped group (Maister et al., 2015; Peck et al., 2013; Todd et al., 2012).

d Being encouraged to pay attention to or confront instances of discrimination (Becker & Swim, 2011; Dickter & Newton, 2013; Gulker et al., 2013).

d Believing that prejudice can be learned and unlearned and is therefore malleable rather than a fixed, unchanging reality (Carr et al., 2012; Neel & Shapiro, 2012).

d Learning that race is more ambiguous and socially determined than simply a genetic, fixed category (Williams & Eberhardt, 2008).

d Taking multicultural views that recognize but also value group differences, that reveal new ways of thinking, and that acknowledge how different groups have in- fluenced each other throughout history, as opposed to pretending to not notice dis- tinctions between groups (Apfelbaum et al., 2008; Holoien & Shelton, 2012; Plaut et al., 2009; Rattan & Ambady, 2012; Rosenthal & Levy, 2013; Tadmor et al., 2012).

d Thinking of multiculturalism as an approach that is inclusive of everyone, includ- ing the majority group (Plaut et al., 2011).

How can these types of thinking be promoted? One important factor no doubt is education. More generally, it is at the cultural level that much potential for positive change can be found. Exposure to images and individuals that reflect the diversity within social groups, for example, can help weaken stereotypes and com- bat their automatic activation. These images might also change people’s tendency to see groups as relatively fixed entities and help them see groups as dynamic entities with less rigid borders.

80220_ch05_ptg01_hr_154-207.indd 204 27/01/16 5:19 PM

205 Reducing the Problem: Social Psychological Solutions

Motivations, norms, and values can and often do change over time. Here again popular culture is a key player. People—especially younger people—look to images in popular culture as well as to their peers and role models for information about what attitudes and behaviors are in fashion or out of date. Popular culture has played an important role in many of these changes. For example, many people cite celebrities such as Ellen DeGeneres or television shows like Modern Family for helping to change their attitudes toward gay rights (Lang, 2015). Similarly, seeing antiprejudice campaigns that promote ideas like “Black Lives Matter” or “It Gets Better” go viral around the world in response to violence against targeted groups no doubt affects many people’s attitudes.

We also look to our peers to get a sense of the local norms around us, including norms about stereotypes and prejudice (e.g., Fein et al., 2003; Nelson, 2015; Pryor et al., 2013). In a particular high school, a senior might feel comfortable calling a friend a “fag” and mean little by it and think nothing of it. Yet several months later as a first-semester college student, this student might realize how wrong that is and feel guilty for ever having done so. If this lesson is learned, it’s more likely to have been learned by watching and interacting with one’s peers than from hav- ing been lectured to about diversity and sensitivity by a campus speaker. Learn- ing these norms can motivate us to adopt them. Legislating against hate speech, unequal treatment, and hostile environments can also be an important weapon, of course. Although they can create resistance and backlashes, laws and policies that require behavior change can—if done right, with no suggestion of compromise and with important leaders clearly behind them—cause hearts and minds to follow (Aronson, 1992). Laura Barron and Mikki Hebl (2013), for example, recently found across a series of studies that laws that make employment discrimination against gays illegal (which many states in the United States do not currently have) do make a significant and positive difference in individuals’ attitudes and behaviors.

Elizabeth Levy Paluck (2011) has conducted some fascinating field studies show- ing the power of peer and cultural influences. In one study, students in several high schools were trained to be “peer leaders” who would confront expressions of inter- group prejudice. These students not only engaged in more antiprejudice behaviors, but these behaviors spread to their friends. Paluck (2009) also examined the role of the media in promoting positive norms in an unusual year-long field experiment in Rwanda, which has been the site of ter- rible war, genocide, and intergroup con- flict, particularly between the Hutus and Tutsis. She had Rwandans listen to a radio soap opera (radio being the most important form of mass media there) over the course of a year. She randomly assigned half to listen to a soap opera about conflicts that paralleled real conflicts in the country but that were solved in ways that modeled in- tergroup cooperation and communication, nonviolence, and opposition to prejudice. The other half of the participants listened to a soap opera about health issues. At the end of the year, those Rwandans who listened to the soap opera promoting positive intergroup norms had significantly more positive feelings about intergroup cooperation, trust, and interactions.

After a series of racist incidents, fans denounce racism at the World Cup in Brazil, in September 2014.

Ya su

yo sh

i C hi

ba /A

FP /G

et ty

Im ag

80220_ch05_ptg01_hr_154-207.indd 205 27/01/16 5:19 PM

206 Chapter 5 Stereotypes, Prejudice, and Discrimination

Studies like these show that a source for much positive change stems from what is at the very core of social psychology: the social nature of the human animal. Some of our baser instincts, such as intergroup competition that breeds intergroup biases, may always be present, but we also can learn from each other the thoughts, values, and goals that make us less vulnerable to perpetuating or being the targets of stereotypes, prejudice, and discrimination.

1. Prejudice and discrimination operate not only on an indi- vidual level, in which anyone can be both a perpetrator and a target, but also on institutional and cultural levels, which involve practices that promote the domination of one group over another.

2. Many stories in the news as well as data from numerous studies provide compelling evidence that racism, sexism, and other forms of prejudice and discrimination persist, but also that progress has been made.

3. Stereotypes, prejudice, and discrimination today often operate implicitly, without conscious intent or aware- ness on the part of perceivers, but with significant consequences.

4. Ambivalent sexism reflects both hostile sexism, charac- terized by negative and resentful feelings toward women, and benevolent sexism, characterized by affectionate, chivalrous, but potentially patronizing feelings toward women.

5. There are some striking sex differences in occupational choices and in the treatment individuals experience in the workplace.

6. Stereotype threat can impair the performance and af- fect the identity of members of stereotyped or devalued groups. Slight changes in a setting can reduce stereotype threat and its negative effects significantly.

7. People tend to exaggerate the differences between in- groups and outgroups, treat ingroup members more favorably (especially when feeling threatened or uncer- tain), and see outgroup members as more homogenous and often in dehumanizing ways.

8. We learn information relevant to stereotypes, prejudice, and discrimination without even realizing it by absorb- ing what we see around us in our culture, groups, and families.

9. A fundamental effect of stereotyping is that it influ- ences people’s perceptions and interpretations of the behaviors of group members, causing them to perceive confirmation of their stereotype-based expectancies.

10. Among the factors that have been found to be success- ful to reduce stereotyping, prejudice, and discrimina- tion are: having intergroup contact (especially if the ideal conditions for contact are met), perceiving con- nections and a shared identity with people from other groups, adopting a multicultural rather than colorblind perspective, and being intrinsically (rather than extrin- sically) motivated to reduce prejudiced responses.

Top 10 Key Points in Chapter 5

Review

80220_ch05_ptg01_hr_154-207.indd 206 27/01/16 5:19 PM

207 Review

The Nature of the Problem: Persistence and Change

Children do not tend to show biases based on race; it is only after they become adolescents that they learn to respond to people differently based on race. F False. Children learn about social categories quite

early and use stereotypes when they are very young. Children show biases in favor of their racial ingroup on both explicit and implicit measures.

interracial interactions tend to go better and to reduce the perceptions of racism if a colorblind mentality is used, which denies or minimizes any acknowledgment of racial differences. F False. Research has shown that this approach often

backfires and makes members of racial minority groups more, rather than less, uncomfortable; a multicultural approach that acknowledges and positively values racial and ethnic differences is often more effective.

are likely to experience stereotype threat and therefore

underperform if the task is described as one that is diagnostic of their sports intelligence. White students tend to show the opposite effect: Their performance is worse if the task is described as reflecting natural athletic ability.

Causes of the Problem: intergroup, Motivational, Cognitive, and Cultural Factors

Being reminded of one’s own mortality makes people put things into greater perspective, thereby tending to reduce ingroup–outgroup distinctions and hostilities. F False. Research has shown that when people feel

threatened by thoughts of their own mortality, they tend to seek greater affiliation with their ingroups and exhibit greater prejudice against outgroups, in part to reaffirm their sense of place and purpose in the world.

People’s very quick judgments are not influenced by a stereotype unless they actually believe the stereotype to be true.

F False. Even very brief exposure to a member of a stereotyped group can activate the stereotype about the group, even if the person does not believe the stereotype.

Putting Common Sense to the Test

ambivalent sexism (166) aversive racism (160) contact hypothesis (196) discrimination (158) illusory correlation (191) implicit racism (162) ingroup favoritism (183) ingroups (177) jigsaw classroom (199) modern racism (160)

outgroup homogeneity effect (177)

outgroups (177) prejudice (158) racism (157) realistic conflict theory (182) relative deprivation (182) sexism (158) social categorization (176) social dominance orientation (180)

social identity theory (183) social role theory (189) stereotype content model (181) stereotypes (158) stereotype threat (172) stigmatized (171) subliminal presentation (193) superordinate goal (182) system justification theory (180)

Key Terms

80220_ch05_ptg01_hr_154-207.indd 207 27/01/16 5:19 PM

Attitudes

This chapter examines social influences on attitudes. We define attitudes and then discuss how they are measured and when they are related to behavior. Then we consider two methods of changing attitudes. First, we look at source, message, and audience factors that win persuasion through the media of communication. Second, we consider theories and research showing that people often change their attitudes as a consequence of their own actions.

80220_ch06_ptg01_hr_208-261.indd 208 27/01/16 5:20 PM

Attitudes 6 The Study of Attitudes | 210

How Attitudes Are Measured

How Attitudes Are Formed

The Link Between Attitudes and Behavior

Persuasion by Communication | 223

Two Routes to Persuasion

The Source

The Message

The Audience

Culture and Persuasion

Persuasion by Our Own Actions | 245

Role Playing: All the World’s a Stage

Cognitive Dissonance Theory: The Classic Version

Cognitive Dissonance Theory: A New Look

Alternative Routes to Self-Persuasion

Ethical Dissonance

Cultural Influences on Cognitive Dissonance

Changing Attitudes | 259 Review

209ad am

ka z/

G et

ty Im

ag es

80220_ch06_ptg01_hr_208-261.indd 209 27/01/16 5:20 PM

210 Chapter 6 Attitudes

ISIS. Taxes. Same-sex marriage. The death penalty. Gun control. Is- raelis and Palestinians. Abortion. Climate change. Immigration. Anyone who has followed recent events in the United States—or anywhere else in the world, for that matter—knows how passionately people feel about the issues of the day. Attitudes and the mechanisms of attitude change, or persuasion, are a vital part of human social life. This chapter addresses three sets of questions: (1) What is an attitude, how can it be measured, and what is its link to behavior? (2) What kinds of persuasive messages lead people to change their attitudes? (3) Why do we often change our attitudes as a result of our own actions?

The Study of Attitudes Are you a Democrat, Republican, or Independent? Should marijuana be legalized? Would you rather listen to alternative rock, country-western, or hip-hop? Do you prefer drinking Coke or Pepsi, water, or fruit juice? Do you have an iPhone or an Android? Should terrorism be contained by war or conciliation?

As these questions suggest, each of us has positive and negative reactions to various persons, objects, and ideas. These reactions are called attitudes. Skim the chapters in this book, and you’ll see just how pervasive attitudes are. You’ll see, for example, that self-esteem is an attitude we hold about ourselves, that attrac- tion is a positive attitude toward another person, and that prejudice is a negative attitude often directed against certain groups. Indeed, the study of attitudes—

what they are, where they come from, how they can be measured, what causes them to change, and how they interact with behavior—is central to the whole field of social psychology (Bohner & Dickel, 2011; Crano & Prislin, 2014; Perloff, 2010; Maio & Haddock, 2015).

An attitude is a positive, negative, or mixed eval- uation of an object that is expressed at some level of intensity—nothing more, nothing less. Like, love, dislike, hate, admire, and detest are the kinds of words that peo- ple use to describe their attitudes. It’s important to real- ize that attitudes cannot simply be represented along a single continuum ranging from wholly positive to wholly negative—as you might expect if attitudes were like the volume button on a remote control unit or the lever on a thermostat that raises or lowers temperature. Rather, as depicted in d Figure 6.1, our attitudes can vary in strength along both positive and negative dimensions. In other words, we can react to something with positive af- fect, with negative affect, with ambivalence (mixed emo- tions), or with apathy and indifference (Cacioppo et al., 1997). Some people more than others are troubled by this type of inconsistency (Newby-Clark et al., 2002). In fact, at times people have both positive and negative reactions

to the same attitude object without feeling conflict because they are conscious of one reaction but not the other. Someone who is welcoming toward racial minori- ties but harbors unconscious prejudice is a case in point (Wilson et al., 2000).

Each and every one of us routinely forms positive and negative evaluations of the people, places, objects, and ideas we encounter. We like some things but not

Putting Common SenSe to the Test

T F Researchers can tell if someone has a positive or negative attitude by measuring physiological arousal.

T F In reacting to persuasive communications, people are influenced more by superficial images than by logical arguments.

T F People are most easily persuaded by commercial messages that are presented without their awareness.

T F The more money you pay people to tell a lie, the more they will come to believe it.

T F People often come to like what they suffer for.

Circle Your Answer

attitude A positive, negative, or mixed reaction to a person, object, or idea.

80220_ch06_ptg01_hr_208-261.indd 210 27/01/16 5:20 PM

211 The Study of Attitudes

others. This attitude formation process is often quick, automatic, and “implicit”—much like a reflex action (Bargh et al., 1996; De Houwer, 2014; Ferguson, 2007).

You might assume that a person’s attitude represents a unique relation between that person and a specific attitude object. There are two ways, however, in which our attitudes reveal a lot about us as individuals. First, people differ in terms of their tendency in general to like or dislike things. Consider an array of very different and unrelated attitude objects: How do you feel about bicycles? What about crossword puzzles, camping, Japan, taxes, politics, architecture, chess, statistics, Hulu, Netflix, religion, and bottled water? Curious as to whether people have tendencies in general to like or dislike things, which they called dispositional attitudes, Justin Hepler and Dolores Albarracín (2013) found that when they asked research participants to rate how much they liked or disliked a long list of unrelated things, some individuals on average tended to report positive attitudes; others on average were negative (also see Eschleman et al., 2015).

A second way in which our attitudes reveal something about us as individuals is that people differ not only in terms of whether they tend to like or dislike things but in the extent to which how quickly and how strongly they react. Think about yourself. Do you form opinions easily? Do you tend to have strong likes and dislikes? Or do you tend to react in more guarded, less effusive ways? Individu- als who describe themselves as high rather than low in the need for evaluation are more likely to view their daily experiences in highly judgmental terms and they are more opinionated—positive and negative—on a whole range of social, moral, and political issues (Bizer et al., 2004; Jarvis & Petty, 1996).

Before we examine the elusive science of attitude measurement, let’s stop for a moment and ponder this question: Why do human beings bother to form and have attitudes? Does forming a positive or negative judgment of people, objects, and ideas serve any useful purpose? Over the years, researchers have found that attitudes serve important functions—such as enabling us to judge quickly and without much thought whether something we first encounter is good or bad, help- ful or hurtful, and to be sought or avoided (Maio & Olson, 2000). The downside is that having preexisting attitudes can lead us to be closed-minded, bias how we interpret new information, and make us more resistant to change. For example, Russell Fazio and others (2000) found that people who were focused on their posi- tive or negative attitudes toward computerized faces, compared to those who were not, were later slower to notice when the faces were “morphed” and no longer the same.

j How Attitudes Are Measured In 1928, Louis Thurstone published an article entitled “Attitudes Can Be Measured.” What Thurstone failed to fully anticipate, however, is that attitude measurement is tricky business. At one point, several years ago, over 500 different methods were available to determine an individual’s attitudes (Fishbein & Ajzen, 1972).

Self-Report Measures The easiest way to assess a person’s attitude about something is to ask. All over the world, public opinions are recorded on a range of issues in politics, the economy, health care, foreign affairs, science and technology,

d F i g u R e 6 . 1

Four Possible Reactions to Attitude Objects

As shown, people evaluate objects along both positive and negative dimensions. As a result, our attitudes can be positive, negative, ambivalent, or indifferent. Cacioppo et al., 1997.

Low

Lo w

High

H ig

Po si

tiv e

re ac

tio n

Negative reaction

Dual attitudes (ambivalence)

Indifference

Positive attitude

Negative attitude

80220_ch06_ptg01_hr_208-261.indd 211 27/01/16 5:20 PM

212 Chapter 6 Attitudes

sports, entertainment, religion, and lifestyles. Simply by asking people, public opinion surveys conducted by Harris, Gallup, Pew Research Center, and other polling organizations have recently revealed that 33% of Ameri- cans will never consider buying or leasing a self-driving car; 82% believe that childhood vaccinations should be mandatory; 74% would delete them- selves from Internet search results if they could; 70% do not “feel engaged or inspired at their jobs;” and 88% trust online reviews of products and services as much as personal recom- mendations. Still other surveys have shown that Americans prefer to watch football over baseball and like to eat

chocolate ice cream more than vanilla and other flavors; that more than half say they surf the Internet while watching TV; that one in five adults has a tattoo; and that Christmas is America’s favorite holiday, followed by Thanksgiving and Halloween.

Self-report measures are direct and straightforward. But attitudes are sometimes too complex to be measured by a single question. As you may recall from Chapter 2, one problem recognized by public opinion pollsters is that responses to attitude questions can be influenced by their wording, the order and context in which they are asked, and other extraneous factors (Sudman et al., 2010; Tourangeau et al., 2000). In one survey, the National Opinion Research Center asked hundreds of Americans if the U.S. government spent too little money on “assistance to the poor” and 65% said yes. Yet when the same question was asked using the word “welfare” instead, only 20% said the government spent too little (Schneiderman, 2008). In a second survey, researchers asked more than 2,000 registered voters about their belief in the phenomenon of “global warming” or “climate change.” Democrats uniformly endorsed the proposition at a high rate, but the number of Republicans who did so increased from 44% when asked about global warming to 60% when asked about climate change (Schuldt et al., 2011).

Recognizing the shortcomings of single-question measures, survey research- ers have developed more sophisticated methods (Fowler, 2014). Often single questions are replaced by multiple-item questionnaires known as attitude scales (Robinson et al., 1991; 1998). Attitude scales come in different forms, perhaps the most popular being the Likert Scale, named after its inventor, Rensis Likert (1932). In this technique, respondents are presented with a list of statements about an attitude object and are asked to indicate on a multiple-point scale how strongly they agree or disagree with each statement. Each respondent’s total attitude score is derived by summing responses to all the items. However, regardless of whether attitudes are measured by one question or a full-blown scale, the results should be taken with caution. All self-report measures assume that people honestly ex- press their true opinions. Sometimes this assumption is reasonable and correct, but often it is not. Wanting to make a good impression, people are often reluctant to admit to their failures, vices, weaknesses, unpopular opinions, and prejudices.

One approach to this problem is to increase the accuracy of self-report mea- sures. To get respondents to answer attitude questions more truthfully, researchers have sometimes used the bogus pipeline, an elaborate mechanical device that

As in this January 7, 2014 unity rally in Paris, where 3 million gathered to pay respect for the victims of a terrorist attack on the Charlie Hebdo magazine, people can be very passionate about the attitudes they hold. Their message here: “Not Afraid.” On November 13, 2015, not quite two years later, Paris was once again rocked by a cluster of heinous terrorist attacks.

A P

Im ag

es /T

hi ba

ul t C

am us

To view new public opinion poll results online, you can visit: http://www. harrisinteractive.com/, http://www. gallup.com/home.aspx, and http:// www.pewresearch.org/.

attitude scale A multiple-item questionnaire designed to measure a person’s attitude toward some object.

bogus pipeline A phony lie- detector device that is sometimes used to get respondents to give truthful answers to sensitive questions.

80220_ch06_ptg01_hr_208-261.indd 212 27/01/16 5:20 PM

213 The Study of Attitudes

supposedly records our true feelings physiologically, like a lie- detector test. Not wanting to get caught in a lie, respondents tend to answer attitude questions more honestly, and with less positive spin, when they think that any deception would be exposed by the bogus pipeline (Jones & Sigall, 1971; Roese & Jamieson, 1993). In one study, people were more likely to admit to drinking too much, using cocaine, having frequent oral sex, and not exercising enough when the bogus pipeline was used than when it was not (Tourangeau et al., 1997). In another study, adolescents were more likely to admit to smok- ing when the bogus pipeline was used than when it was not (Adams et al., 2008).

Covert Measures A second general approach to the self- report problem is to collect indirect, covert measures of attitudes that cannot be controlled. One possibility in this regard is to use observable behavior such as facial expressions, tone of voice, and body language. In one study, Gary Wells and Richard Petty (1980) secretly videotaped college students as they listened to a speech and noticed that when the speaker took a position that the students agreed with (that tuition costs should be lowered), most made vertical head movements. But when the speaker took a contrary position (that tuition costs should be raised), head movements were in a horizontal di- rection. Without realizing it, the students had signaled their attitudes by nodding and shaking their heads.

Although behavior provides clues, it is far from perfect as a measure of attitudes. Sometimes we nod our heads be- cause we agree; at other times, we nod to be polite. The problem is that people monitor their overt behavior just as they monitor self-reports. But what about internal physiolog- ical reactions that are difficult if not impossible to control? Does the body betray how we feel? In the past, researchers tried to divine attitudes from involuntary physical reactions such as perspiration, heart rate, and pupil dilation. The re- sult, however, was always the same: Measures of arousal reveal the intensity of one’s attitude toward an object but not whether that attitude itself is positive or negative. On the physiological record, love and hate look very much the same (Petty & Cacioppo, 1983).

Although physiological arousal measures cannot distin- guish between positive and negative attitudes, some inter- esting alternatives have been discovered. One is the facial electromyograph (eMg). As shown in d Figure 6.2, certain muscles in the face contract when we are happy and differ- ent facial muscles contract when we are sad. Some of the muscular changes cannot be seen with the naked eye, how- ever, so the facial EMG is used. To determine whether the EMG can be used to measure the affect associated with at- titudes, John Cacioppo and Richard Petty (1981) recorded facial muscle activity of college students as they listened to a message with which they agreed or disagreed. The agreeable message increased activity in the cheek muscles—the facial pattern that is characteristic of happiness. The disagreeable

Pollsters and attitude researchers are well aware that self-reports cannot always be trusted, especially on sensitive topics.

“What I drink and what I tell the pollsters I drink are two different things.”

Le o

Cu llu

m /T

he N

ew Y

or ke

r C ol

le ct

io n/

Ca rt

oo nb

an k.

co m

d F i g u R e 6 . 2

The Facial eMg: A Covert Measure of Attitudes?

The facial EMG makes it possible to detect differences between positive and negative attitudes. Notice the major facial muscles and recording sites for electrodes. When people hear a message with which they agree rather than disagree, there is a relative increase in EMG activity in the depressor and zygomatic muscles but a relative decrease in corrugator and frontalis muscles. These changes cannot be seen with the naked eye. From Cacioppo, J. T. and Petty, R. E., “Electromyograms as mea- sures of extent and affectivity of information processing,” American Psychologist vol 36 (pp. 441–456). Copyright © 1981 by the Ameri- can Psychological Association. Reprinted by permission.

Frontalis

Corrugator

Zygomatic

Depressor

80220_ch06_ptg01_hr_208-261.indd 213 27/01/16 5:20 PM

214 Chapter 6 Attitudes

message sparked activity in the forehead and brow area—the facial patterns that are associated with sadness and distress. Outside observers who later watched the participants were unable to see these subtle changes. Apparently, the muscles in the human face reveal smiles, frowns, feelings of disgust, and other reactions to attitude objects that might otherwise be hidden from view (Cacioppo et al., 1986; Tassinary & Cacioppo, 1992).

From a social neuroscience perspective, electrical activity in the brain may also assist in the measure of attitudes. In 1929, Hans Burger invented a machine that could detect, amplify, and record “waves” of electrical activity in the brain us- ing electrodes pasted to the surface of the scalp. The instrument is called an elec- troencephalograph, or EEG, and the information it provides takes the form of line tracings called brain waves. Based on an earlier discovery that certain patterns of electrical brain activity are triggered by exposure to stimuli that are novel or unexpected, Cacioppo and others (1993) had participants list 10 items they liked and 10 they did not like within various object categories (fruits, sports, movies, universities, etc.). Later, these participants were brought into the laboratory, wired to an EEG, and presented with a list of category words that depicted objects they liked and disliked. The result: Brain-wave patterns that are normally triggered by inconsistency increased more when a disliked stimulus appeared after a string of positive items or when a liked stimulus was shown after a string of negative items than when either stimulus evoked the same attitude as the items that preceded it.

Today, social psychologists are also starting to use new forms of brain imaging in the measurement of attitudes. In one study, researchers used fMRI to record brain activity in participants as they read names of famous—and infamous— figures such as John F. Kennedy, Bill Cosby, and Adolf Hitler. When the names were read, they observed in participants greater activity in the amygdala, a brain structure associated with emotion, regardless of whether or not participants were asked to evaluate the famous figures (Cunningham et al., 2003). In a study focused on political attitudes, other researchers used fMRI to record brain activity in opinionated men during the 2004 presidential election as they listened to posi- tive and negative statements about their preferred candidate. Although brain areas associated with cognitive reasoning were unaffected during these presentations, activity increased in areas typically associated with emotion (Westen et al., 2006). These studies suggest that people react automatically to positive and negative attitude objects and that these attitudes may be measurable by electrical activity in the brain.

The implicit Association Test (iAT) When it comes to covert measurement, one particularly interesting development is based on the notion that each of us has all sorts of implicit attitudes that we cannot self-report in questionnaires because we are not even aware of having these attitudes (Fazio & Olson, 2003). To measure these unconscious attitudes, a number of indirect methods have been developed (De Houwer et al., 2009; Nosek et al., 2011; Payne & Lundberg, 2014). The most popular is the implicit Association Test (iAT) created by Anthony Greenwald, Mahzarin Banaji, Brian Nosek, and others. As we saw in Chapter 5, the IAT measures the sheer speed—in fractions of a second—with which people associate pairs of concepts (Greenwald et al., 1998). To see how it works, visit the IAT website by logging onto Project Implicit at https://implicit.harvard.edu/ implicit/ or installing a free IAT app onto your Android or iPhone. On the first page, you will read that “Project Implicit investigates thoughts and feelings that exist outside of conscious awareness or conscious control.”

Researchers can tell if someone has a positive or negative attitude by measuring physiological arousal.

FAlSe

facial electromyograph (eMg) An electronic instrument that records facial muscle activity associated with emotions and attitudes.

implicit attitude An attitude, such as prejudice, that one is not aware of having.

implicit Association Test (iAT) A covert measure of unconscious attitudes derived from the speed at which people respond to pairings of concepts—such as black or white with good or bad.

80220_ch06_ptg01_hr_208-261.indd 214 27/01/16 5:20 PM

215 The Study of Attitudes

To take a test that measures your implicit racial attitudes, you go through a series of stages. First, you are asked to categorize black or white faces as quickly as you can, for example, by pressing a left-hand key in response to a black face and a right-hand key for a white face. Next, you are asked to categorize a set of words, for example, by pressing a left-hand key for positive words (love, laughter, friend) and a right-hand key for negative words (war, failure, evil). Once you become familiar with the categorization task, the test combines faces and words. You may be asked, for example, to press the left-hand key if you see a black face or positive word and a right-hand key for a white face or negative word. Then, in the fourth stage, the opposite pairings are presented—black or negative, white or positive. Black and white faces are then interspersed in a quick sequence of trials, each time paired with a positive or negative word. In rapid-fire succes- sion, you have to press one key or another in response to stimulus pairs such as black-wonderful, black-failure, white-love, black-laughter, white-evil, white-awful, black-war, and white-joy. As you work through the list, you may find that some pairings are harder and take longer to respond to than others. In general, people are quicker to respond when liked faces are paired with positive words and dis- liked faces are paired with negative words than the other way around. Using the IAT, your implicit attitudes about African Americans can thus be detected by the speed it takes you to respond to black-bad/white-good pairings relative to black- good/white-bad pairings. The test takes only 10–15 minutes to complete. When you’re done, you receive the results of your test and an explanation of what it means (see d Figure 6.3).

d F i g u R e 6 . 3

The implicit Association Test

Through a sequence of tasks, the IAT measures implicit racial attitudes toward, for example, African Americans, by measuring how quickly people respond to black-bad/ white-good word pairings relative to black-good/white-bad pairings. Most white Americans are quicker to respond to the first type of pairings than to the second, which suggests that they do not as readily connect black-good and white-bad. From Kassin, S. Essentials of psychology. Copyright © 2004. Reprinted by permission of Pearson Education, Inc., Upper Saddle River, NJ.

White or

bad

Black or

good

Black or

bad

White or

good

80220_ch06_ptg01_hr_208-261.indd 215 27/01/16 5:20 PM

216 Chapter 6 Attitudes

From 1998 to the present, visitors to the IAT website completed millions of tests. In questionnaires, interviews, public opinion polls, and Internet surveys, people don’t tend to express stereotypes, prejudices, or other unpopular attitudes. Yet on the IAT, respondents have exhibited a marked implicit preference for self over other, white over black, young over old, straight over gay, able over disabled, thin over obese, and the stereotype that links males with careers and females with family (Greenwald et al., 2003; Nosek et al., 2002). Because more and more researchers are using these kinds of indirect measures, social psychologists who study attitudes find themselves in the midst of a debate over what IAT scores mean, how the implicit attitudes revealed in the IAT are formed and then changed, how well these attitudes predict or influence societally important behaviors, and how they differ from the more explicit attitudes that we consciously hold and report (Blanton et al., 2009; Gawronski & Bodenhausen, 2006; Greenwald et al., 2015; Oswald et al., 2015; Petty et al., 2009).

Do implicit attitudes matter? Do millisecond differences in response times on a computerized test really predict behavior in real-world settings of consequence? And what does it mean when one’s implicit and explicit attitudes clash? The im- portance of these questions cannot be overstated. If the IAT reveals unconscious prejudices that people do not self-report, should individuals be scrutinized in the laboratory for hidden motives underlying potentially unlawful behaviors—as when a police officer shoots a black suspect, fearing that he or she is armed; as when an employer hires a male applicant over a female applicant, citing his credentials as opposed to an act of discrimination; or as when a jury chooses to convict a Latino defendant on the basis of ambiguous evidence?

Eager to use the IAT for predicting social behaviors of consequence, some re- searchers have speculated about the relevance of implicit attitudes in the domains of law (Lane et al, 2007) and politics (Gawronski et al., 2015). But is their specu- lation justified? Some researchers are critical of claims concerning the predictive validity of the IAT, citing the need for more and stronger behavioral evidence (Blanton et al., 2009; Oswald et al., 2015). Based on a meta-analysis of 122 IAT studies involving 15,000 participants, Greenwald and others (2009) conceded that people’s implicit attitudes are generally less predictive of behavior than their ex- plicit attitudes. They also found, however, that IAT measures are better when it comes to socially sensitive topics for which people often conceal or distort their self-reports. In a poignant illustration of this point, one research team adminis- tered to a large group of psychiatric patients an IAT that measured their implicit associations between self and suicide. Over the next six months, patients appear- ing in the emergency room because of a suicide attempt had a stronger implicit association between self and suicide than those who appeared with other types of psychiatric emergencies (Nock et al., 2010).

j How Attitudes Are Formed How did you become liberal or conservative in your political values? Why do you favor or oppose same-sex marriage? What draws you toward or away from orga- nized religion?

Are Attitudes inherited? One hypothesis, first advanced by Abraham Tesser (1993), is that strong likes and dislikes are rooted in our genetic makeup. Research shows that on some issues the attitudes of identical twins are more similar than those of fraternal twins and that twins raised apart are as similar to each other as those who are raised in the same home. This pattern of evidence suggests that

80220_ch06_ptg01_hr_208-261.indd 216 27/01/16 5:20 PM

217 The Study of Attitudes

people may be predisposed to hold certain attitudes. Indeed, Tesser found that when asked about attitudes for which there seems to be a predisposition (such as attitudes toward sexual promiscuity, religion, and the death penalty), research participants were quicker to respond and less likely to alter their views toward social norms.

Tesser speculated that individuals are disposed to hold certain strong attitudes as a result of inborn physical, sensory, and cognitive skills, temperament, and personality traits. Other twin studies, too, as well as complex comparisons within extended families, have supported the notion that people differ in their attitudes toward a range of social and political issues in part because of genetically rooted differences in their biological makeup (Hatemi et al., 2010; Olson et al., 2001). There are different ways in which a person’s inborn tendencies may influence social and political attitudes. In one study, for example, researchers brought 40 adults with strong political views into the lab for testing and found that those who physiologically were highly reactive to sudden noise and other unpleasant stimuli were more likely to favor capital punishment, the right to bear arms, de- fense spending, and other policies seen as protective against domestic and foreign threats (Oxley et al., 2008).

Are Attitudes learned? Whatever dispositions nature provides to us, our most cherished attitudes often form as a result of our exposure to attitude objects; our history of rewards and punishments; the attitudes that our parents, friends, and enemies express; the social and cultural context in which we live; and other types of experiences. In a classic naturalistic study, Theodore Newcomb (1943) surveyed the political attitudes of students at Bennington College in Vermont. At the time, Bennington was a women’s college that drew its students from conservative and mostly affluent families. Once there, however, the students encountered professors and older peers who held more liberal views. Newcomb found that as the women moved from their first year to graduation, they became progressively more liberal.

Chances are, these identical twins have more in common than being truck driving partners. Research suggests that people may also be genetically predisposed to hold certain attitudes.

Jo di

C ob

b/ N

at io

na l G

eo gr

ap hi

c Im

ag e

Co lle

ct io

n/ A

la m

80220_ch06_ptg01_hr_208-261.indd 217 27/01/16 5:20 PM

218 Chapter 6 Attitudes

(In the 1936 presidential election, 62% of first-year Bennington students preferred the Republican Landon to the Democrat Roosevelt, compared to only 43% of sopho- mores and 15% of juniors and seniors.) This link between cultural environment and attitudes is particularly evident in the current political landscape of America—a “house divided” into red states and blue states by geography, culture, and ideology (Seyle & Newman, 2006).

Clearly, attitudes are formed through basic processes of learning. For example, a number of studies have shown that people can form strong positive and negative attitudes toward neutral objects that somehow are linked to emotionally charged stimuli. At the start of the twentieth century, Russian physiologist Ivan Pavlov (1927) discovered that dogs would naturally and reflexively salivate in response to food in the mouth. He then discovered that by repeatedly ringing a bell—a neutral stimulus—before the food was placed in the mouth, the dog would eventually start to salivate at the sound of the bell itself. This process by which organisms learn to associate a once neutral stimulus with an inherently positive or negative response is a basic and powerful form of learning. In fact, Pavlov found that once a dog was conditioned to salivate to one tone, it went to “generalize” by respond- ing to other sounds that were similar but not identical.

It is now clear that this form of learning can help to explain the develop- ment of social attitudes. In a classic first study, college students were presented with a list of adjectives that indicate nationality (German, Swedish, Dutch, Ital- ian, French, and Greek), each of which was repeatedly presented with words that were known to have very pleasant (happy, gift, sacred) or unpleasant (bitter, ugly, failure) connotations. When the participants later evaluated the nationalities by name, they were more positive in their ratings of those that had been paired with pleasant words than with unpleasant words (Staats & Staats, 1958). And as discovered by Pavlov nearly a hundred years ago, research shows that when an attitude is changed toward one object, attitudes toward similar and related objects are often changed as well (Glaser et al., 2015).

More recent studies of evaluative conditioning have shown that implicit and explicit attitudes toward neutral objects can form by their association with posi- tive and negative stimuli, even in people who are not conscious of this association (Hofmann et al., 2010; Olson & Fazio, 2001; Sweldens et al., 2014; Walther et al., 2011). That’s why political leaders all over the world wrap themselves in a na- tional flag to derive a benefit from positive associations, while advertisers strategi- cally pair their products with sexy models, uplifting music, celebrities, nostalgic images, and other positive emotional symbols. In a series of laboratory studies, researchers found that people came to prefer brands of a consumer product that were paired with humorous ads more than those that were associated with non- humorous ads (Strick et al., 2009).

j The link Between Attitudes and Behavior People take for granted the notion that attitudes influence behavior. We assume that voters’ opinions of opposing candidates predict the decisions they will make on Election Day, that consumer preference for one product over competing prod- ucts will influence the purchases they make, and that feelings of prejudice will trigger bad acts of discrimination. Yet as sensible as these assumptions seem, the link between attitudes and behavior has proved far from perfect.

Sociologist Richard LaPiere (1934) was the first to notice that attitudes and be- havior don’t always go hand in hand. In the 1930s, during the height of the Great Depression, LaPiere took a young Chinese American couple on a three-month,

evaluative conditioning The process by which we form an attitude toward a neutral stimulus because of its association with a positive or negative person, place, or thing.

80220_ch06_ptg01_hr_208-261.indd 218 27/01/16 5:20 PM

219 The Study of Attitudes

10,000-mile car trip, visiting 250 restaurants, campgrounds, and hotels across the United States. Although prejudice against Asians was widespread at the time, the couple was refused service only once. Yet when LaPiere wrote back to the places they had visited and asked if they would accept Chinese patrons, more than 90% of those who returned an answer said they would not. Self-reported attitudes did not correspond with behavior.

This study was provocative but seriously flawed. LaPiere measured attitudes several months after his trip, and during that time the attitudes may have changed. He also did not know whether those who responded to his letter were the same people who had greeted the couple in person. It was even possible that the Chi- nese couple were served wherever they went because they were accompanied by LaPiere—or because businesses were desperate during hard economic times.

Despite these limitations, LaPiere’s study was the first of many to reveal a lack of correspondence between attitudes and behavior. In 1969, Allan Wicker reviewed the applicable research and concluded that attitudes and behavior are correlated only weakly, if at all. Sobered by this conclusion, researchers were puz- zled: Could it be that the votes we cast do not follow from our political opinions, that consumers’ purchases are not based on their attitudes toward a product, or that discrimination is not related to underlying prejudice? Is the study of attitudes useless to those interested in human social behavior? Not at all. During subse- quent years, researchers went on to identify the conditions under which attitudes and behavior are correlated. Thus, when Stephen Kraus (1995) meta-analyzed all of this research, he concluded that “attitudes significantly and substantially predict future behavior” (p. 58). In fact, he calculated that there would have to be 60,983 new studies reporting a zero correlation before this conclusion would have to be revised. Based on a meta-analysis of 41 additional studies, Laura Glasman and Dolores Albarracín (2006) went on to identify some of the conditions under which attitudes most clearly predict future behavior.

Attitudes in Context One important condition is the level of correspondence, or similarity, between attitude measures and behavior. Perhaps the reason that LaPiere (1934) did not find a correlation between self-reported prejudice and discrimination was that he had asked proprietors about Asians in general but then observed their actions toward only one couple. To predict a single act of discrimination, he should have measured people’s more specific attitudes toward a young, well-dressed, attrac- tive Chinese couple accompanied by an American professor.

Icek Ajzen and Martin Fishbein (1977) analyzed more than 100 studies and found that attitudes correlate with behavior only when attitude measures closely match the behavior in question. Illustrating the point, Andrew Davidson and James Jaccard (1979) tried to use attitudes to predict whether women would use birth control pills within the next two years. Attitudes were measured in a series of questions ranging from very general (“How do you feel about birth control?”) to very specific (“How do you feel about using birth control pills during the next two years?”). The more specific the initial attitude question was, the better it predicted the behavior. Other researchers as well have replicated this finding (Kraus, 1995).

The link between our feelings and our actions should also be placed within a broader context. Attitudes are one determinant of social behavior, but there are other determinants as well. This limitation formed the basis for Fishbein’s (1980) theory of reasoned action, which Ajzen (1991) later expanded into his theory of planned behavior. According to these theories, our attitudes influence our behav- ior through a process of deliberate decision making, and their impact is limited in four respects (see d Figure 6.4).

theory of planned behavior The theory that attitudes toward a specific behavior combine with subjective norms and perceived control to influence a person’s actions.

80220_ch06_ptg01_hr_208-261.indd 219 27/01/16 5:20 PM

220 Chapter 6 Attitudes

First, as just described, behavior is influenced less by general attitudes than by attitudes toward a specific behavior. Second, behavior is influenced not only by attitudes but by subjective norms— our beliefs about what others think we should do. As we’ll see in Chapter 7, so- cial pressures to conform often lead us to behave in ways that are at odds with our inner convictions. Third, according to Ajzen, attitudes give rise to behavior only when we perceive the behavior to be within our control. To the extent that people lack confidence in their ability to engage in some behavior, they are unlikely to form an intention to do so. Fourth, although attitudes (along with subjective norms and perceived control) contribute to an intention to behave in a particular manner, people often do not or cannot follow through on their intentions.

A good deal of research now sup- ports the theories of reasoned action and

planned behavior (Ajzen & Fishbein, 2005). Indeed, this general approach, which places the link between attitudes and behaviors within a broader context, has successfully been used to predict a wide range of practical behaviors—such as using condoms, obeying speed limits, washing hands and other food safety hab- its, donating blood, complying with medical regimens, and reducing risky sexual behaviors (Albarracín et al., 2001; Conner et al., 2013; Elliott et al., 2003; Milton & Mullan, 2012; Rich et al., 2015; Tyson et al., 2014).

Strength of the Attitude According to the theories of reasoned action and planned behavior, specific attitudes combine with social factors to produce behavior. Sometimes attitudes have more influence on behavior than do the other factors; some- times they have less influence. In large part, it depends on the importance, or strength, of the attitude. Each of us has some views that are nearer and dearer to the heart than others. Computer jocks become attached to PCs or to Macs, religious fundamentalists care deeply about issues pertaining to life and death, and political activists have fiery passions for one political party or policy over others. In each case, the attitude is held with great confidence and is difficult to change (Petty & Krosnick, 1995).

Why are some attitudes stronger than others? David Boninger and others (1995) identified three psychological factors that consistently seem to distinguish between our strongest and weakest attitudes. These investigators asked people to reflect on their views toward defense spending, gun control, the legalization of marijuana, abortion rights, and other issues. They found that the attitudes people held most passionately were those that concerned issues that (1) directly affected their own self-interests; (2) related to deeply held philosophical, political, and religious values; and (3) were of concern to their close friends, family, and social ingroups. This last, highly social, point is important. Research shows that when people are surrounded by others who are like-minded, the attitudes they hold are stronger and more resistant to change (Visser & Mirabile, 2004).

d F i g u R e 6 . 4

Theory of Planned Behavior

According to the theory of planned behavior, attitudes toward a specific behavior combine with subjective norms and perceived behavior control to influence a person’s intentions. These intentions, in turn, guide but do not completely determine behavior. This theory places the link between attitudes and behavior within a broader context. From Ajzen, Organizational Behavior and the Human Decision Process, vol 50, pp. 179–211. Copyright © 1991 Elsevier. Reprinted with permission.

Attitude toward a behavior

Perceived behavior control

Subjective norm Intention Behavior

80220_ch06_ptg01_hr_208-261.indd 220 27/01/16 5:20 PM

221 The Study of Attitudes

Several factors indicate the strength of an attitude and its link to behavior. One is that people tend to behave in ways that are consistent with their attitudes when they are well informed. For example, college students were asked which of two candidates they preferred in an upcoming local election for mayor. Those who knew the factual campaign issues were later the most likely to actually vote for their favored candidate (Davidson et al., 1985). In another study, col- lege students were questioned about their views on various environmental issues and later were asked to take action—to sign petitions, participate in a recycling project, and so on. Again, the more informed the students were, the more con- sistent their attitudes about the environment were with their behavior (Kallgren & Wood, 1986).

The strength of an attitude is indicated not only by the amount of information on which it is based but also by how that information was acquired. Research shows that attitudes are more stable and more predictive of behavior when they are born of direct personal experience than when based on indirect, secondhand information. In a series of experiments, for example, Russell Fazio and Mark Zanna (1981) introduced two groups of participants to a set of puzzles. One group worked on sample puzzles; the other group merely watched someone else work on them. All participants were then asked to rate their interest in the puzzles (attitude) and were given an opportunity to spend time on them (behavior). As it turned out, attitudes and behaviors were more consistent among participants who had previously sampled the puzzles.

Third, an attitude can be strengthened, ironically, by an attack against it from a persuasive message. According to Zakary Tormala and Richard Petty (2002), people hold attitudes with varying degrees of certainty, and they be- come more confident in their positions after they successfully resist chang- ing that attitude in response to a persuasive communication. In one study, researchers confronted university students with an unpopular proposal to add senior comprehensive exams as a graduation requirement. Each student read a pro-exam argument that was described as strong or weak, after which they were asked to write down counterarguments and indicate their attitude toward the policy. The result: Students who continued to oppose the policy despite reading what they thought to be a strong argument became even more certain of their opinion.

Additional studies have shown that this effect depends on how satisfied peo- ple are with their own resistance. When people resist a strong message and be- lieve that they have done so in a compelling way, they become more certain of their attitude and more likely to form a behavioral intention that is consistent with it. When people resist a persuasive message “by the skin of their teeth,” however, and see their own counterarguments as weak, they become less certain of their initial attitude and more vulnerable to subsequent attack (Tormala et al., 2006). Even if a person’s belief in his or her own thoughtful response is incorrect, it can influence the strength of the attitude in question (Barden & Petty, 2008).

A fourth key factor is that strong attitudes are highly accessible to awareness, which means that they are quickly and easily brought to mind (Fazio, 1990). To return to our earlier examples, computer jocks think often about their computer preferences, and political activists think often about their allegiances to political parties. It turns out that many attitudes—not just those we feel strongly about— easily pop to mind by the mere sight or even just the mention of an attitude object (Bargh et al., 1992). When this happens, the attitude can trigger behavior in a quick, spontaneous way or by leading us to think carefully about how we feel and how to respond (Fazio & Towles-Schwen, 1999).

80220_ch06_ptg01_hr_208-261.indd 221 27/01/16 5:20 PM

222 Chapter 6 Attitudes

Attitudes in a Cultural Context Bo Kyung Park and colleagues (2013) tell this story: “In the hallways of a local library in Korea, it is not unusual to see in- dividuals in front of a vending machine, wavering between two beverage options for a while, eventually making a choice by doing ‘eeny, meeny, miny, moe.’ Thus a vending machine in the library has the unique option, ‘Random.’ If you press this option, the machine will give you Coke or Sprite randomly! Such an idea may sound hilarious or absurd to North Americans. However, this vending machine anecdote nicely illustrates a cultural difference regarding preference and choice” (pp. 106-107).

In Western cultures that value independence, it is common to see our attitudes as a part of who we are, embodying our values, tastes, preferences, and person- alities. Making choices is an exercise of our attitudes and preferences. From that perspective, it is natural to expect that our likes and dislikes will remain relatively consistent over time and predictive of behavior. In many East Asian cultures, however, where independence, choice, and personal preference are less highly valued, a person’s attitude may well not show this level of consistency. Indeed, Hila Riemer and colleagues (2014) have observed that whereas Western views of attitudes are often person-centric in these ways, in other parts of the world at- titudes depend more on contextual factors such as social norms, others’ expecta- tions, roles, and obligations.

To the extent that norms can change over time and across situations, people in non-Western cultures may well exhibit less consistency in their attitudes. Research by Wilken and others (2011) supports this hypothesis. In one study, American and Japanese participants were asked to report on their favorite musical artists, TV shows, restaurants, and other preferences—and to estimate how long these prefer- ences have lasted. Overall, the Japanese participants reported liking their favorites for a shorter period of time than Americans did. In a second study, participants evaluated how trendy various items were, such as the iPod, Harry Potter books, and the Wii. When questioned about those same items one year later, American participants stayed relatively consistent in their evaluations; Japanese participants reported significant changes in their attitudes.

If attitudes in Western cultures are consistent parts of a person, which we carry with us over time and across situations, it stands to reason that our attitudes and behavior would be highly correlated. But does it also stand to reason that attitudes would be less predictive of behavior in non-Western cultures, where context matters as well? In one study, North American and Indian participants rated how much they liked everyday items like watches, shoes, and shirts. Later when they had to choose one item from a set of four, Indians were less likely than Americans to choose the watch, shoes, or shirts they said they liked the best (Savani et al., 2008). For Indian participants, their choices presumably depended on factors other than personal preferences. Is the link between attitudes and be- havior stronger in some cultures than others? It is possible. At this point, however, more research is needed to test this specific proposition.

To sum up: Research on the link between people’s attitudes and behavior leads to an important conclusion. Our evaluations of an object do not always determine our actions because other factors must be taken into account. How- ever, when attitudes are strong and specific to a behavior, at least in Western cultures, the effects are beyond dispute. Under these conditions, voting is in- fluenced by political opinions, consumer purchasing is affected by product at- titudes, and racial discrimination is rooted in feelings of prejudice. Attitudes can be important determinants of behavior. The question is, how can attitudes be changed?

80220_ch06_ptg01_hr_208-261.indd 222 27/01/16 5:20 PM

223 Persuasion by Communication

On a day-to-day basis, we are all involved in the process of changing attitudes. On TV, on Facebook pages and blogs, in pop-up ads, magazines, and billboards, advertisers flood us with ad campaigns designed to sell cars, soft drinks, credit cards, sneakers, prescription drugs, airlines, new movies, and travel destinations. Likewise, politicians make speeches, run commercials, pass out bumper stickers, and kiss babies to win votes. Attitude change is sought whenever parents social- ize their children, scientists advance theories and seek funding, religious groups seek converts, financial analysts recommend stocks, or trial lawyers argue cases to a jury. Some appeals work; others do not. Some are soft and subtle; others are hard and blatant. Some serve the public interest, whereas others serve commercial interests. The point is, there is nothing inherently evil or virtuous about changing attitudes—a process known as persuasion. We do it all the time.

If you wanted to change someone’s attitude on an issue, you’d probably try to do it by making a persuasive communication. Appeals made in person, over the Internet, or through the mass media rely on the spoken word, the written word, and the image or video that is worth a thousand words. What determines whether an appeal succeeds or fails? To understand why certain approaches are effective whereas others are not, social psychologists have long sought to understand how and why persuasive communications work. For that, we need a road map of the persuasion process.

j Two Routes to Persuasion It’s a familiar scene in American politics: Every four years, two or more presi- dential candidates launch extensive—and expensive—campaigns for office. In a way, if you’ve seen one election, you’ve seen them all. The names and dates may change, but over and over again, opposing candidates accuse each other of duck- ing substantive issues and turning the election into a high stakes, money driven, mud-slinging popularity contest.

True or not, these criticisms show that politicians are keenly aware that they can win votes through two different methods. They can stick to policy, issues, and rational argumentation using the power of words or they can base their appeals on other grounds. Interestingly, these “other grounds” can well determine who wins an election. In The Political Brain, Drew Westen (2007) presents a wealth of research evidence indicating that in the marketplace of politics, emotions trump reason. Based on a combination of laboratory experiments and public opinion polls, other political psychologists agree (Brader, 2006; Neuman et al., 2007). Out- side the realm of politics too, influence can be quick and automatic. In Split-Sec- ond Persuasion, Kevin Dutton (2010) described how Buddhist monks, magicians, advertisers, con artists, hostage negotiators, and other “super-persuaders” use simplicity, empathy, an air of self-confidence, and other disarming tactics to effect instant persuasion.

To account for the two alternative approaches to persuasion, Richard Petty and John Cacioppo (1986) proposed a dual-process model of persuasion. This model assumes that we do not always process communications the same way. When people think hard and critically about the contents of a message, they are said to take a central route to persuasion and are influenced by the strength and quality of the arguments. When people do not think hard or critically about the contents of a message but focus instead on other cues, they take a peripheral route to persuasion. As we’ll see, the route taken depends on whether one is

Persuasion by Communication

persuasion The process by which attitudes are changed.

central route to persuasion The process by which a person thinks carefully about a communication and is influenced by the strength of its arguments.

peripheral route to persuasion The process by which a person does not think carefully about a communication and is influenced instead by superficial cues.

80220_ch06_ptg01_hr_208-261.indd 223 27/01/16 5:20 PM

224 Chapter 6 Attitudes

D ig

ita l F

oc us

/A la

m y

willing and able to scrutinize the information contained in the message itself. Over the years, this model has pro- vided an important framework for understanding the fac- tors that elicit persuasion (Petty & Wegener, 1998).

The Central Route In the first systematic attempt to study persuasion, Carl Hovland and colleagues (1949, 1953) started the Yale Communication and Attitude Change Program. They proposed that for a persuasive message to have influence, the recipients of that message must learn its contents and be motivated to accept it. According to this view, people can be persuaded only by an argument they attend to, comprehend, and retain in memory for later use. Regardless of whether the message takes the form of a live personal appeal, a newspaper editorial, a Sunday sermon, a TV commercial, or a pop-up window on a website, these basic requirements remain the same.

A few years later, William McGuire (1969) reiterated the information- processing steps necessary for persuasion and like the Yale group before him dis- tinguished between the learning, or reception, of a message, a necessary first step, and its later acceptance. In fact, McGuire (1968) used this distinction to explain the surprising finding that a recipient’s self-esteem and intelligence are unrelated to persuasion. In McGuire’s analysis, these characteristics have opposite effects on reception and acceptance. People who are smart or high in self-esteem are better able to learn a message, but are less likely to accept its call for a change in at- titude. People who are less smart or low in self-esteem are more willing to accept the message, but they may have trouble learning its contents. Overall, then, nei- ther group is generally more vulnerable to persuasion than the other—a prediction that is supported by a good deal of research (Rhodes & Wood, 1992).

Anthony Greenwald (1968) and others then argued that persuasion requires a third, intermediate step: elaboration. To illustrate, imagine you are offered a job and your prospective employer tries to convince you over lunch to accept. You listen closely, learn the terms of the offer, and understand what it means. But if it’s a really important interview, your head will spin with questions as you weigh all the pros and cons and contemplate the implications: What would it cost to move? Is there potential for advancement? Am I better off staying where I am? When confronted with personally significant messages, we don’t listen merely

In U.S. presidential politics, candidates try to win votes by addressing the issues, as in debates and speeches delivered from a podium (the central route) or through the use of banners, balloons, music, and other theatrics (the peripheral route).

elaboration The process of thinking about and scrutinizing the arguments contained in a persuasive communication.

Jo e

Ra ed

le /G

et ty

Im ag

es N

ew s/

G et

ty Im

ag es

M ar

k W

ils on

/G et

ty Im

ag es

N ew

s/ G

et ty

Im ag

80220_ch06_ptg01_hr_208-261.indd 224 27/01/16 5:20 PM

225 Persuasion by Communication

to collect information; we think about that information. When this happens, the message is effective to the extent that it leads us to focus on favorable rather than unfavorable thoughts.

These theories of attitude change all share the assumption that the recipi- ents of persuasive appeals are attentive, active, critical, and thoughtful of every word spoken. This assumption is correct—some of the time. When it is and when people consider a message carefully, their reaction to it depends on the strength of its contents. In these instances, messages have greater impact when they are easily learned rather than difficult, when they are memorable rather than forgettable, and when they stimulate a good deal of favorable rather than unfavorable elaboration. Ultimately, strong arguments are persuasive and weak arguments are not.

On the central route to persuasion, the process is eminently rational. It’s im- portant to note, however, that thinking hard and carefully about a persuasive message does not guarantee that the process is objective or that it necessarily promotes truth seeking. At times, each of us prefers to hold a particular attitude and become biased in the way we process information (Petty & Wegener, 1998). Among college students who were politically conservative or liberal, the tendency to agree with a social welfare plan was influenced more—rapidly, strongly, and persistently—by whether it was said to have the support of Democrats or Repub- licans than by the logical merits of the policy itself (Cohen, 2003; Smith et al., 2012). Similarly, college students were less likely to be persuaded by a proposed tuition hike to fund campus improvements when the increase would take effect in one year, thus raising the personal stakes, than by a proposal to raise tuition in eight years (Darke & Chaiken, 2005).

There is one additional complicating factor to consider. Several years ago, Petty and his colleagues (2002) proposed the self-validation hypothesis that people not only “elaborate” on a persuasive communication with positive or negative attitude-relevant thoughts; they also seek to assess the validity of these thoughts. Those thoughts that we hold with high confidence will have a strong impact on our attitudes, as predicted by the Dual-Process Model of persuasion; but those we hold with low confidence will not have a strong impact on our attitudes. This means that various aspects of a persuasive communication—such as whether the source is a knowledgeable expert, and whether the position he or she advocates is one with which we agree or disagree—can affect the confidence we have in our own thoughts and, in turn, our attitudes (see Briñol & Petty, 2009).

The Peripheral Route “The receptive ability of the masses is very limited, their understanding small; on the other hand, they have a great power of forgetting.” The author of this cynical assessment of human nature was Adolf Hitler (1933, p. 77). Believing that human beings are incompetent processors of information, Hitler relied in his propaganda on the use of slogans, uniforms, swastika-covered flags, a special salute, and other symbols. For Hitler, “meetings were not just occasions to make speeches; they were carefully planned theatrical productions in which settings, light- ing, background music, and the timing of entrances were devised to maximize the emotional fervor of an audience” (Qualter, 1962, p. 112).

Do these ploys work? Can the masses be so handily manipulated into persua- sion? History shows that they can. Audiences are not always thoughtful. Some- times people do not follow the central route to persuasion but instead take a shortcut through the peripheral route. Rather than try to learn about a message and think through the issues, they respond with little effort on the basis of super- ficial peripheral cues.

80220_ch06_ptg01_hr_208-261.indd 225 27/01/16 5:20 PM

226 Chapter 6 Attitudes

On the peripheral route to persuasion, people will often evaluate a communi- cation by using simple-minded heuristics, or rules of thumb (Chaiken, 1987; Chen & Chaiken, 1999). If a communicator has a good reputation, speaks fluently, or writes well, we tend to assume that his or her message must be correct. And when a speaker has a reputation for being honest, people think less critically about the contents of his or her communication (Priester & Petty, 1995). Likewise, we assume that a message must be correct if it contains a long litany of arguments or statistics or an impressive list of supporting experts, if it’s familiar, if it elicits cheers from an audience, or if the speaker seems to argue against his or her own interests. In some cases, people will change their attitudes simply because they know that an argument has majority support (Giner-Sorolla & Chaiken, 1997).

On the mindless peripheral route, people are also influenced by a host of fac- tors that are not relevant to attitudes—such as cues from their own body move- ments. Increasingly, social psychologists are coming to appreciate the extent to which human thought is embodied—that the way we think and feel about things is influenced by the physical position, orientation, and movements of our bodies (Lakoff & Johnson, 1999; Niedenthal et al., 2005).

Several studies illustrate attitude embodiment effects. In one study, partici- pants were coaxed into nodding their heads up and down (as if saying yes) or shaking their heads from side to side (as if saying no) while listening via head- phones to an editorial, presumably to test whether the headphones could endure the physical activity. Those coaxed into nodding later agreed more with the argu- ments than those coaxed into shaking their heads from side to side (Wells & Petty, 1980). In other studies, participants viewed graphic symbols or word-like stimuli (surtel, primet) while using an exercise bar to either stretch their arms out (which mimics what we do to push something away) or flex their arms in (which we do to bring something closer). Participants later judged these stimuli to be more pleas- ant when they were associated with the flexing of the arm than when they were associated with the stretching-out motion (Cacioppo et al., 1993; Priester et al., 1996). Even our attitudes toward consumer products can be influenced by bodily sensations. In one study, for example, participants evaluated the appearance of vases, flowers, and other products placed at a distance as more appealing when they stood on a soft, comfortable carpet than on a hard tile floor (Meyers-Levy et al., 2010).

Route Selection Thanks to Petty and Cacioppo’s (1986) two-track distinction be- tween the central and peripheral routes, it is easy to understand why the persuasion process seems so logical on some occasions yet so illogical on others—why voters may select candidates according to issues or images, why juries may base their ver- dicts on evidence or a defendant’s appearance, and why consumers may base their purchases on marketing reports or on product images. The process that is engaged depends on whether the recipients of a persuasive message have the ability and the motivation to take the central route or whether they rely on peripheral cues instead.

To understand the conditions that lead people to process information on one route or the other, it’s helpful to view persuasive communication as the outcome of three factors: a source (who), a message (says what and in what context), and an audience (to whom). Each of these factors steers a recipient’s approach to the communication. If a source speaks clearly, if the message is important, if there is a bright, captive, and involved audience that cares deeply about the issue and has time to absorb the information, then audience members will be willing and able to take the effortful central route. But if the source speaks at a rate too fast to comprehend, if the message is trivial or too complex to process, or if audience

80220_ch06_ptg01_hr_208-261.indd 226 27/01/16 5:20 PM

227 Persuasion by Communication

members are distracted, pressed for time, or uninterested, then the less strenuous peripheral route is taken.

d Figure 6.5 presents a road map of persuasive communication. In the next three sections, we will follow this map from the input factors (source, message, and audience) through the central or peripheral processing routes to reach the final destination: persuasion.

j The Source Golfer Tiger Woods is a living legend, one of the most gifted athletes of our time. Until recently, he was also paid more millions of dollars per year than just about anyone else—to endorse Nike, American Express, and other products. Woods was considered a highly effective spokesman until 2009 when various extramarital affairs were exposed, causing the breakup of his marriage and the demise of his championship caliber golf game. He has struggled since that time and is no longer the top-ranked golfer in the world or the most highly sought spokesperson. What does the story of Tiger Woods tells us about source effects in persuasion? More specifically, what makes some communicators in general more effective than others? As we’ll see, there are two key attributes: credibility and likability.

Credibility Imagine you are waiting in line in a supermarket and you catch a glimpse of this large boldfaced headline: “Doctors Discover Cure for AIDS!” As your eye wanders across the front page, you discover that you are reading a supermarket tabloid that features aliens from another planet. What would you think? In contrast, imagine that you are reading through scientific periodicals in a library when you come across a similar article, but this time it appears in the New England Journal of Medicine. Now what would you think?

Chances are, you’d react with more excitement to the medical journal than to the tabloid, even though both sources report the same news item. In a study conducted during the Cold War era of the 1950s, American participants read a speech that advocated for the development of nuclear submarines. The speech

d F i g u R e 6 . 5

Two Routes to Persuasion

Based on aspects of the source, message, and audience, recipients of a communication take either a central or peripheral route to persuasion. On the central route, people are influenced by strong arguments and evidence. On the peripheral route, persuasion is based more on heuristics and other superficial cues. This two-process model helps explain how persuasion can seem logical on some occasions and illogical on others.

Source Message

Audience

High ability and motivation

Low ability or motivation

Central route

Peripheral route

Persuasion

Input

Processing Strategy

Output

en ga

ge L

ea rn

in g®

in reacting to persuasive communications, people are influenced more by superficial images than by logical arguments.

FAlSe

80220_ch06_ptg01_hr_208-261.indd 227 27/01/16 5:20 PM

228 Chapter 6 Attitudes

elicited more agreement when it was attributed to an eminent American physicist than when the source was said to be the Soviet government–controlled newspa- per (Hovland & Weiss, 1951). Likewise, when participants read a lecture favoring more lenient treatment of juvenile offenders, they changed their attitudes more when they thought the speaker was a judge than when they believed the speaker was a convicted drug dealer (Kelman & Hovland, 1953). Recent research confirms that high-credibility sources are generally more persuasive than low-credibility sources (Pornpitakpan, 2004).

Why are some sources more credible than others? Why were the medical journal, the physicist, and the judge more credible than the tabloid, govern- ment-controlled newspaper, and drug dealer? For communicators to be seen as credible, they must have two characteristics: competence and trustworthiness. Competence refers to a speaker’s ability. People who are knowledgeable, smart, or well spoken or who have impressive credentials are persuasive by virtue of their expertise (Hass, 1981). Experts can have a disarming effect on us. We assume they know what they’re talking about. So when they speak, we listen. And when they take a position, often we yield. Indeed, research shows that people pay attention more closely to experts than to nonexperts and scrutinize their arguments more carefully (Tobin & Raymundo, 2009).

The impact of experts on our attention, confidence, and attitudes, is not simple or uniform. The effect depends on how we feel about the attitude they advocate—say, on politically charged issues such as gun control, immigration pol- icy, or climate change. As suggested by the self-validation hypothesis described earlier, a highly credible source who argues for a position we tend to favor (yay!) bolsters our confidence and existing attitude—more than someone less credible would. In this case, there is less need to scrutinize the expert than the support-

ive but questionable nonexpert. But a highly cred- ible source who advocates for a position we oppose (yikes!) poses a real threat to our confidence and existing attitude—more than someone less credible. In this instance, we need to scrutinize the expert more than the nonexpert. Research is consistent with this nuanced hypothesis: People scrutinize nonex- perts more than experts when they advocate a posi- tion we agree with, but they scrutinize nonexperts more when they advocate a position we oppose (Clark et al., 2012; Clark & Evans, 2014; Clark & Wegner, 2013).

Expertise is only one aspect of credibility. To have credibility, however, sources must also be trustwor- thy—that is, they must be seen as willing to report their knowledge truthfully and without compromise. What determines whether we trust a communica- tor? To some extent, we make these judgments on the basis of stereotypes. In 2014, in an update to a survey conducted many times over the years, the Gallup Organization asked 1,000 Americans to rate how honest people are from various occupational categories. As shown in Table 6.1, nurses topped the list as the most trusted occupational group. Car salesmen, members of Congress, and advertisers were the least trusted.

In December of 2014, a Gallup poll was conducted to determine the level of honesty attributed to people from various occupational groups. Indicated alongside are the percentages of respondents who rated each group as “high” or “very high” in honesty.

m T a b l e 6 . 1

Who Do You Trust?

Occupation % Very high/High

Nurses 80

Medical doctors 65

Pharmacists 65

Police officers 48

Clergy 46

Bankers 23

Lawyers 21

Business executives 17

Advertising practitioners 10

Car salespeople 8

Members of Congress 7

Please tell me how you would rate the honesty and ethical standards of people in these different fields—very high, high, average, low, or very low?

80220_ch06_ptg01_hr_208-261.indd 228 27/01/16 5:20 PM

229 Persuasion by Communication

In judging the credibility of a source, common sense arms us with a simple rule of caution: Beware of people who have something to gain from successful persuasion. If a speaker has been paid off, has an ax to grind, or is simply telling us what we want to hear, we suspect some degree of bias. This rule sheds light on a classic dilemma in advertising concerning the value of celebrity spokespersons: The more products a celebrity endorses, the less trustworthy he or she appears to consumers (Tripp et al., 1994). In the courtroom, the same rule of caution can be used to evaluate witnesses. In one study, research participants served as jurors in a mock trial involving a man who claimed that his exposure to an industrial chemical at work had caused him to contract cancer. Testifying in support of this claim was a biochemist paid either $4,800 or $75 for his expert testimony. You might think that jurors would be more impressed by the scientist who com- manded the higher fee. Yet while he was highly paid, the expert was perceived to be a “hired gun” and was as a result less believable and less persuasive (Cooper & Neuhaus, 2000).

The self-interest rule has other interesting implications. One is that peo- ple are impressed by others who take unpopular stands or argue against their own interests. When research participants read a political speech accusing a large corporation of polluting a local river, those who thought the speechmaker was a pro-environment candidate addressing a staunch environmentalist group perceived him to be biased, whereas those who thought he was a pro-business candidate talking to company supporters assumed he was sincere (Eagly et al., 1978).

Trust is also established by speakers who do not purposely try to change our views. Thus, people are influenced more when they think that they are acciden- tally overhearing a communication than when they receive a sales pitch clearly intended for their ears (Walster & Festinger, 1962). That’s why advertisers some- times use the “overheard communicator” trick, in which the source appears to tell a buddy about a new product that works. As if eavesdropping on a personal conversation, viewers assume that what one friend says to another can be trusted. The self-interest rule also has great relevance in law, which is why people are far

At an October 2014 press conference at Tsinghua University in Beijing, Facebook co-founder Mark Zuckerberg stunned the crowd by speaking and answering questions in Mandarin. In his effort to bring Facebook to China, “Zuck” may have enhanced his persuasive appeal by appearing more similar and, hence, more likeable.

La nd

80220_ch06_ptg01_hr_208-261.indd 229 27/01/16 5:20 PM

230 Chapter 6 Attitudes

more likely to believe a crime suspect’s admissions of guilt than his or her denials (Kassin, 2012).

likability More than anything else, the celebrity star power of Tiger Woods was based on his athletic prowess, his popularity, and his winning smile. Before the revelations that ended Woods’ marriage and derailed his game, he was seen as a likable person. But does that quality necessarily enhance someone’s impact as a communicator? Yes. In his classic bestseller, How to Win Friends and Influence People, Dale Carnegie (1936) said that being liked and being persuasive go hand in hand. The question is, what makes a communicator likable? As we’ll see in Chap- ter 9, two factors that spark attraction are similarity and physical attractiveness.

A study by Diane Mackie and others (1990) illustrates the persuasive power of similarity. Students enrolled at the University of California, Santa Barbara, read a strong or a weak speech that argued against continued use of the SATs in college admissions. Half the participants were led to believe that the speech was written by a fellow UCSB student; the other half thought the author was a student from the University of New Hampshire. Very few participants were persuaded by the weak arguments. In contrast, many of those who read the strong message did change their attitudes, but only when they believed it was given by a fellow UCSB student.

Just as source similarity can spark persuasion, dissimilarity can have the op- posite inhibiting effect. In a study of people’s taste in music, Clayton Hilmert and others (2006) introduced participants to a confederate who seemed to like the same or different kinds of music, such as rock, pop, country, or classical. Others did not meet a confederate. When later asked to rate a particular song, participants were positively influenced by the similar confederate’s opinion and negatively influenced by the dissimilar confederate’s opinion. In fact, although the effect is more potent when the points of similarity seem relevant to the atti- tude in question (Berscheid, 1966), the participants in this study were also more or less persuaded by a confederate whose similarities or differences were wholly

Ro n

Tu re

nn e/

N BA

E/ G

et ty

Im ag

D ie

t C ok

e/ Sp

la sh

N ew

s/ Co

rb is

When he was just out of high school, basketball star LeBron James (left) signed a multimillion dollar contract with Nike. Also paid millions, music star Taylor Swift is surrounded by kittens in this TV commercial for Diet Coke (top right) and Oscar-winning actor Bradley Cooper is the new, and first-ever, face of Haagen-Dazs ice cream (bottom right). Can celebrities sell products? Targeting the peripheral route to persuasion, the advertising industry seems to think so.

Sp la

sh /H

aa ge

n D

az s/

Sp la

sh N

ew s/

Co rb

80220_ch06_ptg01_hr_208-261.indd 230 27/01/16 5:20 PM

231 Persuasion by Communication

unrelated to music—for example, when the confederate had similar or different interests in shopping, world politics, museums, foods, or social media websites.

The effect of source similarity on persuasion has obvious implications for those who wish to exert influence. We’re all similar to one another in some re- spects. We might agree in politics, share a common friend, have similar tastes in food, or enjoy spending summers on the same beach. If aware of the social ben- efits of similarity and the social costs of dissimilarity, the astute communicator can use common bonds to enhance his or her impact on an audience.

Advertising practices presuppose that beauty is also persuasive. After all, on- line, magazine, and TV ads routinely feature supermodels who are young, tall, and slender supermodels (for women) or muscular (for men) and who sport hard bodies, glowing complexions, and radiant smiles. Sure, these models can turn heads, you may think, but can they change minds? In a study that addressed this question, Shelly Chaiken (1979) had male and female college students approach others on campus. They introduced themselves as members of an organization that wanted the university to stop serving meat during breakfast and lunch. In each case, these student assistants gave reasons for the position and then asked respondents to sign a petition. The result: Attractive sources were able to get 41% of respondents to sign the petition, whereas those who were less attractive suc- ceeded only 32% of the time. Additional research has shown that attractive male and female salespersons elicit more posi- tive attitudes and purchasing intentions from customers than less attractive salespersons, even when they are up front about their desire to make a sale (Reinhard et al., 2006).

When What You Say is More important Than Who You Are To this point, it must seem as if the source of a persuasive communication is more important than the communication it- self. Is this true? Certainly there are enough real-life examples— as when books used to skyrocket to the top of the best-seller list because Oprah Winfrey recommended them. The advertis- ing industry has long debated the value of high-priced celebrity endorsements. David Ogilvy (1985), who was called “the king of advertising,” used to say that celebrities are not effective be- cause viewers know they’ve been bought and paid for. Ogilvy was not alone in his skepticism. Still, many advertisers scramble furiously to sign famous entertainers and athletes. From Tiger Woods to Tina Fey, Rihanna, LeBron James, Peyton Manning, Taylor Swift, Scarlett Johansson, Bono, Danica Patrick, Beyonce Knowles, Katy Perry, Jay Z, and Bradley Cooper, TV commercials regularly feature a parade of stars. The bigger the star, they say, the more valuable the testimonial.

Compared with the contents of a message, does the source really make the big difference that advertisers pay for? Are we so impressed by the expert, so enamored of the physical talent, and so drawn to the charming face that we embrace whatever they have to say? And are we so scornful of ordinary or unattractive people that their presentations fall on deaf ears? In light of what is known about the central and peripheral routes to persuasion, the answer to these questions is “it depends.”

First, a recipient’s level of involvement plays an important role. When a mes- sage has personal relevance to your life, you pay attention to the source and think critically about the message, the arguments, and the implications. When

Advertisers are so convinced that beauty sells products that they pay millions of dollars for supermodels to appear in their ads. Shown here, supermodel Kate Moss appears in an ad for Italian fashion house Versace.

Ve rs

ac e/

A dv

er tis

in g

A rc

hi ve

80220_ch06_ptg01_hr_208-261.indd 231 27/01/16 5:20 PM

232 Chapter 6 Attitudes

a message does not have personal relevance, how- ever, you may take the source at face value and spend little time scrutinizing the information. In a classic study, Richard Petty and others (1981) had students listen to a speaker who proposed that all seniors should be required to take comprehensive exams in order to graduate. Three aspects of the communication situation were varied. First, partici- pants were led to believe that the speaker was either an education professor at Princeton University or a high school student. Second, participants heard either well-reasoned arguments and hard evidence or a weak message based only on anecdotes and per- sonal opinion. Third, participants were told either that the proposed exams might be used the following year (Uh oh, that means me!) or that they would not take effect for another 10 years (Who cares, I’ll be long gone by then!).

As predicted, personal involvement determined the relative impact of the expertise of the source and the quality of speech. Among participants who would not be affected by the proposed change, at- titudes were based largely on the speaker’s credibil- ity: The professor was persuasive; the high school student was not. Among participants who thought that the proposed change would affect them directly, attitudes were based on the quality of the speaker’s proposal. Strong arguments were persuasive; weak arguments were not. As depicted in d Figure 6.6, people followed the source rather than the message under low levels of involvement, illustrating the pe- ripheral route to persuasion. But message factors did outweigh source characteristics under high levels of involvement, when participants cared enough to take the central route to persuasion. Likewise, re- search has shown that the tilt toward likable and attractive communicators is reduced when recipients take the central route (Chaiken, 1980).

There is a second limit to source effects. It is of- ten said that time heals all wounds. Well, time may also heal the effects of a bad reputation. Hovland and Weiss (1951) varied communicator credibility (for example, the physicist versus the Soviet- controlled newspaper) and found that the change had a large and immediate effect on persuasion.

But when they measured attitudes again four weeks later, the effect had van- ished. Over time, the attitude change produced by the high-credibility source had decreased and the change caused by the low-credibility source had increased. This finding of a delayed persuasive impact of a low-credibility communicator is called the sleeper effect.

To explain this unforeseen result, the Hovland research group proposed the discounting cue hypothesis. According to this hypothesis, people immediately

d F i g u R e 6 . 6

Source Versus Message: The Role of Audience involvement

People who were high or low in their personal involvement heard a strong or weak message from an expert or nonexpert. For high-involvement participants (left), persuasion was based on the strength of arguments, not on source expertise. For low- involvement participants (right), persuasion was based more on the source than on the arguments. Source characteristics have more impact on those who don’t care enough to take the central route. Reprinted by permission from Richard E. Petty.

Unfavorable

Favorable

Po st

co m

m un

ic at

io n

at tit

ud e

–.2

−.4

High Low Involvement

Weak argument

Strong argument

High Low

Unfavorable

Favorable

Po st

co m

m un

ic at

io n

at tit

ud e

−.2

−.4

Involvement

Expert source

Nonexpert source

sleeper effect A delayed increase in the persuasive impact of a noncredible source.

80220_ch06_ptg01_hr_208-261.indd 232 27/01/16 5:20 PM

233 Persuasion by Communication

discount the arguments made by noncredible communicators, but over time, they dissociate what was said from who said it. In other words, we tend to remem- ber the message but forget the source (Pratkanis et al., 1988). To examine the role of memory in this process, Kelman and Hovland (1953) reminded a group of participants of the source’s identity before reassessing their attitudes. If the sleeper effect was caused by forgetting, they reasoned, then it could be eliminated through reinstatement of the link between the source and the message. As shown in d Figure 6.7, they were right. When attitudes were measured after three weeks, participants who were not reminded of the source showed the usual sleeper effect. Yet those who did receive a source reminder did not. For these latter participants, the effects of high and low credibility endured. Recent studies by cognitive psy- chologists have confirmed that over time, people “forget” the connection between information and its source (Underwood & Pezdek, 1998).

The sleeper effect generated a good deal of controversy. There was never a doubt that credible communicators lose some impact over time. But researchers had a harder time finding evidence for delayed persuasion by noncredible sources. Exasperated at one point by their own failures to obtain this result, Paulette Gillig and Anthony Greenwald (1974) wondered, “Is it time to lay the sleeper effect to rest?” As it turned out, the answer was no. More recent research showed that the sleeper effect is reliable provided that participants do not learn who the source is until after they have received the original message (Greenwald et al., 1986; Kumkale & Albarracín, 2004).

To appreciate the importance of timing, imagine that you’re searching for music online when you come across what appears to be a review of a new CD. Before you begin reading, however, you notice in the fine print that this so-called

d F i g u R e 6 . 7

The Sleeper effect

In Experiment 1, participants changed their immediate attitudes more in response to a message from a high-credibility source than from a low-credibility source. When attitudes were measured again after three weeks, the high-credibility source had lost impact and the low-credibility source had gained impact—the sleeper effect. In Experiment 2, the sleeper effect disappeared when participants were reminded of the source. Kelman & Hovland, 1953.

Experiment 1: Now You See It

Experiment 2: Now You Don't

Immediate 3 weeks

High-credibility source Low-credibility source

A tti

tu de

c ha

ng e

Time Interval

A tti

tu de

c ha

ng e

Immediate 3 weeks Time Interval

“The truth is always the strongest argument.”

—Sophocles

80220_ch06_ptg01_hr_208-261.indd 233 27/01/16 5:20 PM

234 Chapter 6 Attitudes

review is really an advertisement. Aware that you can’t always trust what you read, you skim the ad and reject it. Now imagine the same situation, except that you read the entire ad before realizing what it is. Again you reject it. But notice the difference. This time, you have read the message with an open mind. You may still reject it, but after a few weeks, the information will have sunk in to influence your evaluation of the music. This scenario illustrates the sleeper effect.

j The Message On the peripheral route to persuasion, audiences are influenced heavily, maybe too heavily, by various source characteristics. But when people care about an issue, the strength of a message determines its impact. On the central route to persuasion, what matters most is whether a scientist’s theory is supported by the data or whether a company has a sound product. Keep in mind, however, that the target of a persuasive appeal comes to know a message only through the medium of communication—what a person has to say and how that person says it.

informational Strategies Communicators often struggle with how to structure and present an argument to maximize its impact. Should a message be long and crammed with facts or short and to the point? Is it better to present a highly partisan, one-sided message or to take a more balanced, two-sided approach? How should arguments be ordered—from strongest to weakest or the other way around? These are the kinds of questions often studied by persuasion researchers—including those interested in marketing, advertising, and consumer behavior.

Often the most effective strategy to use will depend on whether members of the audience process the message on the central or the peripheral route. Consider the length of a communication. When people process a message lazily, with their eyes and ears half-closed, they often fall back on a simple heuristic: The longer a message, the more valid it must be. In this case, a large number of words gives the superficial appearance of factual support regardless of the quality of the arguments (Petty & Cacioppo, 1984; Wood et al., 1985). Thus, as David Ogilvy (1985) concluded from his years of advertising experience, “The more facts you tell, the more you sell” (p. 88).

When people process a communication carefully, however, length is a two- edged sword. If a message is long because it contains lots of supporting informa- tion, then longer does mean better. The more supportive arguments you can offer or the more sources you can find to speak on your behalf, the more persuasive your appeal will be (Harkins & Petty, 1981). But if the added arguments are weak or if the new sources are redundant, then an alert audience will not be fooled by length alone. When adding to the length of a message dilutes its quality, an appeal might well lose impact (Friedrich et al., 1996; Harkins & Petty, 1987).

When two opposing sides try to persuade the same audience, order of pre- sentation becomes a relevant factor as well. During the summer of 2012, before the November presidential election, the Republicans held their national conven- tion a few days before the incumbent Democrats held theirs. These events were watched on television by millions of voters. Do you think the order in which they were scheduled gave one party an advantage? If you believe that information that is presented first has more impact, you’d predict a primacy effect (advantage to the Republicans). If you believe that the information presented last has the edge, you’d predict a recency effect (advantage to the Democrats).

There are good reasons for both predictions. On the one hand, first impres- sions are important. On the other hand, memory fades over time, and often peo- ple recall only the last argument they hear before making a decision. In light of

80220_ch06_ptg01_hr_208-261.indd 234 27/01/16 5:20 PM

235 Persuasion by Communication

these contrasting predictions, Norman Miller and Donald Camp- bell (1959) searched for the “missing link” that would determine the relative effects of primacy and recency. They discovered that the missing link is time. In a study of jury simulations, they had people (1) read a summary of the plaintiff’s case; (2) read a sum- mary of the defendant’s case; and (3) make a decision. The re- searchers varied how much time separated the two messages and then how much time elapsed between the second message and the decisions. When participants read the second message right after the first and then waited a whole week before reporting their opinion, a primacy effect prevailed and the side that came first was favored. Both messages faded equally from memory, so only the greater impact of first impressions was left. When participants made a decision immediately after the second message but a full week after the first, however, there was a recency effect. The sec- ond argument was fresher in memory, thus favoring the side that went last. Using these results as a guideline, let’s revisit our origi- nal question: What is the impact on Election Day of how national conventions are scheduled? Think for a moment about the placement and timing of these events. The answer appears in Table 6.2.

Message Discrepancy Persuasion is a process of changing attitudes. This ob- jective is not easy to achieve. As a general rule, people are motivated to defend their opinions and attitudes, which they do, in part, through selective exposure to information that supports their views (Hart et al., 2009). Given an opportunity to advocate for attitude change, communicators confront what is perhaps the most critical strategic question: How extreme a position should they take? How discrep- ant should a message be from the audience’s existing position in order to have the greatest impact? Common sense suggests two opposite answers. One approach is to take an extreme position in the hope that the more change you advocate, the more you will get. Another approach is to exercise caution and not push for too much change so that the audience will not reject the message outright. Which approach seems more effective? Imagine trying to convert your politically conservative friends into liberals or the other way around. Would you stake out a radical position in order to move them toward the center or would you preach moderation so as not to be cast aside?

Research shows that communica- tors should adopt the second, more cautious approach. To be sure, some discrepancy is needed to produce a change in attitude. But the relationship to persuasion can be pictured as an upside-down U with the most change being produced at moderate amounts of discrepancy (Bochner & Insko, 1966). A study by Kari Edwards and Edward Smith (1996) helps explain why taking a more extreme position is counterproductive. These investigators first measured people’s attitudes on a number of hot social issues—for exam- ple, whether lesbian and gay couples

Research on the sleeper effect shows that people often remember the message but forget the source.

“It is a superb vision of America, all right, but I can’t remember which candidate projected it.”

Ed F

is he

r/ Th

e N

ew Y

or ke

r C ol

le ct

io n/

Ca rt

oo n

Ba nk

.c om

m T a b l e 6 . 2

effects of Presentation order and Timing on Persuasion

Conditions Results

1. Message 1 Message 2 One week Decision Primacy

2. Message 1 One week Message 2 Decision Recency

3. Message 1 Message 2 Decision None

4. Message 1 One week Message 2 One week Decision None

A study by Miller and Campbell (1959) demonstrated the effect of presentation order and the timing of opposing arguments on persuasion. As applied to our example, the Democratic and Republican conventions resemble the fourth row of this table. From these results, it seems that the scheduling of such events is fair, promoting neither primacy nor recency.

80220_ch06_ptg01_hr_208-261.indd 235 27/01/16 5:20 PM

236 Chapter 6 Attitudes

should adopt children, whether employers should give preference in hiring to mi- norities, and whether the death penalty should be abolished. Several weeks later, they asked the same people to read, think about, and rate arguments that were either consistent or inconsistent with their own prior attitudes. The result: When given arguments to read that preached attitudes that were discrepant from their own, the participants spent more time scrutinizing the material and judged the ar- guments to be weak. Clearly, people tend to refute and reject persuasive messages they don’t agree with. In fact, the more personally important an issue is to us, the more stubborn and resistant to change we become (Zuwerink & Devine, 1996).

Fear Appeals Many trial lawyers say that to win cases they have to appeal to jurors through the heart rather than through the mind. The evidence is important, they admit, but what matters most is whether the jury reacts to their client with an- ger, disgust, sympathy, or sadness. Of course, very few messages are entirely based on rational argument or on emotion.

Fear is a particularly primitive and powerful emotion, serving as an early warn- ing system that signals danger. Neuroscience research shows that fear is aroused in- stantly in response to pain, stimulation from noxious substances, or threat, enabling us to respond quickly without having to stop to think about it (LeDoux, 1996). The use of fear-based appeals to change attitudes is common. Certain religious cults have used scare tactics to indoctrinate new members. So do public health organizations that graphically portray the damage done to those who smoke cigarettes, eat too much junk food, drink too much alcohol, text while driving, or have unprotected sex.

Negative campaigning in American politics is prevalent, though the effects on voters are not clear (Lau & Rovner, 2009). From the large number of attack ads that flood the political scene, it would certainly seem that candidates, their consul- tants, and Super PACS strongly believe in the power of attacking their opponents by arousing fear about the consequences of voting for them. Presidential cam- paign ads are more negative than ever. Yet the most hard-hitting and controversial ever was a TV commercial that aired just once, on September 7, 1964. In an ad to reelect Democratic president Lyndon Johnson, who was running against Republi- can Barry Goldwater, a young girl pictured in a field counted to 10 as she picked the petals off a daisy. As she reached 9, an adult voice broke in to count down from 10 to 0, followed by a blinding nuclear explosion and this message: “Vote for President Johnson on November 3. The stakes are too high for you to stay home.”

The effects of fear arousal in politics are evident. Guided by Terror Manage- ment Theory (Greenberg et al., 1997; Pyszczynski et al., 2004; see Chapter 3) and the prediction that a deeply rooted fear of death motivates people to rally around their leaders as a way to ward off anxiety, Mark Landau and others (2004) found that college students expressed more support for then-President George W. Bush and his policies when they were reminded of their own mortality or subliminally exposed to images of 9/11 than when they were not. This result is not limited to the laboratory. Analyzing patterns of government-issued terror warnings and Gallup polls, Robb Willer (2004) found that increased terror alerts were predict- ably followed by increases in presidential approval ratings.

Is fear similarly effective for commercial purposes? What about using fear to promote health and safety? If you’re interested in public service advertising, visit the website of the Ad Council, the organization that created Smokey Bear (“Only you can prevent forest fires”) and the crash test dummies (“Don’t be a dummy— buckle up”). In recent years, the Ad Council has run campaigns on a range of issues such as the dangers of using of steroids, flu prevention, AIDS prevention, cyber bullying, and the online sexploitation of youth. To get people to change be- havior in these domains, is it better to arouse a little nervousness or a full-blown

80220_ch06_ptg01_hr_208-261.indd 236 27/01/16 5:20 PM

237 Persuasion by Communication

anxiety attack? To answer this question, social psychologists over the years have compared communications that vary in the levels of fear they arouse. In the first such study, Irving Janis and Seymour Feshbach (1953) found that high levels of fear did not generate increased agreement with a persuasive communication. Since then, however, research has shown that appeals that arouse high levels of fear can be highly effective (de Hoog et al., 2007).

Fear arousal increases the incentive to change for those who do not actively resist it, but its ultimate impact depends on the strength of the arguments and on whether the message also contains clear and reassuring advice on how to cope with the threatened danger (Keller, 1999; Leventhal, 1970; Rogers, 1983). This last point is important. Without specific instructions on how to cope, people feel helpless, and they panic and tune out the message. In one study, for example, participants with a chronic fear of cancer were less likely than others to detect the logical errors in a message that called for regular cancer checkups (Jepson & Chaiken, 1990). When clear instructions are included, however, high dosages of fear can be effec- tive. In the past, research had shown that antismoking films elicit more negative attitudes toward cigarettes when they show gory lung-cancer operations than when they show charts filled with dry statistics (Leventhal et al., 1967) and that films about driving safety are more effective when they show bloody accident victims than when they show plastic crash test dummies (Rogers & Mewborn, 1976). In a meta-analysis of 105 studies, however, Natascha de Hoog and others (2007) found that communications that arouse fear do not have to be gruesome to be effective.

Public health organizations often use fear, or scare tactics, to change health-related attitudes and behavior. In 2014, a public service ad was aired on the dangers of texting while driving. In what starts out as a fun car ride with friends, this driver takes her eye of the road to text. All of a sudden her car is T-boned by a truck, hurls over and over in slow motion, and lands with a silent thud. In a spot aimed at teenagers, the theme is: “U drive. U text. U pay.”

80220_ch06_ptg01_hr_208-261.indd 237 27/01/16 5:20 PM

238 Chapter 6 Attitudes

The more personally vulnerable people feel about a threat- ened outcome, the more attentive they are to the message and the more likely they are to follow its recommendations.

Positive emotions It’s interesting that just as fear helps induce a change in attitude, so does positive emotion. In one study, people were more likely to agree with a series of controversial arguments when they snacked on peanuts and soda than when they did not eat (Janis et al., 1965). In another study, participants liked a television commercial more when it was embedded within a program that was upbeat rather than sad (Mathur & Chattopadhyay, 1991). Research shows that people are “soft touches” when they’re in a good mood. Depending on the situation, food, drinks, a soft reclining chair, warm and tender memories,

a success experience, breathtaking scenery, laughter, and good music can lull us into a positive emotional state and ripe for persuasion (Schwarz et al., 1991).

According to Alice Isen (1984), people see the world through rose-colored glasses when they are in a good mood. Filled with high spirits, we become more sociable, more generous, and generally more positive in our outlook. We also make decisions more quickly and with relatively little thought. The result: Positive feelings activate the peripheral route to persuasion, facilitating change and allowing superficial cues to take on added importance (Petty et al., 1993; Worth & Mackie, 1987).

What is it about feeling good that leads us to take shortcuts to persuasion rather than the more effortful central route? There are three possible explana- tions. One is that a positive emotional state is cognitively distracting, causing the mind to wander and impairing our ability to think critically about the persuasive arguments (Mackie & Worth, 1989; Mackie et al., 1992). A second explanation is that when people are in a good mood, they assume that all is well, let down their guard, and become somewhat lazy processors of information (Schwarz, 1990). A third explanation is that when people are happy, they become motivated to savor the moment and maintain their happy mood rather than spoiling it by thinking critically about new information (Wegener & Petty, 1994).

This last notion raises an interesting question: What if happy people are pre- sented with a positive and uplifting persuasive message? Would they still appear cog- nitively distracted, or lazy, or would they pay close attention in order to prolong the rosy glow? To find out, Duane Wegener and others (1995) showed some college stu- dents a funny segment from The Late Show with David Letterman. Others, less fortu- nate, watched a somber scene from an HBO movie called You Don’t Have to Die. The students were then asked to read and evaluate either an uplifting article they agreed with about a new plan to cut tuition or a distressing article they disagreed with about a new plan to raise tuition. In half the cases, the article they read contained strong arguments; in the other cases, the arguments were weak. Did the students read the material carefully enough to distinguish between the strong and weak arguments? Those in the somber condition clearly did. Among those in the happy condition, however, the response depended on whether they expected the message to be one they wanted to hear. When the happy students read about a tuition increase, they tuned out and were equally persuaded by the strong and weak arguments. When they read about the proposal to cut tuition, however, they were persuaded more when the arguments were strong than when they were weak. Because they were in a good mood and were receiving an agreeable message that would not spoil it, these happy students took the effortful central route to persuasion.

Suggesting that a fear of death leads people to rally around their leaders, public opinion polls have shown that as terror threat levels increase, so do presidential approval ratings.

H om

el an

d Se

cu rit

y— Pu

bl ic

D oc

um en

In a highly controversial appeal to fear, a TV commercial that aired only once during the 1964 presidential campaign, pictured this young girl in an open field, a blinding nuclear explosion, and the message: “Vote for President Johnson on November 3. The stakes are too high for you to stay home.”

Ly nd

on B

. J oh

ns on

L ib

ra ry

80220_ch06_ptg01_hr_208-261.indd 238 27/01/16 5:20 PM

239 Persuasion by Communication

Subliminal Messages In 1957, Vance Packard published The Hidden Persuad- ers, an exposé of Madison Avenue. As the book climbed the best-seller list, it awakened in the public a fear of being manipulated by forces they could not see or hear. What had Packard uncovered? In the 1950s, amid growing fears of com- munism and the birth of rock ‘n’ roll, a number of advertisers were said to have used subliminal advertising, the presentation of commercial messages outside of conscious awareness. It all started in a drive-in movie theater in New Jersey, where the words “Drink Coke” and “Eat popcorn” were secretly flashed on the screen during intermissions for a third of a millisecond. Although the audience never noticed the message, Coke sales were said to have increased 18% and pop- corn sales 58% over a six-week period (Brean, 1958).

This incident was followed by several oth- ers. A Seattle radio station presented subaudible anti-TV messages during its programs (“TV is a bore”), and department stores played music tapes over public address systems that con- tained subaudible warnings about theft (“If you steal, you’ll get caught”). Later, in books entitled Subliminal Seduction (1973) and The Age of Ma- nipulation (1989), William Bryan Key charged that advertisers routinely sneak faint sexual im- ages in visual ads to heighten the appeal of their products. Several years ago, concerns were also raised about subliminal messages in rock music. In one case, the families of two boys who com- mitted suicide blamed the British rock group Judas Priest for subliminal lyrics (“Do it”) that promoted Satanism and suicide (National Law Journal, 1990). Although the families lost their case, it’s clear that many people believe in the power of hidden persuaders.

At the time the story about the New Jer- sey theater broke, research on the topic was so sketchy and the public so outraged by the sin- ister implications that the matter was quickly dropped. But soon there was renewed interest in subliminal influences and new research de- velopments. In one recent field study, for ex- ample, researchers played traditional German or French music on alternating days for two weeks at a supermarket display of wines. Keep- ing track of sales, the researchers found that of the total number of wines bought, 83% were German on German-music days and 65% were French on French-music days. Yet when asked the reasons for their choices, customers did not cite the music as a factor, suggesting that they were not aware of the effect it had on them (North et al., 1999).

Current uses of subliminal influence are varied. In what is now a multimillion- dollar industry, companies today sell self-help tapes, DVDs, and MP3 downloads that play New Age music or nature images and sounds—and also contain fleet- ing messages or “affirmations” that promise to help you relax, lose weight, make

For years, advertisers have defended against the charge that they embed suggestive and sexual images in print ads. This piece by the American Association of Advertising Agencies defends against this claim.

A m

er ic

an A

ss oc

ia tio

n of

A dv

er tis

in g

A ge

nc ie

80220_ch06_ptg01_hr_208-261.indd 239 27/01/16 5:20 PM

240 Chapter 6 Attitudes

friends, raise self-esteem, make money, improve athletic performance, and even improve your sex life. Can subliminal messages reflexively trigger behavior without our awareness? In 1982, Timothy Moore re- viewed the existing research and concluded that “what you see is what you get”—nothing, “complete scams.” Moore was right. The original story about the Coke-and-popcorn messages at the New Jersey theater was later exposed as a publicity stunt and hoax (Pratkanis, 1992). To further complicate matters, controlled experiments using subliminal self-help CDs that promise to raise self-esteem, improve memory, or lose weight show that these products offer no therapeutic benefits (Greenwald et al., 1991; Merikle & Skanes, 1992).

If there is no solid evidence of subliminal influence, why, you may wonder, does research demonstrate perception without awareness in studies of priming (described elsewhere in this book) but not in studies of subliminal persuasion? If you think about it, the two sets of claims are different. In the laboratory, subliminal exposures have a short-term effect on simple judgments and actions. But in claims of subliminal persuasion, the exposure is presumed to have long-term effects on eat- ing, drinking, consumer purchases, voter sentiment, or even the most profound of violent acts, suicide. Psychologists agree that people can process information at an unconscious level, but they’re quick to cau- tion that this processing is “analytically limited” (Greenwald, 1992).

Perhaps people perceive subliminal cues but are not persuaded into action unless they are motivated to do so. To test this hypoth- esis, Erin Strahan and others (2002) brought thirsty college students into the laboratory for a marketing study and provided drinking water to some but not to others. Then, as part of a test administered by computer, they subliminally exposed these students to neutral words (pirate, won) or to thirst-related words (thirst, dry). Did the subliminal “thirsty” message later lead the students, like automatons, to drink more in a taste test of Kool-Aid beverages? Yes and no. d Figure 6.8 shows that the subliminal thirst primes had little impact on students

whose thirst had just been quenched, but they quite clearly increased consumption among those who were thirsty and had been deprived of water. For a subliminal mes- sage to influence behavior, it has to strike “while the iron is hot.”

Other researchers have since extended this interesting effect in important ways. In one study, participants who were subliminally presented with the name of a specific soft drink, Lipton Ice, were later more likely to report that they would select that particular brand over others—provided they were thirsty (Karremans et al., 2006). In a second study, participants were subliminally presented with a logo for one brand of dextrose (a sugar pill) or another, after which they worked on a cognitive task that required intense concentration. The results showed that when given an opportunity to enhance their concentration, participants were more likely to select and consume the subliminally advertised brand than the other brand—provided they were mentally tired and in need of a boost (Bermeit- inger et al., 2009).

j The Audience Although source and message factors are important, the astute communicator must also take his or her audience into account. Presentation strategies that succeed with some people fail with others. Audiences on the central route to persuasion, for

People are most easily persuaded by commercial messages that are presented without their awareness.

FAlSe

d F i g u R e 6 . 8

Subliminal influence

Thirsty and nonthirsty research participants were subliminally exposed to neutral or thirst-related words. Afterward they participated in a beverage taste test in which the amount they drank was measured. You can see that the subliminal thirst cues had little impact on nonthirsty participants but that they did increase consumption among those who were thirsty. Apparently, subliminal cues can influence our behavior when we are otherwise predisposed. Strahan et al., 2002.

Thirst primes Neutral primes

Thirsty Not thirsty

200

A m

ou nt

c on

su m

ed (

m L)

190

130

170

150

180

120

140

160

Subliminal Exposure

80220_ch06_ptg01_hr_208-261.indd 240 27/01/16 5:20 PM

241 Persuasion by Communication

example, bear little resemblance to those found strolling along the peripheral route. In this section, we’ll see that the impact of a message is influenced by two additional factors: the recipient’s personality and his or her expectations.

Right from the start, social psycholo- gists tried to identify types of people who were more or less vulnerable to persua- sion. But it turned out that very few in- dividuals are consistently easy or difficult to persuade. Based on this insight, the search for individual and group differ- ences is now guided by an “interactionist” perspective. Assuming that each of us can be persuaded more in some settings than in others, researchers look for an appropri- ate match between characteristics of the message and the audience. So, what kinds of messages turn you on?

The Need for Cognition Earlier, we saw that people tend to process informa- tion more carefully when they are highly involved. Involvement can be deter- mined by the importance and self-relevance of a message. According to Cacioppo and Petty (1982), however, there are also individual differences in the extent to which people become involved and take the central route to persuasion. Specifi- cally, they have found that individuals differ in the extent to which they enjoy and participate in effortful cognitive activities, or, as they call it, the need for cogni- tion (NC). People who are high rather than low in their need for cognition like to work on hard problems, search for clues, make fine distinctions, and analyze situations. These differences can be identified by the items contained in the Need for Cognition Scale, some of which appear in Table 6.3.

The need for cognition has interesting implications for changing attitudes. If people are prone to approach or avoid effortful cognitive activities, then a com- municator could prepare messages unique to a particular audience. In theory, the high-NC audience should receive information-oriented appeals and the low-NC audience should be treated to appeals that rely on the use of peripheral cues. The theory is fine, but does it work? Can a message be customized to fit the informa- tion-processing style of its recipients?

In a test of this hypothesis, participants read an editorial that consisted of either a strong or a weak set of arguments. As predicted, the higher their NC scores were, the more the participants thought about the material, the better they later recalled it, and the more persuaded they were by the strength of its arguments (Cacioppo et al., 1983). In contrast, people who are low in the need for cognition are persuaded by cues found along the peripheral route, such as a speaker’s reputation and physical appearance, the overt reactions of others in the audience, and a positive mood state (Cacioppo et al., 1996). At times, they are mindlessly influenced by a reputable source even when his or her arguments are weak (Kaufman et al., 1999).

Self-Monitoring Just as people high in the need for cognition crave informa- tion, other personality traits are associated with an attraction to other kinds of messages. Consider the trait of self-monitoring. As described in Chapter 3, high

need for cognition (NC) A personality variable that distinguishes people on the basis of how much they enjoy effortful cognitive activities.

From Cacioppo, J. T., and Petty, R. E., “The need for cognition,” Journal of Personality and Social Psychology vol 42 (pp. 116–131). Copyright © 1982 by the American Psychological Association. Reprinted by permission.

m T a b l e 6 . 3

need for Cognition Scale: Sample Items

Are you high or low in the need for cognition? These statements are taken from the NC Scale. If you agree with items 1, 3, and 5 and disagree with items 2, 4, and 6, you would probably be regarded as high in NC.

1. I really enjoy a task that involves coming up with new solutions to problems.

2. Thinking is not my idea of fun.

3. The notion of thinking abstractly is appealing to me.

4. I like tasks that require little thought once I’ve learned them.

5. I usually end up deliberating about issues even when they do not affect me personally.

6. It’s enough for me that something gets the job done; I don’t care how or why it works.

80220_ch06_ptg01_hr_208-261.indd 241 27/01/16 5:20 PM

242 Chapter 6 Attitudes

self-monitors regulate their behavior from one situation to another out of concern for public self-presentation. Low self-monitors are less image conscious and behave according to their own beliefs, values, and preferences. In the context of persuasion, high self-monitors may be particularly responsive to messages that promise desir- able social images. Whether the product is beer, soft drinks, blue jeans, or cars, or smart phones, this technique is common in advertising, where often the image is the message.

To test the self-monitoring hypothesis, Mark Snyder and Kenneth DeBono (1985) showed image- or information-oriented print ads to high and low self- monitors. In an ad for Irish Mocha Mint coffee, for example, a man and woman were depicted as relaxing in a candlelit room over a steamy cup of coffee. The image-oriented version promised to “Make a chilly night become a cozy evening,” while the informational version offered “a delicious blend of three great flavors— coffee, chocolate, and mint.” As predicted, high self-monitors were willing to pay more for products after reading imagery ads; low self-monitors were influenced more by information-oriented appeals. Smidt and DeBono (2011) found a similar result for energy drinks. They found that high self-monitors rated the drinks more favorably when they had an image-oriented name; low self-monitors preferred the drinks more when they had a self-descriptive name. Imagery can even influence the way that high self-monitors evaluate a product, independent of its quality. DeBono and others (2003) presented people with one of two perfume samples packaged in more or less attractive bottles. Whereas low self-monitors preferred the more pleasant-scented fragrance, high self-monitors preferred whatever scent came from the more attractive bottles.

Regulatory Fit Setting aside your political views, do you find yourself drawn to some types of speeches, arguments, editorials, and television commercials more than others? Joseph Cesario and others (2004) proposed that people are more likely to be influenced by messages that fit their frame of mind and “feel right.” In particular, they noted that in an effort to regulate their own emotion state, some individuals are promotion-oriented (drawn to the pursuit of success, achievement, and their ideals), whereas others are more prevention-oriented (protective of what they have, fearful of failure, and vigilant about avoiding loss). Do these differing outlooks on life make people more responsive to some types of persuasive mes- sages than others? To find out, these researchers presented two versions of an article advocating for a new after-school children’s program. They found that pro- motion-motivated participants were more persuaded by the article when the argu- ments in it were framed in promotional terms (“because it will advance children’s education and support more children to succeed”), while prevention-motivated participants were persuaded more when the very same arguments were framed in more defensive terms (“because it will secure children’s education and prevent more children from failing”).

In a follow-up study, Cesario and Higgins (2008) found that audience mem- bers are also influenced when a speaker’s nonverbal style fits their motivational orientations. Watching a high school teacher deliver the same communication about a new after-school program, promotion-motivated participants were more receptive when the speaker exhibited an “eager” delivery style (fast, animated, and forward leaning, with hand gestures projecting outward), whereas preven- tion-oriented participants were more receptive when he displayed a cautious style (slow, precise, and backward leaning, with hand gestures pushing in). In studies conducted in France, in the context of messages that advertised healthful diet, Lucia Bosone and others (2015) also found that persuasion is maximized when

80220_ch06_ptg01_hr_208-261.indd 242 27/01/16 5:20 PM

243 Persuasion by Communication

there is a regulatory fit between audience members and a role model whose ap- proach is either positive (“always very healthy thanks to a balanced diet”) or negative (always sick because of an unbalanced diet”).

There are plenty of other ways that you may be more comfortable with some types of messages than others. We saw earlier that some of us are high in the need for cognition, enjoying effortful forms of reasoning and problem solving. Some of us are also high in the need for affect, seeking out and enjoying feelings of strong emotion. In matters of persuasion, these traits lead people to be more receptive to messages that are presented in primarily cognitive or emotional terms (Haddock et al., 2008).

Forewarning and Resistance When our attitudes or values come under attack, we can succumb to the challenge and change the attitude or we can resist it and maintain the attitude. There are different means of resistance. In a series of studies, Julia Jacks and Kimberly Cameron (2003) asked people to describe and rate the ways that they manage to resist persuasion in their at- titudes on abortion or the death penalty. They identified seven strategies, the most common being attitude bolstering (“I think about all the reasons I believe the way I do”) and the least common being source derogation (“I look for faults in the person who challenges my belief”). These means of resistance are listed in Table 6.4.

What leads people to invoke these mechanisms of resistance? Does it help to be forewarned that your attitude is about to come under attack? Perhaps the toughest audience to persuade is the one that knows you’re coming. When people are aware that someone is trying to change their attitude, they become more likely to resist. All they need is some time to collect their thoughts and come up with a good defense. Jonathan Freedman and David Sears (1965) first discovered this when they told high school seniors to expect a speech on why teenagers should not be allowed to drive (an unpopular position, as you can imagine). The students were warned either 2 or 10 minutes before the talk began or not at all. Those who were the victims of a sneak attack were the most likely to succumb to the speaker’s position. Those who had a full 10 minutes’ warning were the least likely to agree. To be forewarned is to be forearmed. But why?

At least two processes are at work here. To un- derstand them, let’s take a closer look at what fore- warning does. Participants in the Freedman and Sears (1965) study were put on notice in two ways: (1) They were informed of the position the speaker would take, and (2) they were told that the speaker intended to change their opinion. Psychologically, these two aspects of forewarning have different effects. The first effect is purely cognitive. Know- ing in advance what position a speaker will take enables us to come up with counterarguments and, as a result, to become more resistant to change. To explain this effect, William McGuire (1964) drew an analogy: Protecting a person’s attitudes from persuasion, he said, is like inoculating the human body against disease. In medicine, injecting a small dose of infection into a patient stimulates the body

m T a b l e 6 . 4

Strategies for Resisting Persuasion

Strategy example

Attitude bolstering “I reassure myself of facts that support the validity of my belief.”

Counterarguing “I would talk to myself and play devil’s advocate.”

Social validation “I also rely on others with the same opinion to be there for me.”

Negative affect “I tend to get angry when someone tries to change my beliefs.”

Assertions of confidence “I doubt anybody could change my viewpoint.”

Selective exposure “Most of the time I just ignore them.”

Source derogation “I look for faults in the person presenting the challenging belief.”

80220_ch06_ptg01_hr_208-261.indd 243 27/01/16 5:20 PM

244 Chapter 6 Attitudes

to build up a resistance to it. According to this inoculation hypothesis, an atti- tude can be immunized the same way. As with flu shots and other vaccines, our defenses can be reinforced by exposure to weak doses of the opposing position before we actually encounter the full presentation. Studies of negative political ads show that inoculation can be used to combat the kinds of attack messages that sometimes win elections (Pfau et al., 1990).

Simply knowing that someone is trying to persuade us also sparks a moti- vational reaction as we brace ourselves to resist the attempt regardless of what position is taken. As a TV viewer, you have no doubt heard the phrase “And now, we pause for a message from our sponsor.” What does this warning tell us? Not knowing yet who the sponsor is, even the grouchiest among us is in no position to object. Yet imagine how you would feel if an experimenter said to you, “In just a few minutes, you will hear a message prepared according to well-established prin- ciples of persuasion and designed to induce you to change your attitudes.” If you are like the participants who actually heard this warning, you might be tempted to reply, “Oh yeah? Try me!” Indeed, participants rejected that message without counterargument and without much advance notice (Hass & Grady, 1975).

When people think that someone is trying to change their attitude or other- wise manipulate them, a red flag goes up. That red flag is called psychological reactance. According to Jack Brehm’s theory of psychological reactance, all of us want the freedom to think, feel, and act as we (not others) choose. When we sense that a cherished freedom is being threatened, we become motivated to maintain it. And when we sense that a freedom is slipping away, we try to restore it (Brehm & Brehm, 1981). One possible result is that when a com- municator comes on too strongly, we may react with negative attitude change by moving in the direction that is the opposite of the one being advocated— even, ironically, when the speaker’s position agrees with our own (Heller et al., 1973). Sometimes, the motive to protect our freedom to think as we choose trumps our desire to hold a specific opinion. Reactance can trigger resistance to persuasion in two ways. Once aroused, the reactant target of at- tempted persuasion may simply shut down in a reflex-like response or disagree in a more thoughtful manner by questioning the credibility of the source and counterarguing the message (Silvia, 2006).

It’s important to realize that forewarning does not always increase resistance to persuasion because the effects are not that simple. Based on a meta-analysis of 48 experiments, Wendy Wood and Jeffrey Quinn (2003) found that when people are forewarned about an impending persuasive appeal on a topic that is person- ally not that important, they start to agree before they even receive the mes- sage so as not to appear vulnerable to influence. Yet when people are forewarned about a persuasive appeal on a topic of personal importance, they feel threatened and think up counterarguments to bolster their attitude. This cognitive response strengthens their resistance to change once that appeal is delivered. It takes men- tal effort to resist a persuasive appeal. When people know that someone will try to influence them, they conserve their energy and generate the necessary counter- arguments (Janssen et al., 2010).

j Culture and Persuasion A communication is persuasive to the extent that the source is favorable and the message, however it is presented, meets the psychological needs of its audience. In this regard, cultural factors also play a subtle but important role.

“To do just the opposite is also a form of imitation.”

—Lichtenberg

inoculation hypothesis The idea that exposure to weak versions of a persuasive argument increases later resistance to that argument.

psychological reactance The theory that people react against threats to their freedom by asserting themselves and perceiving the threatened freedom as more attractive.

80220_ch06_ptg01_hr_208-261.indd 244 27/01/16 5:20 PM

245 Persuasion by Our Own Actions

In earlier chapters, we saw that cultures differ in the extent to which people are oriented toward individualism or collectivism. In light of these differences, Sang-Pil Han and Sharon Shavitt (1994) compared the contents of magazine advertisements in the United States, an individualistic country, and Korea, a country with a more collectivistic orientation. They found that American advertising campaigns were focused more on personal benefits, individuality, competition, and self-improvement (“She’s got a style all her own”; “Make your way through the crowd”), and that Korean ads appealed more to the integrity, achievement, and well-being of one’s ingroups (“An exhilarating way to provide for your family”; “Celebrating a half-century of partnership”). Clearly, there are different ways to appeal to the members of these two cultures.

In a second study, Han and Shavitt created two sets of ads for vari- ous products. One set portrayed individuals (“Treat yourself to a breath- freshening experience”), and the other set featured groups (“Share this breath-freshening experience”). Both sets were presented to American and Korean participants. The result: Americans were persuaded more by individualistic ads, and Koreans preferred collectivistic ads. Similar differences are found in the way celebrity endorsements are used in the two cultures. In the United States, celebrities tend to portray them- selves using or talking directly about a product; in Korean commercials that appeal to belongingness, family, and traditional values, celebrities are more likely to play the role of someone else without being singled out (Choi et al., 2005).

As people from all over the world come into contact with each other through travel, satellite television, international trade agreements, and the Internet, cultural values begin to change. Just as humans develop as they get older, cultures sometimes change over time from one gen- eration to the next. Recent and substantial modernization efforts in China—home to roughly one out of every five people on the planet—il- lustrate the point. There is so much recent change that Zhang and Shavitt (2003) sought to compare the contents of television commercials, which are primarily directed at the traditional mass market, with advertise- ments in new magazines that specifically target 18- to 35-year-old, ed- ucated, high-income citizens who constitute China’s “X-generation.” Based on their analysis of 463 ads, they found that although traditional and collectivist values predominated on mainstream TV, magazine ads were characterized by more modern and individualistic impulses. To be persuasive, a message should appeal to the culturally shared values of its audience.

In a series of print ads, Apple Computer featured Albert Einstein and other creative geniuses throughout history who dared to “think different.” In a marketing campaign that paid tribute to individualism, Apple saluted “The crazy ones. The misfits. The rebels. The troublemakers. The round pegs in the square holes. The ones who see things differently.”

Im ag

e Co

ur te

sy o

f T he

A dv

er tis

in g

A rc

hi ve

Persuasion by Our Own Actions Anyone who has ever acted on stage knows how easy it is to become so absorbed in a role that the experience seems real. Feigned laughter can make an actor feel happy, and crocodile tears can turn into sadness. Even in the high stakes of real life, the effect can be dramatic. In 1974, Patty Hearst, a shel- tered college student from a wealthy California family, was kidnapped. By the

80220_ch06_ptg01_hr_208-261.indd 245 27/01/16 5:20 PM

246 Chapter 6 Attitudes

time she was arrested months later, she was a gun-toting revolutionary who called herself Tania. How could someone be so totally converted? In Hearst’s own words, “I had thought I was humoring [my captors] by parroting their cli- chés and buzzwords without believing in them . . . In trying to convince them I convinced myself” (Hearst, 1982).

j Role Playing: All the World’s a Stage The Patty Hearst case illustrates the powerful effects of role playing. Of course, you don’t have to be kidnapped or terrorized to know how it feels to be coaxed into behavior that is at odds with your sense of who you are. People frequently engage in attitude-discrepant behavior as part of a job, for example, or to fit into a group. As commonplace as this seems, it raises a profound question. When we play along, saying and doing things that are privately discrepant from our own attitudes, do we begin to change those attitudes as a result? How we feel can determine the way we act. Is it also possible that the way we act can determine how we feel?

Many years ago, Irving Janis (1968) theorized that attitude change would persist more when it is inspired by our own behavior than when it stems from a passive exposure to a persuasive communication. Janis conducted a study in which one group of participants listened to a speech that challenged their posi- tions on a topic and others were handed an outline and asked to give the speech themselves. As predicted, participants changed their attitudes more after giving the speech than after listening to it (Janis & King, 1954). According to Janis, role playing works to change attitudes because it forces people to learn the message.

People tend to remember arguments they come up with on their own better than they remember argu- ments provided to them by other people (Slamecka & Graff, 1978). In fact, attitude change is more en- during even when people who read a persuasive message merely expect that they will later have to communicate it to others (Boninger et al., 1990).

But there’s more to role playing than improved memory. The effects of enacting a role can be stag- gering, in part because it is so easy to confuse what we do or what we say with how we really feel. Think about the times you’ve dished out compli- ments you didn’t mean or flashed a smile at some- one you didn’t like or nodded your head in response to a statement you disagreed with. We often shade what we say just to please a particular listener. What’s fascinating is not that we make adjustments to suit others but that this role playing has such powerful effects on our own private attitudes. For example, participants in one study read about a man and then described him to someone else, who supposedly liked or disliked him. As you might ex- pect, they described the man in more positive terms when their listener was favorably disposed. In the

process, however, they also convinced themselves. At least to some extent, “say- ing is believing” (Higgins & Rholes, 1978).

Recent research into this process of self-generated persuasion has shown that more attitude change is produced by having people generate arguments themselves

Posing as a revolutionary named Tania, Patty Hearst was converted by the role her captors forced her to play. “In trying to convince them I convinced myself,” she said.

Be tt

m an

n/ Co

rb is

80220_ch06_ptg01_hr_208-261.indd 246 27/01/16 5:20 PM

247 Persuasion by Our Own Actions

than listen passively to others making the same arguments. Often in life, people hold attitudes they wish they could change—as when we wish we could like a job more than we do, or our favorite junk foods less. In these instances, people may try to talk themselves deliberately into something they do not already believe (Maio & Thomas, 2007). This research raises an interesting question: What produces more change in attitude, formulating arguments in order to convince yourself of something, or trying to convince someone else? The answer is, it depends. In a series of experi- ments, researchers found that when college students were instructed to advocate for a policy that was largely consistent with their own attitudes (that their university’s tuition should be lowered), more self-persuasion occurred when their intended audience was another student. Not knowing how their target felt about the issue, these students tried harder to generate good arguments and moved themselves in the process. However, when instructed to advocate for a position that they opposed (that tuition should be raised), more self-persuasion occurred among students who had sought to convince themselves as opposed to another student. Knowing full well that self-persuasion on the issue would be particularly chal- lenging, these students tried harder and generated better arguments (Briñol et al., 2012).

The kind of self-persuasion that occurs when people advocate for something, as when they play a role, is a fascinating process. Consider the implications. We know that attitudes influence behavior, as when people help those whom they like and hurt those whom they dislike. But research on role playing emphasizes the flip side of the coin—that behavior can change attitudes. Perhaps we come to like people because we have helped them and dislike people whom we have hurt. To change a person’s inner feelings, then, maybe we should begin by focus- ing on their behavior. Why do people experience changes of attitude in response to changes in their own behavior? One answer to this question is provided by the theory of cognitive dissonance.

j Cognitive Dissonance Theory: The Classic Version Many social psychologists believe that people are strongly motivated by a desire for cognitive consistency—a state of mind in which one’s beliefs, attitudes, and behaviors are compatible with each other (Abelson et al., 1968). Cognitive con- sistency theories seem to presuppose that people are generally logical. However, Leon Festinger (1957) turned this assumption on its head. Struck by the irrational- ities of human behavior, Festinger proposed cognitive dissonance theory, which states that a powerful motive to maintain cognitive consistency can give rise to irrational, sometimes maladaptive behavior.

According to Festinger, all of us hold many cognitions about ourselves and the world around us. These cognitions include everything we know about our own beliefs, attitudes, and behavior. Although generally our cognitions coex- ist peacefully, at times they clash. Consider some examples. You say you’re on a diet, yet you just dove headfirst into a tub of chocolate fudge brownie ice cream. Or you waited in line for hours to get into a concert, and then the band proved to be disappointing. Or you baked for hours under the hot summer sun while listening to your iPod even though you knew the health risks. Each of these scenarios harbors inconsistency and conflict: You’ve already committed yourself to a course of action, yet you realize that the action is inconsistent with your attitude.

Under certain conditions, discrepancies such as these can evoke an unpleas- ant state of tension known as cognitive dissonance. But discrepancy doesn’t

“If you don’t like something, change it. If you can’t change it, change your attitude.”

—Maya Angelou

cognitive dissonance theory Theory holding that inconsistent cognitions arouses psychological tension that people become motivated to reduce.

80220_ch06_ptg01_hr_208-261.indd 247 27/01/16 5:20 PM

248 Chapter 6 Attitudes

always produce dissonance. If you broke a diet for a Thanksgiving dinner with family, your indiscretion would not lead you to experience dissonance. Or if you mistakenly thought the ice cream you ate was low in calories only to find out the truth later, then, again, you would not experience much dissonance. As we’ll see, what really hurts is the knowledge that you committed yourself to an attitude-discrepant behavior freely and with some knowledge of the consequences. When that happens, dissonance is aroused, and you become motivated to reduce it. As shown in Table 6.5, there are many possible ways to reduce dissonance—such as rationalizing that others in one’s ingroup are also hypocrites (McKimmie et al., 2009), denying personal respon- sibility for the behavior (Gosling et al., 2006), and trivializing the issue in question (Starzyk et al., 2009). Of course, sometimes the easiest way to re- duce dissonance is to change your attitude to bring it in line with your behavior.

Right from the start, cognitive dissonance theory captured the imagination of social psychology. Festinger’s basic proposition is simple, yet its implications are far-reaching. In this section, we examine three research areas that demonstrate the breadth of what dissonance theory has to say about attitude change.

Justifying Attitude-Discrepant Behavior: When Doing is Believing Imagine for a moment that you are a participant in the classic study by Leon Festinger and J. Merrill Carlsmith (1959). As soon as you arrive, you are greeted by an experi- menter who says that he is interested in various measures of performance. Wonder- ing what that means, you all too quickly find out. The experimenter hands you a wooden board containing 48 square pegs in square holes and asks you to turn each peg a quarter turn to the left, then a quarter turn back to the right, then back to the left, then back again to the right. The routine seems endless. After 30 minutes, the experimenter comes to your rescue. Or does he? Just when you think things are look-

ing up, he hands you another board, another assign- ment. For the next half-hour, you are to take 12 spools of thread off the board, put them back, take them off, and put them back again. By now, you’re just about ready to tear your hair out. As you think back over better times, even the first task begins to look good.

Finally, you’re done. After one of the longest hours of your life, the experimenter lets you in on a secret: There’s more to this experiment than meets the eye. You were in the control group. To test the effects of motivation on performance, other partici- pants are being told that the experiment will be fun and exciting. You don’t realize it, but you are now being set up for the critical part of the study. Would you be willing to tell the next participant that the ex-

periment is enjoyable? As you hem and haw, the experimenter offers to pay for your services. Some participants are offered $1; others are offered $20. In either case, you agree to help out. Before you know it, you find yourself in the waiting

One way to reduce dissonance is to minimize the importance of the conflict.

“It’s a crazy idea, but it just might work.”

Re pr

in te

d fr

om t

he W

al l

St re

et J

ou rn

al .

Pe rm

is si

on ,

Ca rt

oo n

Fe at

ur es

S yn

di ca

m T a b l e 6 . 5

Ways to Reduce Dissonance

Techniques examples

Change your attitude. “I don’t really need to be on a diet.”

Change your perception of the behavior.

“I hardly ate any ice cream.”

Add consonant cognitions. “Chocolate ice cream is very nutritious.”

Minimize the importance of the conflict.

“I don’t care if I’m overweight—life is short!”

Reduce perceived choice. “I had no choice; the ice cream was served for this special occasion.”

“I need to be on a diet, yet I just dove head first into a tub of chocolate fudge brownie ice cream.” If this were you, how would you reduce dissonance aroused by the discrepancy between your attitude and your behavior?

80220_ch06_ptg01_hr_208-261.indd 248 27/01/16 5:20 PM

249 Persuasion by Our Own Actions

room trying to dupe an unsuspecting fellow student (who is really a confederate).

By means of this elaborate staged presentation, partici- pants were goaded into an attitude-discrepant behavior, an action that was inconsistent with their private attitudes. They knew how dull the experiment really was, yet they raved about it. Did this conflict arouse cognitive disso- nance? It depends on how much the participants were paid. Suppose you were one of the lucky ones offered $20 for your assistance. By today’s standards, that payment would be worth $80—surely a sufficient justification for telling a little white lie, right? Feeling well compensated, these par- ticipants experienced little if any dissonance. But wait. Sup- pose you were paid only $1. Surely your integrity is worth more than that, don’t you think? In this instance, you have insufficient justification for going along, so you need a way to cope. According to Festinger (1957), unless you can deny your actions (which is not usually possible), you’ll feel pressured to change your attitude about the task. If you can convince yourself that the experiment wasn’t that bad, then saying it was interesting is all right.

The results were as Festinger and Carlsmith had pre- dicted. When the experiment was presumably over, partici- pants were asked how they felt about the peg-board tasks. Those in the control group who did not mislead a confeder- ate openly admitted that the tasks were boring. So did those in the $20 condition, who had ample justification for what they did. However, participants who were paid only $1 rated the experiment as somewhat enjoyable. Having engaged in an attitude-discrepant act without sufficient justification, these participants reduced cognitive dissonance by chang- ing their attitude. The results can be seen in d Figure 6.9.

Two aspects of this classic study are noteworthy. First, it showed the phenomenon of self-persuasion: When peo- ple behave in ways that contradict their attitudes, they sometimes go on to change those attitudes without any exposure to a persuasive communication. Demonstrat- ing the power of this phenomenon, Michael Leippe and Donna Eisenstadt (1994) found that white college students who were coaxed into writing essays in favor of new scholarship funds only for black students later reported more favorable attitudes in general toward African Americans. The second major contribution of Festinger and Carlsmith’s results is that they contradicted the time-honored belief that big rewards produce greater change. In fact, the more money participants were offered for their inconsistent behavior, the more justified they felt and the less likely they were to change their attitudes.

Just as a small reward provides insufficient justification for attitude-discrepant behavior, mild punishment is insufficient deterrence for attitude-discrepant non- behavior. Think about it. What happens when people refrain from doing some- thing they really want to do? Do they devalue the activity and convince themselves that they never really wanted to do it in the first place? In one study, children were prohibited from playing with an attractive toy by being threatened with a mild or a severe punishment. All participants refrained. As cognitive dissonance theory predicts, however, only those faced with the mild punishment—an insufficient

d F i g u R e 6 . 9

The Dissonance Classic

Participants in a boring experiment (attitude) were asked to say that it was enjoyable (behavior) to a fellow student. Those in one group were paid $1 to lie; those in a second group were offered $20. Members of a third group, who did not have to lie, admitted that the task was boring. So did the participants paid $20, which was ample justification for telling a lie. Participants paid only $1, however, rated the task as more enjoyable. Behaving in an attitude-discrepant manner without justification, these latter participants reduced dissonance by changing their attitude. Based on Festinger, L. and Carlsmith, J. M., “Cognitive Conse- quences and Forced Compliance,” Journal of Abnormal and Social Psychology vol 58 (pp. 203–210). Copyright © 1959 by the American Psychological Association. Reprinted by permission.

No lie $20 lie $1 lie

Ra tin

g of

ta sk

e nj

oy m

en t

insufficient justification A condition in which people freely perform an attitude-discrepant behavior without receiving a large reward.

insufficient deterrence A condition in which people refrain from engaging in a desirable activity, even when only mild punishment is threatened.

80220_ch06_ptg01_hr_208-261.indd 249 27/01/16 5:20 PM

250 Chapter 6 Attitudes

deterrent—later showed disdain for the forbidden toy. Those who confronted the threat of severe punishment did not (Aronson & Carlsmith, 1963). Once again, cognitive dissonance theory turned common sense on its head: The less severe the threatened punishment, the greater the attitude change produced.

Justifying effort: Coming to like What We Suffer For Have you ever spent tons of money or tried really hard to achieve something, only to discover later that it wasn’t worth all the effort? This kind of inconsistency between effort and outcome can arouse cognitive dissonance and motivate a change of heart toward the unsatisfying outcome. The hypothesis is simple but profound: We alter our attitudes to justify our suffering.

In a classic test of this hypothesis, Elliot Aronson and Judson Mills (1959) invited female students to take part in a series of group discussions about sex. But there was a hitch. Because sex is a sensitive topic, participants were told that they would have to pass an “embarrassment test” before joining the group. The test consisted of reading sexual material aloud in front of a male experimenter. One group of participants experienced what amounted to a severe initiation in which they had to recite obscene words and lurid passages taken from paperback nov- els. A second group underwent a mild initiation in which they read a list of more ordinary words pertaining to sex. A third group was admitted to the discussions without an initiation test.

Moments later, all participants were given headphones and permitted to eaves- drop on the group they would soon be joining. Actually, what they heard was a tape-recorded discussion about “secondary sex behavior in the lower animals.” It was dreadfully boring. When it was over, participants were asked to rate how much they liked the group members and their discussion. Keep in mind what dis- sonance theory predicts: The more time or money or effort you choose to invest in something, the more anxious you will feel if the outcome proves disappointing. One way to cope with this inconsistency is to alter your attitudes. That’s exactly

what happened. Participants who had endured a severe initiation rated the discussion group more favorably than did those who had endured little or no initiation.

Social embarrassment is not the only kind of “effort” we feel we need to justify to ourselves. As a general rule, the more you pay for something—whether you pay in physical exertion, pain, time, or money—the more you will come to like it. This principle has provocative im- plications for hazing practices in fraternities and sorori- ties, on sports teams, and in the military. Research even suggests that the harder psychotherapy patients have to work at their own treatment, the more likely they are to feel better when that treatment is over (Axsom, 1989; Axsom & Cooper, 1985).

Justifying Difficult Decisions: When good Choices get even Better Whenever we make diffi- cult decisions—whether to marry, what school to attend, where to live, or what job to accept—we feel dissonance. By definition, a decision is difficult when the alternative

courses of action are about equally desirable. Marriage offers comfort and stability; staying single enables us to seek out exciting new relationships. One job might pay more money; the other may offer more interesting work. Once people make tough

The more money you pay people to tell a lie, the more they will come to believe it.

FAlSe

Cognitive dissonance theory would predict that this character, if he survives, will come to like the “frat” - a lot.

Co nd

e N

as te

80220_ch06_ptg01_hr_208-261.indd 250 27/01/16 5:20 PM

251 Persuasion by Our Own Actions

decisions like these, they are at risk because as negative aspects of the chosen alter- natives and positive aspects of the unchosen alternatives are at odds with their deci- sions. According to dissonance theory, people rationalize whatever they decide by exaggerating the positive features of the chosen alternative and the negative features of the unchosen alternative.

In an early test of this hypothesis, Jack Brehm (1956) asked female partici- pants to evaluate various consumer products, presumably as part of a marketing research project. After rating a toaster, a coffee pot, a radio, a stopwatch, and other products, participants were told that they could take one home as a gift. In the high-dissonance condition, they were offered a difficult choice between two items they found equally attractive. In the low-dissonance group, they were of- fered an easier choice between a desirable and an undesirable item. After receiv- ing the gift, participants read a few research reports and then reevaluated all the products. The results provided strong support for dissonance theory. In the low- dissonance group, the participants’ post-decision ratings were about the same as their pre-decision ratings. But in the high-dissonance condition, ratings increased for the chosen item and decreased for the item that was not chosen. Participants torn between two equivalent alternatives coped by reassuring themselves that they had made the right choice.

This phenomenon appears in a wide range of settings. For example, Robert Knox and James Inskter (1968) took dissonance theory to the racetrack and found that bettors who had already placed $2 bets on a horse were more optimistic about winning than were those still standing in line. This type of optimism may even begin to set in once a thoughtful decision is made, even before the bet is placed (Brownstein et al., 2004). Similarly, Dennis Regan and Martin Kilduff (1988) vis- ited several polling stations on Election Day and found that voters were more likely to think that their candidates would win when they were interviewed after submitting their ballots than when they were interviewed before they submitted them. Since bets and votes cannot be taken back, people who had committed to a decision were moti- vated to reduce post-decision dissonance. So they convinced themselves that the decision they made was right.

j Cognitive Dissonance Theory: A New look

Following in Festinger’s bold footsteps, gen- erations of social psychologists have studied and refined the basic theory (Cooper, 2007; Harmon-Jones & Mills, 1999). Nobody dis- putes the fact that when people are gently coaxed into performing an attitude-discrepant behavior, they often go on to change their attitudes. In fact, people will feel discomfort and change their attitudes when they disagree with others in a group (Matz & Wood, 2005) or even when they observe inconsistent behavior from others with whom they identify—a process of vicari- ous dissonance (Cooper & Hogg, 2007).

Researchers have also examined possible perceptual consequences of cogni- tive dissonance. In one study, Emily Balcetis and David Dunning (2007) took col- lege students to the crowded center of campus and asked them to put on—and

People often come to like what they suffer for.

TRue

Suggesting that people need to justify difficult irrevocable decisions to quell the dissonance they arouse, researchers found that gamblers who had already bet on a horse rated themselves as more certain of winning than those who were still waiting to place a bet.

W ill

ia m

M an

ni ng

/T er

ra /C

or bi

80220_ch06_ptg01_hr_208-261.indd 251 27/01/16 5:20 PM

252 Chapter 6 Attitudes

walk around in—a costume consisting of a grass skirt, a coconut bra, a flower lei around the neck, and a plastic fruit basket on the head. Embarrassing as it was to appear this way in public, all the participants walked across campus in this costume. In a high-choice condition, they were led to believe that they could decline in favor of a different task (insufficient justification). In a low-choice con- dition, they were told that no alternative tasks were available (sufficient justifi- cation). How bad was it? Afterward, all the students were asked to estimate the distance they had walked from one point to the other. Needing to justify their embarrassing antics, those in the high-choice condition underestimated how far they had walked relative to those in the low-choice condition. Apparently, the motivation to reduce dissonance can alter our visual representations of the natural environment.

Through systematic research, it became evident early on that Festinger’s (1957) original theory was not to be the last word. People do change their atti- tudes to justify attitude-discrepant behavior, effort, and difficult decisions. But for dissonance to be aroused, certain specific conditions must be present. Thanks to Joel Cooper and Russell Fazio’s (1984) “new look” at dissonance theory, we now have a pretty good idea of what those conditions are.

According to Cooper and Fazio, four steps are necessary for both the arousal and reduction of dissonance. First, the attitude-discrepant behavior must produce unwanted negative consequences. Recall the initial Festinger and Carlsmith (1959) study. Not only did participants say something they knew to be false but they also deceived a fellow student into taking part in a painfully boring experiment. Had these participants lied without causing hardship, they would not have changed their attitudes to justify the action (Cooper et al., 1974). To borrow an expression from schoolyard basketball, “no harm, no foul.” In fact, negative consequences can arouse dissonance even when people’s actions are consistent with their attitudes, as when college students who wrote against fee hikes were led to be- lieve that their essays had backfired, prompting a university committee to favor an increase (Scher & Cooper, 1989).

The second necessary step in the process is a feeling of personal responsibility for the unpleasant outcomes of behavior. Personal responsibility consists of two factors. The first is the freedom of choice. When people believe they had no choice but to act as they did, there is no dissonance and no attitude change (Linder et al., 1967). Had Festinger and Carlsmith coerced participants into raving about the boring experiment, the participants would not have felt the need to further justify what they did by changing their attitudes. But the experimental situation led participants to think that their actions were voluntary and that the choice was theirs. Participants were pressured without realizing it and believed that they did not have to comply with the experimenter’s request.

For people to feel personally responsible, they must also believe that the potential negative consequences of their actions were foreseeable at the time (Goethals et al., 1979). When the outcome could not realistically have been an- ticipated, then there is no dissonance and no attitude change. Had Festinger and Carlsmith’s participants lied in private and found out only later that their state- ments had been tape-recorded for subsequent use, then, again, they would not have felt the need to further justify their behavior.

The third necessary step in the process is physiological arousal. Right from the start, Festinger viewed cognitive dissonance as a state of discomfort and tension that people seek to reduce—much like hunger, thirst, and other basic drives. Re- search has shown that this emphasis was well placed. In a study by Robert Croyle and Joel Cooper (1983), participants wrote essays that supported or contradicted

80220_ch06_ptg01_hr_208-261.indd 252 27/01/16 5:20 PM

253 Persuasion by Our Own Actions

their own attitudes. Some were ordered to do so, but others were led to believe that the choice was theirs. During the session, electrodes were attached to each participant’s fingertips to record levels of physiological arousal. As predicted by cognitive dissonance theory, those who freely wrote attitude-discrepant essays were the most aroused, an observation made by other researchers as well (Elkin & Leippe, 1986). In fact, participants who write attitude-discrepant essays in a “free- choice” situation report feeling high levels of discomfort—which subside once they change their attitudes (Elliot & Devine, 1994).

The fourth step in the dissonance process is closely related to the third. It isn’t enough to feel generally aroused. A person must also make an attribution for that arousal to his or her own behavior. Suppose you just lied to a friend or studied for an exam that was canceled or made a tough decision that you might soon regret. Suppose further that although you are upset, you believe that your discomfort is caused by some external factor, not by your dissonance-producing behavior. Under these circumstances, will you exhibit attitude change as a symptom of cognitive dissonance? Probably not. When participants were led to attribute their dissonance-related arousal to a drug they had supposedly taken (Zanna & Cooper, 1974), to the anticipation of painful electric shocks (Pittman, 1975), or to a pair of prism goggles that they had to wear (Losch & Cacioppo, 1990), attitude change did not occur. d Figure 6.10 summarizes these four steps in the production and reduction of dissonance.

To this day, social psychologists continue to debate the “classic” and “new look” theories of cognitive dissonance. On the one hand, research has shown that attitude-discrepant actions do not always produce dissonance, in part because not everyone cares about being cognitively consistent (Cialdini et al., 1995) and in part because a change in attitude often seems to require the production of nega- tive consequences (Johnson et al., 1995). On the other hand, some researchers have found that mere inconsistency can trigger cognitive dissonance, even with- out negative consequences. For example, Eddie Harmon-Jones and others (1996) had people drink a Kool-Aid beverage that was mixed with sugar or vinegar. They either told the participants (no choice) or asked them (high choice) to state in writing that they liked the beverage and then toss these notes, which were not really needed, into the wastebasket. Afterward, they rated how much they really liked the drink. You may have noticed that this experiment parallels the Festinger and Carlsmith study with one key exception: For participants in the high-choice situation who consumed vinegar and said they liked it, the lie—although it

d F i g u R e 6 . 1 0

Necessary Conditions for the Arousal and Reduction of Dissonance

Research suggests that four steps are necessary for attitude change to result from the production and reduction of dissonance.

Behavior

Step 1 Unwanted negative consequence

Step 2 Personal responsibility

Step 3 Physiological arousal

Step 4 Attribution of arousal to behavior

Attitude change

Antecedent Conditions that Produce Discomfort

Physiological Arousal and Its Interpretation

+ + +

en ga

ge L

ea rn

in g®

80220_ch06_ptg01_hr_208-261.indd 253 27/01/16 5:20 PM

254 Chapter 6 Attitudes

contradicted their true attitudes—did not cause harm to anyone. Did they experi- ence dissonance that they would have to reduce by overrating the vinegar Kool- Aid? Yes. Compared with participants who lied about the vinegar in the no-choice situation, those in the high-choice situation rated its taste as more pleasant. The lie was harmless, but the feeling of inconsistency still forced a change in attitude.

j Alternative Routes to Self-Persuasion It is important to distinguish between the empirical facts uncovered by dissonance researchers and the theory that is used to explain them. The facts themselves are clear: Under certain conditions, people who behave in attitude-discrepant ways go on to change their attitudes. Whether this phenomenon reflects a human need to reduce dissonance, however, is a matter of some controversy. Over the years, three other explanations have been proposed.

Self-Perception Theory Daryl Bem’s (1965) self-perception theory, described in Chapter 3, posed the first serious challenge to dissonance theory. Noting that we don’t always have firsthand knowledge of our own attitudes, Bem proposed that we infer how we feel by observing ourselves and the circumstances of our own behav- ior. This sort of self-persuasion is not fueled by the need to reduce tension or justify our actions. Instead, it is a cool, calm, and rational process in which people interpret ambiguous feelings by observing their own behavior. But can Bem’s theory replace dissonance theory as an explanation of self-persuasion?

Bem confronted this question head-on. What if neutral observers who are not motivated by the need to reduce dissonance were to read a step-by-step descrip- tion of a dissonance study and predict the results? This approach to the problem was ingenious. Bem reasoned that observers can have the same behavioral in- formation as the participants themselves but not experience the same personal conflict. If observers generate the same results as real participants, it shows that dissonance arousal is not necessary for the resulting changes in attitudes.

To test his hypothesis, Bem (1967) described the Festinger and Carlsmith study to observers and had them guess participants’ attitudes. Some were told about the $1 condition, some were told about the $20 condition, and others read about the control group procedure. The results closely paralleled the original study. As ob- servers saw it, participants who said the task was interesting for $20 didn’t mean it; they just went along for the money. But those who made the claim for only $1 must have been sincere. Why else would they have gone along? As far as Bem was concerned, the participants themselves reasoned the same way. No conflict, no arousal—just inference by observation.

So should we conclude that self-perception, not dissonance, is what’s necessary to bring about attitude change? That’s a tough question. It’s not easy to come up with a critical experiment to distinguish between the two theories. Both predict the same results, but for different reasons. And both offer unique support for their own points of view. On the one hand, Bem’s observer studies show that dissonance-like results can be obtained without arousal. On the other hand, the participants of dis- sonance studies do experience arousal, which seems necessary for attitude change to take place. Can we say that one theory is right and the other wrong?

Fazio and others (1977) concluded that both theories are right but in differ- ent situations. When people behave in ways that are strikingly at odds with their attitudes, they feel the unnerving effects of dissonance and change their attitudes to rationalize their actions. When people behave in ways that are not terribly dis- crepant from how they feel, however, they experience relatively little tension and

80220_ch06_ptg01_hr_208-261.indd 254 27/01/16 5:20 PM

255 Persuasion by Our Own Actions

form their attitudes as a matter of inference. In short, highly discrepant behavior produces attitude change through dissonance, whereas slightly discrepant behav- ior produces change through self-perception.

impression-Management Theory Another alternative to a dissonance view of self-persuasion is impression-management theory, which says that what matters is not a motive to be consistent but a motive to appear consistent. Nobody wants to be called fickle or be seen by others as a hypocrite. So we calibrate our attitudes and behaviors publicly in order to present ourselves to others in a particular light (Baumeister, 1982; Tedeschi et al., 1971). Or perhaps we are motivated not by a de- sire merely to appear consistent but by a desire to avoid being held responsible for the bad consequences of our actions (Schlenker, 1982). Either way this theory places the emphasis on our concern for self-presentation. According to this view, partici- pants in the Festinger and Carlsmith study mostly did not want the experimenter to think they had sold out for a paltry sum of money.

If the impression-management approach is correct, then cognitive dissonance does not produce attitude change at all—only reported change. In other words, if research participants were to state their attitudes anonymously or if they were to think that the experimenter could determine their true feelings through covert mea- sures, then dissonance-like effects should vanish. Sometimes the effects do vanish, but other times they do not. In general, studies have shown that although self- persuasion can be motivated by impression management, it can also occur in situations that do not clearly arouse self-presentation concerns (Baumeister & Tice, 1984).

Self-esteem Theories A third competing explanation relates self-persuasion to the self. According to Elliot Aronson, acts that arouse dissonance do so because they threaten the self-concept, making the person feel guilty, dishonest, or hypocriti- cal, and motivating a change in attitude or future behavior (Aronson, 1999; Stone, Wiegand, Cooper, & Aronson et al., 1997). This being the case, perhaps Festinger and Carlsmith’s participants needed to change their attitudes toward the boring task in order to repair damage to the self, not to resolve cognitive inconsistency.

If cognitive dissonance is aroused only by behavior that lowers self-esteem, then people with already low expectations of themselves should not be affected: “If a person conceives of himself as a ‘schnook,’ he will expect to behave like a schnook” (Aronson, 1969, p. 24). In fact, Jeff Stone (2003) found that when college students were coaxed into writing an essay in favor of a tuition increase (a position that contradicted their attitude) and into thinking about their own standards of be- havior, those who had high self-esteem changed their attitude to meet their behav- ior, as dissonance theory would predict, more than those who had low self-esteem.

Claude Steele (1988) took this notion two steps further. First, he suggested that a dissonance-producing situation—engaging in attitude-discrepant behavior, exerting wasted effort, or making a tough decision—sets in motion a process of self-affirmation that serves to revalidate the integrity of the self-concept. Second, this revalidation can be achieved in many ways, not just by resolving dissonance. Self-affirmation theory makes a unique prediction: If the active ingredient in dis- sonance situations is a threat to the self, then people who have an opportunity to affirm the self in other ways will not suffer from the effects of dissonance. Give Festinger and Carlsmith’s $1 participants a chance to donate money, help a victim in distress, or solve a problem, and their self-concepts should bounce back with- out further need to justify their actions.

Research provides support for this hypothesis. For example, Steele and others (1993) gave people positive or negative feedback about a personality test they had

80220_ch06_ptg01_hr_208-261.indd 255 27/01/16 5:20 PM

256 Chapter 6 Attitudes

taken. Next, they had them rate 10 popular music CDs and then offered them a choice of keeping either their fifth- or sixth-ranked CD. Soon after making the decision, par- ticipants were asked to rate the CDs again. As predicted by dissonance theory, most inflated their ratings of the chosen CD relative to the unchosen one. The key word, however, is most. The ratings of the participants who had received positive feedback did not change. Why not? According to Steele, it was because they had just enjoyed a self-affirming experience that was enough to overcome the need to reduce dissonance.

Steele’s research suggests that there are many possible ways for people to repair a dissonance-damaged self. But if these efforts at indirect self-affirmation fail, would cognitive dissonance return and create pressure to make a change in attitude? Yes. In one study, college students were asked (high-choice) or told (low-choice) to deliver an attitude-discrepant speech advocating that a popular campus tradition (running nude on the evening of the first snowfall) be banned. For those in the basic high-choice condition, cognitive dissonance was aroused, creating pressure on them to change their attitude favoring the ban. Students in another high-choice group who were subsequently given an opportunity to self- affirm by expressing some cherished values felt less discomfort and exhibited less attitude change. For them, self-affirmation provided the necessary relief. However, among students in a fourth group (also high in choice) who self-affirmed but then received negative feedback about the values they expressed, cognitive dissonance returned, creating pressure to change their attitude toward the ban. In essence, cognitive dissonance and its impact on attitudes reemerged from the failed at- tempt at self-affirmation (Galinsky et al., 2000).

To summarize, cognitive dissonance theory states that people come to change their attitudes to justify attitude-discrepant behaviors, efforts, and decisions. Self- perception theory argues that the change occurs because people infer how they feel by observing their own behavior. Impression-management theory claims that the attitude change is spurred by concerns about self-presentation. And self- affirmation theory says that the change is motivated by threats to the self-concept (see d Figure 6.11). To this day, new theories are being proposed—for example,

d F i g u R e 6 . 1 1

Theories of Self-Persuasion: Critical Comparisons

Here we compare the major theories of self-persuasion. Each alternative challenges a different aspect of dissonance theory. Self-perception theory assumes that attitude change is a matter of inference, not motivation. Impression-management theory maintains that the change is more apparent than real, reported for the sake of public self-presentation. Self-affirmation theory contends that the motivating force is a concern for the self and that attitude change will not occur when the self-concept is affirmed in other ways.

Is the attitude change motivated by a desire to reduce discomfort?

Does a person's private attitude really change?

Must the change be directly related to the attitude- discrepant behavior?

Yes No

Yes

Yes Yes

Yes

Theories

Cognitive Dissonance

Self- Perception

Impression Management

Self- Affirmation

en ga

ge L

ea rn

in g®

80220_ch06_ptg01_hr_208-261.indd 256 27/01/16 5:20 PM

257 Persuasion by Our Own Actions

the notion that feelings of dissonance are aroused not by inconsistency but by the fact that the conflict impedes our ability to know whether to approach or avoid and how to act (Harmon-Jones et al., 2015).

j ethical Dissonance In recent years, in domains as far and wide as business, law, sports, health and education, many social psychologists have become interested in behavioral eth- ics, the study of how individuals behave when facing temptations to cheat, steal, plagiarize, commit fraud, lie, or otherwise behave unethically. From standing in a supermarket express line with too many groceries to overstating business ex- penses in tax forms, our day-to-day lives provide more examples than any of us would care to admit (Bazerman & Gino, 2012).

Within this new field, researchers have addressed two types of problems. The first focuses on the unintentional lapses in ethics that can occur when otherwise good people do not pay attention and are not sufficiently vigilant, causing “blind spots” in ethical judgment (Bazerman & Tenbrunsel, 2011). Research shows that such lapses are most likely to occur when people are fatigued—which is why participants in a series of laboratory experiments were more likely to lie and cheat when tested in the afternoon than in the morning (Kouchaki & Smith, 2014) and when they are overly focused on the temptation of what is to be gained (Pittarello et al., 2015).

The second approach to the study of ethical behavior focuses on acts of in- tentional wrongdoing that people knowingly commit in order to serve their own interests. From a strictly economic standpoint, individuals are thought to engage in unethical acts when the tangible benefits (monetary and other rewards) exceed the tangible costs (exposure and punishment). In the laboratory, research shows that when payment is based on self-reported performance and cheating cannot be detected, participants tend to inflate their claims in order to increase the amount of money they receive. In one study, participants inflated by 15% the number of arithmetic problems they solved (Mazar et al., 2008). In a second study, they lied about random rolls of dice on 40% of occasions (Shalvi et al., 2015). People are particularly prone to cheat for money after seeing a peer confederate they identify with cheat and get away with it (Gino et al., 2009).

Although people can be induced into unethical behavior by the perception of benefits relative to costs, a growing body of research has shown that human nature is complicated by the fact that most people feel badly about their unethi- cal acts even when they do not fear exposure (Hilbig & Hessler, 2013). To put it another way, behaving in ways that violate our own moral code threatens our self-esteem and arouses an inner state of turmoil that Rachel Barkan and her col- leagues (2012) have called ethical dissonance. Following in Festinger’s footsteps, these researchers and others have suggested that when temptation lures us into the possibility of an unethical behavior, our moral self-concept is threatened both before and after we do it. To help attenuate this threat, we use various self-serving justifications to cope with both the anticipation and the experience of ethical dis- sonance. These processes are illustrated in d Figure 6.12.

Social psychologists have begun to explore some of the pre-violation and post-violation types of justifications that enable unethical behavior by helping violators to do wrong but feel moral. Reviewing this research, Shaul Shalvi and others (2015) describe a number of ways in which people reduce their ethical dissonance—for example, citing social norms to suggest that “everyone is do- ing it”; blaming others or circumstances; rationalizing the good that comes from the misdeed; confessing, apologizing, and offering compensation; and distancing

80220_ch06_ptg01_hr_208-261.indd 257 27/01/16 5:20 PM

258 Chapter 6 Attitudes

themselves from the misdeed by asserting stricter ethical standards for the future and judging other transgressors more harshly.

One particularly interesting form of self-justification, which “enables” people to behave unethically, is called moral licensing, a tendency to justify an antici- pated misdeed by citing good things that we have done. In an article entitled “Moral self-licensing: When being good frees us to be bad,” Anna Merritt and her colleagues (2010) reviewed research suggesting that past good deeds can liberate individuals to engage in unethical behaviors that they would otherwise avoid for fear of feeling or appearing immoral. In one study, participants were randomly assigned to write stories that describe their own positive or negative traits. After- ward, they were asked if they would make a small donation to a charity of their choice. Those who wrote positively about themselves later donated $1.07—only a fifth of the $5.30 donated by those who wrote negative (Sachdeva et al., 2009). In a second study, participants were assigned to shop from an online store that carried a predominance of green, environment-friendly products or conventional products equal in price. Moments later, those who shopped at the green store were more likely to lie to the experimenter to increase the money they earned (Mazar & Zhong, 2010). Overall, more than 90 studies have shown that moral licensing op- erates like something moral balancing scale. When people feel as if they have es- tablished moral credentials, they anticipate less ethical dissonance about engaging in an unethical behavior and become more likely to do so (Blanken et al., 2015).

d F i g u R e 6 . 1 2

Ordinary people often behave badly but still feel moral. How is this discrepancy possible? As shown, temptation may lead us to misbehave, which threatens our moral self-concept both beforehand (when ethical dissonance is anticipated) and afterward (when ethical dissonance is experienced). Once tempted, pre-violation justifications excuse what we are about to do. Afterward, post-violation justifications compensate for what we did. In these ways, people manage to behave immorally yet maintain a moral self-concept. Shalvi et al., 2015.

Tempting situation

Moral violation

Moral self-concept threatened

Post-violation justi�cation

(act is compensated for)

Post-violation justi�cation

(act is excusable)

Moral self-concept maintained

80220_ch06_ptg01_hr_208-261.indd 258 27/01/16 5:20 PM

259 Changing Attitudes

j Cultural influences on Cognitive Dissonance

Over the years, social psychologists have presumed that the cognitive dissonance effects uncovered in 50 years of research and described in this chapter are universal and characteristic of human nature. More and more, however, it appears that cultural context may influence both the arousal and reduction of cognitive dissonance.

In Western cultures, individuals are expected to make de- cisions that are consistent with their personal attitudes and to make those decisions free from outside influences. In East Asian cultures, however, individuals are also expected to make deci- sions that benefit their ingroup members and to take the well- being of others into account in making those decisions. In light of these differences, Etsuko Hoshino-Browne and colleagues (2005) compared the reactions of Canadian and Japanese re- search participants in a post-decision dissonance experiment in which they rank-ordered items on a menu by choosing their top 10 dishes. Then they ranked the list again: Half made the choices for themselves, and the others were asked to imagine a close friend whose tastes they knew and choose on behalf of that friend. Did participants show the classic post-decision jus- tification effect, becoming more positive in their ratings of the chosen items relative to nonchosen items? Yes and no. When they made decisions for themselves, only the Canadian partici- pants exhibited a significant justification effect. When Japa- nese participants made decisions for a friend, however, they exhibited the stronger effect (see d Figure 6.13). Similar re- sults have been found in other studies (Kitayama et al., 2004).

To sum up: Cognitive dissonance is both universal and de- pendent on culture. At times everyone feels and tries to reduce dissonance, but cultures influence the conditions under which these processes occur.

d F i g u R e 6 . 1 3

Cognitive Dissonance as Both universal and Culturally Dependent

Researchers compared Canadian and Japanese research participants in a post-decision dissonance study in which they rank ordered items on a menu, chose their top dishes, then ranked the list again. Half made the choices for themselves; the others were asked to imagine a close friend. When deciding for themselves, only the Canadians exhibited a significant justification effect; when deciding for a friend, however, Japanese participants exhibited the stronger effect. Hoshino-Browne et al., 2005.

Canadian Sp

re ad

o f

A lte

rn at

iv es

Self Friend

Japanese 0

0.7

0.6

0.5

0.4

0.3

0.2

0.1

Changing Attitudes Attitudes and attitude change are an important part of social life. In this chapter, we have seen that persuasion can be achieved in different ways. The most com- mon approach is through communication from others. Faced with newspaper edi- torials, junk mail, books, TV commercials, blogs, websites, and other messages, we take one of two routes to persuasion. On the central route, attitude change is based on the merits of the source and his or her communication. On the periph- eral route, it is based on superficial cues. Either way, the change in attitude often precipitates a change in behavior.

A second, less obvious means of persuasion originates within ourselves. When people behave in ways that run afoul of their true convictions, they often go on to change their attitudes. Once again, there are many routes to change, not just one. Cognitive dissonance, self-perception, impression management, and self-esteem concerns are among the possible avenues. From attitudes to behavior and back again, the processes of persuasion are complex and interwoven.

80220_ch06_ptg01_hr_208-261.indd 259 27/01/16 5:20 PM

260 Chapter 6 Attitudes

1. An attitude is an evaluative—positive or negative— reaction to a person, place, issue, or object.

2. Researchers typically measure attitudes through self- report, as in public opinion surveys, or by covert mea- sures of a person’s behavior—most notably, by using the Implicit Association Test (IAT).

3. Research shows individuals may be genetically predis- posed to hold certain attitudes but that our attitudes are also shaped by personal experiences, observations, and associations.

4. The most common approach to changing attitudes—as seen in marketing, politics, advertising, and religion—is through the presentation of a persuasive communication.

5. There are two pathways by which people are persuaded by a communication: (1) on the “central route,” people think critically about the strength and quality of argu- ments, or (2) on the “peripheral route,” people are in- fluenced by surface cues like a speaker’s appearance or reputation.

6. Research shows that people are more likely to change their attitudes in response to a speaker who is likeable and credible; by arguments that are well developed and not too extreme; when motivated by fear; or when re- laxed by positive emotions.

7. To be effective, persuasive messages should present arguments that appeal to the individual and cultural values of audience members.

8. According to Cognitive Dissonance Theory, people of- ten come to change their attitudes in order to justify their own actions—as when they lie for profit, waste effort on something that proves disappointing; or make a difficult decision they start to regret.

9. For more than fifty years now, social psychologists have debated how and why cognitive dissonance leads peo- ple to change their attitudes—for example, whether the change represents a logical inference or a motivation to protect one’s self-esteem.

10. It now appears that people all over the world will try to reduce dissonance when it arises but that the condi- tions that trigger it are influenced by cultural factors.

Top 10 Key Points in Chapter 6

The Study of Attitudes

Researchers can tell if someone has a positive or negative attitude by measuring physiological arousal. F False. Measures of arousal can reveal how intensely

someone feels, but not whether the person’s attitude is positive or negative.

Persuasion by Communication

in reacting to persuasive communications, people are influenced more by superficial images than by logical arguments. F False. As indicated by the dual-process model of

persuasion, people can be influenced by images or arguments, depending on their ability and motivation to think critically about the information.

People are most easily persuaded by commercial messages that are presented without their awareness. F False. There is no research evidence to support the

presumed effects of subliminal ads.

Persuasion by Our Own Actions

The more money you pay people to tell a lie, the more they will come to believe it. F False. Cognitive dissonance studies show that people

believe the lies they are underpaid to tell as a way to justify their own actions.

People often come to like what they suffer for. T True. Studies show that the more people work or

suffer for something, the more they come to like it as away to justify their effort.

Putting Common Sense to the Test

Review

80220_ch06_ptg01_hr_208-261.indd 260 27/01/16 5:21 PM

Review

attitude (210) attitude scale (212) bogus pipeline (212) central route to persuasion (223) cognitive dissonance theory (247)

elaboration (224) evaluative conditioning (218)

facial electromyograph (emG) (214)

Implicit association Test (IaT) (214)

implicit attitude (214) inoculation hypothesis (244) insufficient deterrence (249) insufficient justification (249)

need for cognition (nC) (241) peripheral route to persuasion (223)

persuasion (223) psychological reactance (244) sleeper effect (232) theory of planned behavior (219)

Key Terms

261

80220_ch06_ptg01_hr_208-261.indd 261 27/01/16 5:21 PM

Conformity

This chapter opens by examining ways in which social influences are “automatic.” We then look at three processes. First, we consider the reasons why people exhibit conformity to group norms. Second, we describe the strategies used to elicit compliance with direct requests. Third, we analyze the causes and effects of obedience to the commands of authority. The chapter concludes with a discussion of the continuum of social influence.

80220_ch07_ptg01_hr_262-307.indd 262 27/01/16 11:28 AM

Conformity 7 Social Influence as “Automatic” | 265 Conformity | 268

The Early Classics

Why Do People Conform?

Majority Influence

Minority Influence

Culture and Conformity

Compliance | 283 Mindlessness and Compliance

The Norm of Reciprocity

Setting Traps: Sequential Request Strategies

Assertiveness: When People Say No

Obedience | 291 Milgram’s Research: Forces of Destructive

Obedience

Milgram in the Twenty-First Century

Lingering Questions

Defiance: When People Rebel

The Continuum of Social Influence | 303

Social Impact Theory

Perspectives on Human Nature

Review

263Ri ch

ar d

M cM

an us

/G et

ty Im

ag es

80220_ch07_ptg01_hr_262-307.indd 263 27/01/16 11:29 AM

264 Chapter 7 Conformity

On a Thursday evening in the summer of 2009, a couple of hun- dred ordinary people who were strangers to one another showed up in San Fran- cisco’s Union Square. At precisely 6 p.m., on cue, the impromptu group belted out the Beatles’ song “With a Little Help from My Friends.” Earlier that week, in England, feathers flew outside of the well-known Yorkshire Museum and Gar- dens when 500 happy Facebook users appeared, pillows in hand, for a massive pillow fight. In both cases, participants had received instructions on the Internet, gathered at a set time and place, did a silly but harmless thing, and dispersed. Illustrating the viral power of mobile apps and the Internet to serve as a vehicle for social influence, other “flash mobs” have formed in cities all over the world. In one particularly striking event, at New Hampshire’s Portsmouth High School 2015 graduation ceremony, the valedictorian concluded his speech by saying, “All you have to do is shake it off.” At that point, Taylor Swift’s pop hit “Shake It Off” started to play over the loudspeaker and all 230 graduates got up and danced.

Sometimes the social influences that move us are not entertaining and funny but potentially hazardous to our health. Consider the unusual events that oc- curred a while back in a Tennessee high school. It started when a teacher noticed

a gas-like smell in her classroom and then came down with a headache, nausea, short- ness of breath, and dizziness. Word spread. Others soon reported the same symptoms, and the school was evacuated. Eighty students and several staff members were taken to a local emergency room. Nothing showed up in blood tests, urine tests, or other medical procedures, nor were gases, pesticides, or other toxins detected. What the investigation did turn up was that students who reported feeling ill that day were more likely than others to have seen or heard about someone with symptoms. Re- porting in the New England Journal of Medi- cine, researchers concluded that the problem stemmed from “mass psychogenic illness”—a profound, almost contagious form of social in- fluence (Jones et al., 2000).

In Outbreak! The Encyclopedia of Extraordinary Social Behavior, Hilary Evans and Robert Bartholomew (2009) note that these types of incidents have occurred throughout human history, causing people to break out in rashes, vomit, bark, twirl, faint, strip naked, or laugh uncontrollably. The behaviors may vary. A few years ago, in Le Roy High School in Western New York, 6 girls—a number that grew to 18—developed a number of mysterious motor tics, uncontrollably writh- ing, shaking, twitching, head jerking, arms swinging, grimacing, and stuttering (Dominus, 2012).

Flash mobs and mass psychogenic illnesses reveal the awesome power of social influence. The effects that people have on each other can also be seen in mundane human events. Thus, sports fans spread the “wave” around massive stadiums, sing in unison, or chant “de-fense” in a spectacular show of unison. TV producers insert canned laughter into sitcoms to increase viewer responsiveness. Politicians trumpet inflated results from their own public opinion polls to attract voters. Bartenders, waiters, waitresses, and street musicians stuff dollar bills into their tip jars as a way to get others to follow suit. And online videos that go viral

Online videos that go viral and spread like a contagious disease have the power to attract millions across the internet. In 2015, Google announced that Psy’s 2012 YouTube music video, “Gangnam Style,” officially broke the 2,147,483,647-view barrier, an all-time record.

Th om

as C

oe x/

G et

ty Im

ag es

“We are discreet sheep; we wait to see how the drove is going and then go with the drove.”

—Mark Twain

80220_ch07_ptg01_hr_262-307.indd 264 27/01/16 11:29 AM

265 Social Influence as “Automatic”

Before we consider the explicit forms of social influence depicted in d Figure 7.1, whereby individuals choose whether or not to “go along,” it’s important to note that as social animals humans are vulnerable to a host of subtle, almost reflex-like influences. Without realizing it, we often crack open an involuntary yawn when we see others yawning, laugh aloud when we hear others laughing, and grimace when we see others in pain. In an early demonstration, Stanley Milgram and

spread like a wildfire, capturing millions across the Internet. As they say, “Monkey see, monkey do.”

You don’t need to be a social psychologist to know that people have an impact on each other’s behavior. The trickier question is, how and with what effect? The term social influ- ence refers to the ways that people are affected by the real and imagined pressures of others (Cialdini & Goldstein, 2004; Kiesler & Kiesler, 1969). The kinds of influences brought to bear on an individual come in different shapes and sizes. In this chapter, we look at social influences that are mindless and automatic. Then we consider three forms of influence that vary in the degree of pressure exerted on an individual—conformity, compliance, and obedience. As depicted in Figure 7.1, confor- mity, compliance, and obedience are not distinct, qualitatively different “types” of influence. In all three cases, the influence may emanate from a person, a group, or an institution. In all instances, the behavior in question may be constructive (help- ing oneself or others), destructive (hurting oneself or others), or neutral. It is useful to note, once again, that social influence varies as points along a continuum according to the degree of pressure exerted on the individual. It is also useful to note that we do not always succumb under pressure. People may conform or maintain their independence from others, they may comply with direct requests or react with assertiveness, or they may obey the commands of authority or oppose power- ful others in an act of defiance. In this chapter, we examine the factors that lead human beings to yield to or resist social influence.

Circle Your Answer

T F When all members of a group give an incorrect response to an easy question, most people most of the time conform to that response.

T F An effective way to get someone to do you a favor is to make a first request that is so large the person is sure to reject it.

T F In experiments on obedience, most participants who were ordered to administer severe shocks to an innocent person refused to do so.

T F As the number of people in a group increases, so does the group’s impact on an individual.

T F Conformity rates vary across different cultures and from one generation to the next.

Putting Common SenSe to the Test

Social Influence as “Automatic”

d F i g u r e 7 . 1

Continuum of Social influence

Social influences vary in the degree of pressure they bring to bear on an individual. People may (1) conform to group norms or maintain their independence, (2) comply with requests or be assertive, and (3) obey or defy the commands of authority.

Yielding to In�uence Resisting In�uence

en ga

ge L

ea rn

in g®

80220_ch07_ptg01_hr_262-307.indd 265 27/01/16 11:29 AM

266 Chapter 7 Conformity

others (1969) had research confederates stop on a busy street in New York City, look up, and gawk at the sixth-floor window of a nearby building. Films shot from behind the window indicated that about 80% of passersby stopped and gazed up when they saw the confederates.

Rudimentary forms of automatic imitation have been observed in various animal species, such as pigeons, monkeys, hamsters, and fish (Heyes, 2011; Zentall, 2012). There is even evidence to suggest that “cultures” are transmitted through imitation in groups of whales, as when humpback whales off the coast of Maine use lobtail feeding, a technique in which they slam their tail flukes onto the water, then dive and exhale, forming clouds of bubbles that envelop schools of prey fish to be gulped. This complex behavior was first observed in 1980. By 1989 it was measur- ably adopted by 50% of the whale population in that area (Rendell & Whitehead, 2001). Even more recently, researchers using a “network-based diffusion analysis” found that up to 87% of whales that adopted this technique learned it by exposure from other humpbacks (Allen et al., 2013). Similar observations in other species have led animal scientists to suggest that many nonhuman animals form and trans- mit cultures in this manner to succeeding generations (Laland & Galef, 2009).

Do humans similarly imitate one another automatically, without thought, effort, or conflict? It appears that we do. Controlled studies of human infants have shown that some time shortly after birth, babies not only look at faces but (to the delight of parents all over the world) often they mimic simple gestures such as moving the head, pursing the lips, and sticking out the tongue (Meltzoff & Moore, 1977; Ray & Heyes, 2010). Studying 162 infants from 6 to 20 months old, Susan Jones (2007) found that imitation developed at different rates for different behaviors. Using parents as models, she found, for example, that infants mimicked opening the mouth wide, tapping their fingers on a table, and waving bye-bye before they

mimicked clapping hands, flexing their fingers, or putting their hands on the head.

You may not realize it, but we humans unwittingly mimic each other all the time. You might not know you’re doing it, but when you are in a conversation with someone, chances are you are subtly mirroring them as you speak: nodding when they do, sitting back, leaning forward, scratching your face, or crossing your legs. To demonstrate, Tanya Chartrand and John Bargh (1999) set up participants to work on a task with a partner, a confederate who exhib- ited the habit of rubbing his face or shaking his foot. Hidden cameras recording the interaction revealed that without realizing it, participants mimicked these motor behaviors, rubbing their face or shaking a foot to match their partner’s behavior. Chartrand and Bargh dubbed this phenomenon the “chameleon effect,” after the reptile that changes colors according to its physical environment (see d Figure 7.2).

There are two possible reasons for this nonconscious form of imitation. Char- trand and Bargh theorized that such mimicry serves an important social function, that being “in sync” in terms of their pace, posture, mannerisms, facial expressions, tone of voice, accents, speech patterns, and other behaviors enables people to in- teract more smoothly with one another. Accordingly, Chartrand and Bargh (1999)

Among humpback whales off the coast of Maine, “lobtail feeding” (a complex behavior that traps prey fish) was first observed in 1981. Through imitation, it soon spread across the entire whale population in the region.

iS to

ck /G

et ty

Im ag

es P

lu s/

G et

ty Im

ag es

80220_ch07_ptg01_hr_262-307.indd 266 27/01/16 11:29 AM

267 Social Influence as “Automatic”

turned the tables in a second study in which they instructed their confederate to match in subtle ways the mannerisms of some par- ticipants but not others. Sure enough, participants who had been mimicked liked the confederate more than those who had not.

Two additional sets of findings further demonstrate the social benefits of mimicry. First, research shows that people mimic others more when they are highly motivated to affiliate—say, because they are similar to these others or are feeling excluded—than when they are not (Chartrand & Lakin, 2013; Chartrand & van Baaren, 2009; Lakin et al., 2008). Second, research shows that when participants interact with others who exhibit negative, antisocial behaviors— say, in their tone of voice—mimicry backfires and causes the par- ticipants to be perceived unfavorably (Smith-Genthôs et al., 2015).

Social mimicry is so powerful that it can influence us even when the mimicker is not a real person. In a study entitled “digi- tal chameleons,” Jeremy Bailenson and Nick Yee (2005) immersed college students, one at a time, in a virtual reality environment in which they found themselves seated at a table across from an avatar, a human-like person that looked something like a three- dimensional cartoon character. This avatar proceeded to argue that students should be required to carry identification cards at all times for security purposes. In half the sessions, his back-and- forth head movements perfectly mimicked the participant’s head movements at a four-second delay. In the other half, he repeated the head movements of an earlier recorded participant. Very few of the students who were mimicked were aware of it. Yet when later asked about the experience, they rated the avatar as more likable and were persuaded by its speech more if it imitated their head movements than the previous participant.

The human impulse to mimic others may have adaptive social value, but these types of effects can also be found in nonsocial situations. In one study, Roland Neumann and Fritz Strack (2000) had people listen to an abstract philosophical speech that was recited on tape in a happy, sad, or neutral voice. Afterward, participants rated their own mood as more positive when they heard the happy voice and as more negative when they heard the sad voice. Even though the speakers and participants never interacted, the speaker’s emotional state was infectious, an auto- matic effect that can be described as a form of “mood contagion.” The same can be true about the way we mimic the language we hear in other people’s expres- sions and speech styles. To illustrate, Molly Ireland and James Pennebaker (2010) found that college students answering essay questions or working from excerpts of fictional writing tended in subtle ways to match the language style of the target material to which they were exposed—for example, in terms of their use of personal pronouns (such as I, you), conjunctions (such as but, while), and quantifiers (such as many, few).

It is also important to realize that mimicry is a dynamic process, as when two people who are walking together or dancing become more and more coordinated over time. To demonstrate, Michael Richardson and others (2005) sat pairs of

d F i g u r e 7 . 2

The Chameleon effect

Based on the study by Chartrand and Bargh (1999), this graph shows the number of times per minute participants rubbed their face or shook their foot when they were with a confederate who was rubbing or shaking his foot. Adapted from Kassin, S., Psychology 3rd ed. Copyright © 1997.

Participant rubs face

Participant shakes foot

N um

be r

of ti

m es

Confederate rubs face Confederate shakes foot

0.8

0.7

0.6

0.5

0.4

0.3

Often we are not aware of the influence other people have on our behavior.

M ic

k St

ev en

s /T

he N

ew Y

or ke

r C ol

le ct

io n/

Th e

Ca rt

oo n

Ba nk

80220_ch07_ptg01_hr_262-307.indd 267 27/01/16 11:29 AM

268 Chapter 7 Conformity

conformity The tendency to change our perceptions, opinions, or behavior in ways that are consistent with group norms.

college students side by side to work on visual problems while swinging a hand- held pendulum as “a distraction task.” The students did not need to be synchro- nized in their swinging tempo to get along or solve the problems. Yet when each could see the other’s pendulum, even without speaking, their tempos gradually converged over time—like two hearts beating as one.

Conformity It is hard to find behaviors that are not in some way affected by exposure to the actions of others. When social psychologists talk of conformity, they specifically refer to the tendency of people to change their perceptions, opinions, and behav- ior in ways that are consistent with group norms.

Using this definition, would you call yourself a conformist or a nonconform- ist? How often do you feel inclined to follow what others are saying or doing? At first, you may deny the tendency to conform and, instead, declare your individu- ality and uniqueness. But think about it. When was the last time you attended a formal wedding dressed in blue jeans or remained seated during the national anthem at a sports event? When was the last time you tweeted an unpopular con- trarian position for others to see? People find it difficult to breach social norms. In an early demonstration of this point, research assistants were recruited to ask subway passengers to give up their seats—a conspicuous violation of the norm of acceptable conduct. Many of the assistants could not carry out their assignment. In fact, some of those who tried it became so anxious that they pretended to be ill just to make their request appear justified (Milgram & Sabini, 1978).

With conformity being so widespread and seemingly so natural to human nature, it is interesting and ironic that research participants in North America who are coaxed into following a group norm will often not admit to being influ- enced. Instead, they try to reinterpret the task and rationalize their behavior as a way to see themselves in as independent (Hornsey & Jetten, 2004). But there is a second reason why people do not see themselves as conformist. In a series of studies, Emily Pronin and others (2007) found that people perceive others to be more conforming than themselves in all sorts of domains—from why they bought an iPad to why they hold a popular opinion. Part of the reason for this asymme- try is that people judge others by their overt behavior and the degree to which it matches what others do, but they tend to judge themselves by focusing inward and introspecting about their thought processes, which blinds them to their own conformity.

People understandably have mixed feelings about conformity. On the one hand, some degree of it is essential if individuals are to maintain communities and coexist peacefully, as when people assume their rightful place in a waiting line. Yet at other times, conformity can have harmful consequences, as when people drink too heavily at parties, cheat on taxes, or tell offensive jokes because they believe others are doing the same. For the social psychologist, the goal is to understand the conditions that promote conformity or independence and the rea- sons for these behaviors.

j The early Classics In 1936, Muzafer Sherif published a classic laboratory study of how norms develop in small groups. His method was ingenious. Male students, who believed they

80220_ch07_ptg01_hr_262-307.indd 268 27/01/16 11:29 AM

269 Conformity

were participating in a visual perception experi- ment, sat in a totally darkened room. Fifteen feet in front of them, a small dot of light appeared for two seconds, after which participants were asked to estimate how far it had moved. This procedure was repeated several times. Although participants didn’t realize it, the dot of light always remained motionless. The movement they thought they saw was merely an optical illusion known as the au- tokinetic effect: In darkness, a stationary point of light appears to move, sometimes erratically, in various directions.

At first, participants sat alone and reported their judgments to the experimenter. After several trials, Sherif found that they settled in on their own stable perceptions of movement, with most estimates ranging from 1 to 10 inches (although one participant gave an estimate of 80 feet!). Over the next three days, people returned to participate openly in three-person groups. As before, lights were flashed and the participants, one by one, an- nounced their estimates. As shown in d Figure 7.3, initial estimates varied considerably, but partici- pants later converged on a common perception. Eventually, each group established its own set of norms.

Some 15 years after Sherif’s demonstration, Solomon Asch (1951) constructed a very different task for testing how people’s beliefs affect the beliefs of others. To appreciate what Asch did, imagine yourself in the following situation. You sign up for a psychology experiment, and when you arrive, you find six other students waiting around a table. Soon after you take an empty seat, the experimenter explains that he is interested in the ability to make visual discriminations. As an example, he asks you and the others to indicate which of three comparison lines is identical in length to a standard line.

That seems easy enough. The experimenter then says that after each set of lines is shown, you and the others should take turns announcing your judgments out loud in the order of your seating position. Beginning on his left, the experi- menter asks the first person for his judgment. Seeing that you are in the next-to-last position, you patiently await your turn. The opening moments pass uneventfully. The task and discriminations are clear and everyone agrees on the answers. On the third set of lines, however, the first participant selects what is quite clearly the wrong line. Huh? What happened? Did he suddenly lose his mind, his eyesight, or both? Before you have the chance to figure this one out, the next four participants choose the same wrong line. Now what? Feeling as if you have entered the twilight zone, you wonder if you misunderstood the task. And you wonder what the oth- ers will think if you have the nerve to disagree. It’s your turn now. You rub your eyes and take another look. What do you see? More to the point, what do you do?

d Figure 7.4 gives you a sense of the bind in which Asch’s participants found themselves—caught between the need to be right and a desire to be liked (Insko et al., 1982; Ross et al., 1976). As you may suspect by now, the other “participants” were actually confederates and had been trained to make incorrect judgments on 12 out of 18 presentations. There seems little doubt that the real participants

d F i g u r e 7 . 3

A Classic Case of Suggestibility

This graph, taken from Sherif’s study, shows how three participants’ estimates of the apparent movement of light gradually converged. Before they came together, their perceptions varied considerably. Once in groups, however, participants conformed to the norm that had developed. Sherif, 1936

1I nc

he s

of p

er ce

iv ed

m ov

em en

t Pregroup Group

Session I Group

Session II Group

Session III

Participant CParticipant BParticipant A

80220_ch07_ptg01_hr_262-307.indd 269 27/01/16 11:29 AM

270 Chapter 7 Conformity

knew the correct answers. In a control group, where they made judgments in isolation, they made almost no errors. Yet Asch’s par- ticipants went along with the incorrect ma- jority 37% of the time—far more often than most of us would ever predict. Not everyone conformed, of course. About 25% refused to agree on any of the incorrect group judg- ments. Yet 50% went along on at least half of the critical presentations, and remaining participants conformed on an occasional basis. Similarly high levels of conformity were observed when Asch’s study was repeated years later and in recent studies involving other cognitive tasks. For example, recent research demonstrates strong conformity ef- fects on memory—as when an eyewitness is influenced by the report of a co-witness

(Gabbert et al., 2003; Horry et al., 2012). Asch-like conformity effects are also found in the perceptual judgments of 3- and 4-year-old children (Corriveau et al., 2009; Corriveau & Harris, 2010).

Let’s compare Sherif’s and Asch’s classic studies of social influence. Obvi- ously, both demonstrate that our visual perceptions can be heavily influenced by others. But how similar are they, really? Did Sherif’s and Asch’s participants exhibit the same kind of conformity and for the same reasons or was the resem- blance in their behavior more apparent than real?

From the start, it was clear that these studies differed in some important ways. In Sherif’s situation, participants were quite literally “in the dark,” so they natu- rally turned to others for guidance. When physical reality is ambiguous and we are uncertain of our own judgments, as in the autokinetic situation, others can serve

as a valuable source of information (Festinger, 1954). In contrast, Asch’s participants found themselves in a much more awkward position. Their task was relatively simple, and they could see with their own eyes which answers were correct. Still, they often followed the incorrect ma- jority. In interviews, many of Asch’s participants reported afterward that they went along with the group even though they were not convinced that the group was right. Many who did not conform said they felt “conspicuous” and “crazy,” like a “misfit” (Asch, 1956, p. 31).

Worldwide, 3.05 billion people, accounting for over 42% of Earth’s population, have access to the Internet (Internet World Stats, 2015). This being the case, you may wonder: Do the social forces that influence people in the face-to-face encounters studied by Sherif and Asch also op- erate in virtual groups whose other members are nameless, faceless, and anonymous? The answer is yes. McKenna and Bargh (1998) observed behavior in various online blogs in which people with common interests posted and responded to messages on a whole range of topics, from obesity and sexual orientation to money and the stock market. The social nature of the medium in this virtual

After two uneventful rounds in Asch’s study, the participant (seated second from the right) faces a dilemma. The answer he wants to give in the third test of visual discrimination differs from that of the first five confederates, who are all in agreement. Should he give his own answers or conform to theirs?

A rc

hi ve

s of

th e

H is

to ry

o f A

m er

ic an

P sy

ch ol

og y/

U ni

ve rs

ity o

f A kr

d F i g u r e 7 . 4

Line Judgment Task used in Asch’s Conformity Studies

Which comparison line—A, B, or C—is the same in length as the standard line? What would you say if you found yourself in the presence of a unanimous majority that answered A or C? The participants in Asch’s experiments conformed to the majority about a third of the time. Asch, 1955

Standard Line Comparison Lines A B C

80220_ch07_ptg01_hr_262-307.indd 270 27/01/16 11:29 AM

271 Conformity

situation was “remote.” Still, these researchers found that in newsgroups that brought together people with “hidden identities” (such as gays and lesbians who had concealed their sexuality), members were highly responsive to social feed- back. Those who posted messages that were met with approval rather than disap- proval later became more active participants of the newsgroup. When it comes to social support and rejection, even remote virtual groups have the power to shape our behavior (Bargh & McKenna, 2004; Kassner et al., 2012; Williams et al., 2000).

j Why Do People Conform? The Sherif and Asch studies demonstrate that people conform for two very differ- ent reasons: one informational, the other normative (Cialdini & Goldstein, 2004; Crutchfield, 1955; Deutsch & Gerard, 1955).

A Need to Be right Through informational influence, people conform be- cause they want to make good and accurate judgments of reality and assume that when others agree on something, they must be right. In Sherif’s autokinetic task, as in other difficult or ambiguous tasks, it’s natural to assume that four eyes are better than two. Hence, research shows that eyewitnesses trying to recall a crime or some other event will alter their recollections and even create false memories in response to what they hear other witnesses report (Gabbert et al., 2003).

When people are in a state of uncertainty, following the collective wisdom of others may prove to be an effective strategy. In the popular TV game show Who Wants to Be a Millionaire? contestants who were stumped on a question were able to invoke one of two human forms of assistance: (1) calling a friend or relative who served as a designated “expert”; or (2) polling the studio audience, which casted votes by computer for instant feedback. Overall, the “experts” were use- ful, as they offered the correct answer 65% of the time. Illustrating the wisdom of crowds, however, the studio audiences were even more useful, picking the right answer 91% of the time (Surowiecki, 2005). These days, relying on the collective wisdom of large numbers of other people is something we do all the time. When you look to buy merchandise on Amazon.com, you will see average customer satisfaction ratings from 1 to 5 stars and product reviews; similar ratings can be found on Zagat, Yelp, and OpenTable for restaurants; Expedia.com and Trip- Advisor for hotels; Rotten Tomatoes for movies, and, well you get the idea.

A Fear of Ostracism In contrast to the informational value of conformity, normative influence leads people to conform because they fear the consequence of rejection that follows deviance. It’s easy to see why. Early on, research showed that individuals who stray from a group’s norm tend to be disliked, rejected, ridi- culed, and outright dismissed (Schachter, 1951). Although some people are more re- silient than others, these forms of interpersonal rejection can be hard to take (Smart Richman & Leary, 2009).

In a series of controlled experiments, people who were socially ostracized—for example, by being neglected, ignored, and excluded in a live or online chatroom con- versation—react with various types of emotional distress, feeling alone, hurt, angry, and lacking in self-esteem (Williams et al., 2002; Gerber & Wheeler, 2009). Even be- ing left out of a three-way text-messaging conversation on a cell phone can have this effect on us (Smith & Williams, 2004). Kipling Williams and Steve Nida (2011) note that the research on this point is clear: Some people become so distressed when they are ignored or excluded from a group, even one that is newly and briefly formed, that they begin to feel numb, sad, angry, or some combination of these emotions. Over

When all members of a group give an incorrect response to an easy question, most people most of the time conform to that response.

FALSe

informational influence Influence that produces conformity when a person believes others are correct in their judgments.

normative influence Influence that produces conformity when a person fears the negative social consequences of appearing deviant.

80220_ch07_ptg01_hr_262-307.indd 271 27/01/16 11:29 AM

272 Chapter 7 Conformity

time, ostracism becomes a form of social death, making it difficult to cope.

Why does being ostracized hurt so much? Why, for example, have some teenage victims of cyber bullying, re- acted with such devastation that they committed suicide? Increasingly, social psychologists are coming to appreciate the extent to which human beings, over the course of evo- lution, have needed each other in order to survive and to flourish. According to Geoff MacDonald and Mark Leary (2005), our need to belong is so primitive that rejection can inflict a social pain that feels just like physical pain. You can sense the connection in the way people describe their emotional reactions to social loss using such words as “hurt,” “brokenhearted,” and “crushed.” Social neuro- science research lends provocative support to this linkage. In brain-imaging studies, for example, young people who were left out by other players in a three-person Internet ball-tossing game called “Cyberball” exhibited elevated neural activity in a part of the brain that is normally as- sociated with physical pain (Eisenberger et al., 2003). In some instances, people who are excluded report feeling a heightened sensitivity to pain; in other cases, the experi- ence leads them to feel numb (Bernstein & Claypool, 2012).

If there is a silver lining, it is that social pain can have positive motivating ef- fects. Once feeling rejected, people seek to re-affiliate with others, which should increase their sensitivity to social perception cues that signal opportunities for in- clusion. In a study that tested this hypothesis, Michael Bernstein and others (2008) found that research participants who were led to feel socially rejected or excluded became more accurate in their ability to distinguish between true smiles, which betray happiness and an openness to interaction, and “masking” smiles, which do not express a genuine emotion. Even young children are highly sensitive to cues that signal ostracism. In one study, 4- and 5-year-old children were shown a cartoon video in which a character repeatedly approached a group at play and was either ac- cepted into the group or excluded. Shortly afterward, all the children were asked to “draw a picture of you and your friend.” Interestingly, those who saw the social re- jection version of the cartoon drew pictures that placed themselves and their friend closer together and that adults rated as more affiliative (Song et al., 2015).

On final point about being ostracized is important: The effect depends on the source of exclusion and the cultural context. In a fascinating series of studies designed to address this question, Ayse Uskul and Harriet Over (2014) compared farmers and herders in the Eastern Black Sea region of Turkey—two groups that shared the same geographical space, national identity, language, and religion. The farmers, who grow tea and other products, show a high level of interdependence within their social network, having to rely heavily on family and neighbors. In contrast, the herders, who sell cattle and dairy products, tend to be more individu- alistic and independent, moving between neighboring towns and interacting with strangers and others outside their immediate social circle. In one study, partici- pants were asked to visualize themselves as the character in a vignette who was socially excluded either by a close other person or by a stranger. They were then asked to rate in various ways how they would feel about themselves. Two impor- tant results consistently emerged. First, both groups were more distressed about being excluded by close others than by strangers. Second, the herders felt worse

Em ily

F la

ke /C

on de

N as

t C ol

le ct

io n

A good deal of conformity is driven by a fear of public disapproval.

80220_ch07_ptg01_hr_262-307.indd 272 27/01/16 11:29 AM

273 Conformity

private conformity The change of beliefs that occurs when a person privately accepts the position taken by others.

public conformity A superficial change in overt behavior without a corresponding change of opinion that is produced by real or imagined group pressure.

about being excluded by strangers than the farmers did. As a cultural subgroup that relies on strangers to make a living, herders did not make the same ingroup- outgroup distinction that the farmers did. In this domain of social influence, as in others, cultural context plays an important role.

Distinguishing Types of Conformity In group settings, both informational and normative influences are typically at work. Consider the Asch experiment. Even though many of his participants said they had conformed just to avoid being differ- ent, others said that they came to agree with their group’s erroneous judgments. Is that possible? At the time, Asch had to rely on what his participants reported in inter- views. Thanks to recent developments in social neuroscience, however, researchers can now peer into the socially active brain. In an ingenious medical school study that illustrates the point, Gregory Berns and others (2005) put 32 adults into a visual-spatial per- ception experiment in which they were asked to “mentally rotate” two geometric objects to determine if they were the same or different (see d Figure 7.5). As in the original Asch study, participants were accompanied by four con- federates who unanimously made incorrect judgments on certain tri- als. Unlike in the original study, however, participants were placed in an fMRI scanner while engaged in the task. There were two note- worthy results. First, participants conformed to 41% of the group’s incorrect judgments. Second, these conforming judgments were ac- companied by heightened activity in a part of the brain that controls spatial awareness—not in areas as- sociated with conscious decision making. These results suggest that the group altered perceptions, not just behavior.

The distinction between the two types of social influence— informational and normative—is important, not just for under- standing why people conform but because the two sources of influence produce different types of conformity: private and public (Allen, 1965; Kelman, 1961). Like beauty, conformity may be skin deep or it may penetrate beneath the surface. Private conformity, also called true acceptance or conversion, describes instances in which others cause us to change not only our overt behavior but our minds as well. To conform at this level is to be truly persuaded that others in a group are correct. In contrast, public conformity (sometimes called compliance, a term used later in this chapter to describe a different form of influence) refers to a more superficial change in behavior. People often respond to normative pressures by

d F i g u r e 7 . 5

Conformity effects on Perception

In this study, participants tried to determine if pairs of geometric objects were the same or different after observing the responses of four unanimous confederates. Participants followed the incorrect group 41% of the time. Suggesting that the group had altered perceptions, not just behavior, fMRI results showed that these conforming judgments were accompanied by increased activity in a part of the brain that controls spatial awareness. Berns et al., 2005

Subject

Group

Same

80220_ch07_ptg01_hr_262-307.indd 273 27/01/16 11:29 AM

274 Chapter 7 Conformity

pretending to agree even when privately they do not. This often happens when we want to curry favor with others. The politician who tells voters whatever they want to hear is a case in point.

How, you might be wondering, can social psychologists ever tell the dif- ference between the private and public conformist? After all, both exhibit the same change in their observable behavior. The difference is that compared with someone who merely acquiesces in public, the individual who is truly persuaded maintains that change long after the group is out of the picture. When this dis- tinction is applied to Sherif’s and Asch’s research, the results come out as ex- pected. At the end of his study, Sherif (1936) retested participants alone and found that their estimates continued to reflect the norm previously established in their group—even among those who were retested a full year after the experi- ment (Rohrer et al., 1954). Similar results were recently reported in a study in which college students rated the attractiveness of various faces, were shown the average higher or lower results from other students, and then re-rated the same faces—one, three, or seven days or three months later. In this situation, the con- formity effect lasted three to seven days (Huang et al., 2014). Yet in contrast to these results, Asch (1956) himself had reported that when he had participants

write their answers privately, so that others in the group could not see, their level of conformity dropped sharply (Deutsch & Gerard, 1955; Mouton et al., 1956).

In a study that showed both processes at work, Robert S. Baron and others (1996) had people in groups of three (one participant, two confederates) act as eyewitnesses: First, they would see a picture of a person, then they would try to pick that person out of a lineup. In some groups, the task was difficult, like Sherif’s, since participants saw each picture only once for half a second. For other groups, the task was easier, like Asch’s, in that they saw each picture twice for a total of 10 seconds. How often did participants conform when the confederates made the wrong identification? It depended on how motivated they were. When the experimenter downplayed the task as only a “pilot study,” the conformity rates were 35% when the task was dif- ficult and 33% when it was easy. But when participants were of- fered a financial incentive to do well, conformity went up to 51% when the task was difficult and down to 16% when it was easy (see d Figure 7.6). With pride and money on the line, the Sherif- like participants conformed more and the Asch-like participants conformed less.

Table 7.1 summarizes the comparison of Sherif’s and Asch’s studies and the depths of social influence that they demonstrate. Looking at this table, you can see that the difficulty of the task is crucial. When reality cannot easily be validated by physical evidence, as in the autokinetic situation, people turn to others for information and conform because they are truly persuaded by that information. When reality is clear, however, the cost of dissent becomes the major issue. As Asch found, it can be difficult to depart too much from others even when you know that they—not you—are wrong. So you play along. Privately you don’t change your mind. But you nod your head in agreement anyway.

d F i g u r e 7 . 6

Distinguishing Types of Conformity

People made judgments under conditions in which they had a high or low level of motivation. Regardless of whether the judgment task was difficult or easy, there were moderate levels of conformity when participants had low motivation (left). But when they were highly motivated (right), participants conformed more when the task was difficult (as in Sherif’s study) and less when it was easy (as in Asch’s study). From Baron, R. et al., (1996). Journal of Personality and Social Psychology vol 71 (pp. 915–927).

Pe rc

en ta

ge o

f c on

fo rm

ity r

es po

ns es

0 Low motivation High motivation

Dif cult (Sherif-like) task Easy (Asch-like) task

80220_ch07_ptg01_hr_262-307.indd 274 27/01/16 11:29 AM

275 Conformity

m T a b l e 7 . 1

Two Types of Conformity

A comparison of Sherif’s and Asch’s studies suggests different kinds of conformity for different reasons. Sherif used an ambiguous task, so others provided a source of information and influenced the participants’ true opinions. Asch used a task that required simple judgments of a clear stimulus, so most participants exhibited occasional public conformity in response to normative pressure but privately did not accept the group’s judgments.

experimental Task Primary effect of group Depth of Conformity Produced

Sherif’s ambiguous autokinetic effect

Informational influence Private acceptance

Asch’s simple-line judgments Normative influence Public conformity

j Majority influence Realizing that people often succumb to pressure from peers is only a first step in understanding the process of social influence. The next step is to identify the situ- ational and personal factors that make us more or less likely to conform. We know that people tend to conform when the social pressure is intense and that they are insecure about how to behave. But what creates these feelings of pressure and insecurity? Here, we look at four factors: the size of the group, a focus on norms, the presence of an ally, and gender.

group Size: The Power in Numbers Common sense would suggest that as the number of other people in a majority increases, so should their impact. Actually, it is not that simple. Asch (1956) varied the size of groups, using 1, 2, 3, 4, 8, or 15 confederates, and he found that conformity increased with group size—but only up to a point. Once there were 3 or 4 confederates, the amount of additional influence exerted by the rest was negligible. Other researchers have obtained similar results (Gerard et al., 1968).

Beyond the presence of three or four others, additions to a group are subject to the law of “diminishing returns” (Knowles, 1983; Mullen, 1983). As we will see later, Bibb Latané (1981) likens the influence of people on an individual to the way lightbulbs illuminate a surface. When a second bulb is added to a room, the effect is dramatic. When the tenth bulb is added, however, its impact is barely felt, if at all. Economists say the same about the perception of money. An additional dollar seems greater to the person who has only three dollars than it does to the person who has 300.

Another possible explanation is that as more and more people express the same opinion, an individual is likely to suspect that they are acting either in “collusion” or as “spineless sheep.” According to David Wilder (1977), what matters is not the actual number of others in a group but one’s perception of how many distinct others who are thinking independently the group includes. Indeed, Wilder found that people were more influenced by two groups of two than by one four-person group and by two groups of three than by one six-per- son group. Conformity increased even more when people were exposed to three two-person groups. When faced with a majority opinion, we do more than just

80220_ch07_ptg01_hr_262-307.indd 275 27/01/16 11:29 AM

276 Chapter 7 Conformity

count the number of warm bodies—we try to assess the number of independent minds.

A Focus on Norms The size of a majority may influence the amount of pressure that is felt, but social norms give rise to conformity only when we know the norms and focus on them. This may sound like an obvious point, yet we often misperceive what is normative, particularly when others are too afraid or too embarrassed to publicly present their true thoughts, feelings, and behaviors.

One common example of this “pluralistic ignorance” concerns perceptions of alcohol usage. In a number of college-wide surveys, Deborah Prentice and Dale Miller (1996) found that most students overestimated how comfortable their peers were with the level of drinking on campus. Those who most over- estimated how others felt about drinking at the start of the school year even- tually conformed to this misperception in their own attitudes and behavior. In contrast, students who took part in discussion sessions that were designed to correct these misperceptions actually consumed less alcohol six months later. These findings are important. Additional research has shown that both male and female students tend to overestimate how frequently their same-sex peers use various substances and the quantities they consume. Whether the substance in question is alcohol, tobacco, or marijuana, the more normative students perceive peer usage to be, the more they consume (Henry et al., 2011). More generally, across a range of situations and intervention purposes, chang- ing people’s perceptions of norms can be used to change their behavior (Miller & Prentice, 2016).

Knowing how others are behaving in a situation is necessary for conformity, but these norms will influence us only when they are brought to our awareness, or “activated.” Robert Cialdini (2003) and his colleagues have demonstrated this point in studies on littering. In one study, researchers had confederates pass out handbills to amusement park visitors and varied the amount of litter that ap- peared in one section of the park (an indication of how others behave in that set- ting). The result: The more litter there was, the more likely visitors were to toss their handbills to the ground (Cialdini et al., 1990).

A second study showed that passersby were most influenced by the prior behavior of others when their attention was drawn to the existing norm. In this instance, people were observed in a parking garage that was either clean or cluttered with cigarette butts, candy wrappers, paper cups, and trash. In half of the cases, the norm that was already in place—clean or cluttered—was brought to participants’ attention by a confederate who threw paper to the ground as he walked by. In the other half, the confederate passed by without incident. As participants reached their cars, they found a “Please Drive Safely” handbill tucked under the windshield wiper. Did they toss the paper to the ground or take it with them? The results showed that people were most likely to conform (by littering more when the garage was cluttered than when it was clean) when the confederate had littered—an act that drew attention to the norm (Cialdini et al., 1991).

An Ally in Dissent: getting by With a Little Help In Asch’s initial experi- ment, unwitting participants found themselves pitted against unanimous majori- ties. But what if they had an ally, a partner in dissent? Asch investigated this issue and found that the presence of a single confederate who agreed with the partici- pant reduced conformity by almost 80%. This finding, however, does not tell us

80220_ch07_ptg01_hr_262-307.indd 276 27/01/16 11:29 AM

277 Conformity

why the presence of an ally was so effective. Was it because he or she agreed with the participant or because he or she disagreed with the majority? In other words, were the views of the participants strengthened because a dissenting confeder- ate offered validating information or because dissent per se reduced normative pressures?

A series of experiments explored these two possibilities. In one, Vernon Allen and John Levine (1969) led participants to believe that they were work- ing together with four confederates. Three of these others consistently agreed on the wrong judgment. The fourth then followed the majority, agreed with the participant, or made a third judgment, which was also incorrect. This last variation was the most interesting: Even when the confederate did not validate their own judgment, participants conformed less often to the majority. In an- other experiment, Allen and Levine (1971) varied the competence of the ally. Some participants received support from an average person. In contrast, oth- ers found themselves supported by someone who wore very thick glasses and complained that he could not see the visual displays. Not a very reassuring ally, right? Wrong. Even though participants derived less comfort from this sup- porter than from one who seemed more competent at the task, his presence still reduced their level of conformity.

Two important conclusions follow from this research. First, it is substantially more difficult for people to stand alone for their convictions than to be part of even a tiny minority. Second, any dissent—whether it validates an individual’s opinion or not—can break the spell cast by a unanimous majority and reduce the normative pressures to conform. In an interesting possible illustration of how uncommon it is for individuals to single-handedly oppose a majority, researchers examined voting patterns on the U.S. Supreme Court from 1953 to 2001. Table 7.2 shows that out of 4,178 decisions in which all nine justices voted, the 8-to-1 split was the least frequent, occurring in only 10% of all decisions (Granberg & Bartels, 2005).

gender Differences Are there gender differences in con- formity? Based on Asch’s initial studies, social psychologists used to think that women, once considered the “weaker” sex, conform more than men. In light of all the research, however, it appears that two additional factors have to be considered. First, sex differences depend on how comfortable people are with the experimental task. In a classic study, Frank Sistrunk and John McDavid (1971) had male and female participants answer ques- tions on stereotypically masculine, feminine, and gender-neutral topics. Along with each question, participants were told the per- centage of others who agreed or disagreed. Although females conformed to the contrived majority more on masculine items, males conformed more on feminine items. There were no sex differences on the neutral questions. This finding suggests that one’s familiarity with the issue at hand, not gender, is what affects conformity (Eagly & Carli, 1981).

A second factor is the type of social situation people face. As a general rule, gender differences are weak and unreli- able. But there is an important exception: In face-to-face en- counters, where people must disagree with each other openly, small differences do emerge. In fact, when participants think

Granberg & Bartels, 2005

m T a b l e 7 . 2

on being a lone Dissenter: Voting Patterns on the U.S. Supreme Court

Out of 4,178 U.S. Supreme Court decisions handed down from 1953 to 2001, the 8-to-1 breakdown involving a lone dissenter was the least common type of vote. This historical observation is consistent with conformity research showing the power of the majority over an individual who lacks an ally

Vote Breakdowns Frequency

9 to 0 35%

8 to 1 10%

7 to 2 14%

6 to 3 20%

5 to 4 21%

80220_ch07_ptg01_hr_262-307.indd 277 27/01/16 11:29 AM

278 Chapter 7 Conformity

they are being observed, women conform more and men conform less than they do in a more private situation. Why does being “in public” create such a diver- gence in behavior? Alice Eagly (1987) argues that in front of others, people worry about how they come across and feel pressured to behave in ways that are viewed as acceptable according to traditional gender-role constraints. At least in public, men feel pressured to behave with fierce independence and autonomy, whereas women are expected to play a gentler, more harmonious role.

From an evolutionary perspective, Vladas Griskevicius and others (2006) sug- gest that people are most likely to behave in gender-stereotyped ways when mo- tivated to attract someone of the opposite sex. Consistent with the stereotypic notion that women tend to like men who distinguish themselves as independent and dominant, whereas men prefer women who are agreeable and cooperative, their research shows that women conform more—and that men conform less— when primed to think about themselves in a romantic situation. But here’s the interesting part: When Matthew Hornsey and others (2015) asked people in a series of experiments to indicate their personal preferences for a romantic partner, both men and women alike were attracted to others who are nonconformists. This pairing of results begs the question: Why did female participants in the first study present themselves as conformist to attract a male partner when men in the sec- ond set of studies consistently expressed a preference for nonconformist women? Although more research is needed, these studies seem to suggest that women

harbor the belief that men like conformist women—a belief, grounded in the past, that may well turn out to be a myth.

This image, from the 2014 documentary CitizenFour, shows ex-CIA employee Edward Snowden, on the days when he leaked classified information from the U.S. National Security Agency (NSA). The information disclosed a massive surveillance program, which spied on U.S. citizens without a warrant. Although his disclosures proved accurate and sparked reform, the U.S. government charged Snowden with the crime of espionage, causing him to flee. He was granted temporary asylum in Russia, where he now lives. Engaging in one of the most profound acts of dissent in modern history, Snowden has been variously called a hero, a whistleblower, a patriot, and a traitor.

Th e

G ua

rd ia

n/ H

an do

ut /G

et ty

Im ag

es N

ew s/

G et

ty Im

ag es

j Minority influence In a book entitled Dissent in Dangerous Times, Austin Sarat (2005) noted that while the freedom to dissent is highly valued in the American national psyche, in- dividual dissenters are often vilified for their beliefs— especially in today’s post-9/11 war on terrorism.

The fact is, it has never been easy for individu- als to express unpopular views and enlist support

for these views from others. Philosopher Bertrand Russell once said, “Con- ventional people are roused to frenzy by departure from convention, largely because they regard such departure as criticism of themselves.” He may have been right. Although people who assert their beliefs against the majority are generally seen as competent and honest, they are also disliked and roundly rejected (Bassili & Provencal, 1988; Levine, 1989). It’s no wonder that most people think twice before expressing unpopular positions. In a series of survey studies of what he called the “minority slowness effect,” John Bassili (2003) asked people about their attitudes on social policy issues such as affirmative action or about their likes and dislikes for various celebrities, sports, foods, places, and activities. Consistently, and regardless of the topic, respondents who held minority opinions were slower to answer the questions than those in the majority.

Resisting the pressure to conform and maintaining one’s independence may be socially difficult, but it is not impossible. History’s most famous heroes, vil- lains, and creative minds are living proof: Joan of Arc, Jesus Christ, Galileo, Charles Darwin, Mahatma Gandhi, and Nelson Mandela, to name just a few,

80220_ch07_ptg01_hr_262-307.indd 278 27/01/16 11:29 AM

279 Conformity

minority influence The process by which dissenters produce change within a group.

were dissenters of their time who continue to capture the imagination. Then there is human behavior in the laboratory. Social psychologists have been so in- trigued by Asch’s initial finding that participants conformed 37% of the time that textbooks such as this one routinely refer to “Asch’s conformity study.” Yet the overlooked flip side of the coin is that Asch’s participants refused to acquiesce 63% of the time—thus also indicating the power of independence, truth telling, and a concern for social harmony (Friend et al., 1990; Hodges & Geyer, 2006; Jetten & Hornsey, 2011).

Twelve Angry Men, a classic film starring Henry Fonda, illustrates how a lone dissenter can resist the pressure to conform and convince others to follow. Al- most as soon as the door of the jury room closes, the jury in this film takes a show-of-hands vote. The result is an 11-to-1 majority in favor of conviction with Fonda the lone holdout. Through 90 minutes of heated deliberation, Fonda works relentlessly to plant a seed of doubt in the minds of his peers. In the end, the jury reaches a unanimous verdict: not guilty.

Sometimes art imitates life; sometimes it does not. In this instance, Henry Fonda’s heroics are highly atypical. When it comes to jury decision making, as we’ll see in Chapter 12, the majority usually wins. Yet in trial juries, as in other small groups, there are occasional exceptions where minorities prevail—as when someone from the majority faction defects. Thanks to Serge Moscovici and others, we now know quite a bit about minority influence and the strategies that astute nonconformists use to act as agents of social change (Gardikiotis, 2011; Maass & Clark, 1984; Moscovici et al., 1985; Mugny & Perez, 1991).

Moscovici’s Theory According to Moscovici, majorities are powerful by virtue of their sheer numbers, whereas nonconformists derive power from the style of their behavior. It is not just what nonconformists say that matters but how they say it. To exert influence, says Moscovici, those in the minority must be forceful, persistent, and unwavering in support of their position. Yet at the same time, they must appear flexible and open-minded. Confronted with a consistent but even- handed dissenter, members of the majority will sit up, take notice, and rethink their own positions.

Why should a consistent behavioral style prove effective? One possible reason is that unwavering repetition draws attention from those in the mainstream, which is a necessary first step to social influence. Another possibility is that consistency signals that the dissenter is unlikely to yield, which leads those in the majority to feel pressured to seek compromise. A third possible reason is that when con- fronted with someone who has the self-confidence and dedication to take an un- popular stand without backing down, people assume that he or she must have a point. Of course, it helps to be seen as part of “us” rather than “them.” Research shows that dissenters have more influence when people identify with them and perceive them to be similar in ways that are relevant and desirable (Turner, 1991; Wood et al., 1996).

Based on a meta-analysis of 97 experiments investigating minority influence, Wendy Wood and her colleagues (1994) concluded that there is strong support for the consistency hypothesis. In one classic study, for example, Moscovici and others (1969) turned Asch’s procedure on its head by confronting people with a minor- ity of confederates who made incorrect judgments. In groups of six, participants took part in what was supposed to be a study of color perception. They viewed a series of slides that all were blue but varied in intensity. For each slide, the partici- pants took turns naming the color. The task was simple, but two confederates an- nounced that the slides were green. When the confederates were consistent—that

80220_ch07_ptg01_hr_262-307.indd 279 27/01/16 11:29 AM

280 Chapter 7 Conformity

idiosyncrasy credits Interpersonal “credits” that a person earns by following group norms.

is, when both made incorrect green judgments for all slides—they had a surprising degree of influence. About a third of all participants incorrectly reported seeing at least one green slide, and 8% of all responses were incorrect. Subsequent research confirmed that the perception of consistency increases minority influence (Clark, 2001; Crano, 2000).

Noting that social dissent can also breed hostility, Edwin Hollander (1958) recommended a different approach. Hollander warned that people who seek positions of leadership or challenge a group without first becoming accepted full- fledged members of that group run the risk that their opinions will fall on deaf ears. As an alternative to Moscovici’s consistency strategy, Hollander suggested that to influence a majority, people should first conform in order to establish their credentials as competent insiders. By becoming members of the mainstream, they accumulate idiosyncrasy credits, or “brownie points.” Once they have accumu- lated enough goodwill within the group, a certain amount of their deviance will then be tolerated. Several studies have shown that this “first conform, then dis- sent” strategy, like the “consistent dissent” approach, can be effective (Bray et al., 1982; Lortie-Lussier, 1987; Rijnbout & McKimmie, 2012).

Processes and Outcomes of Minority influence Regardless of which strat- egy is used, minority influence is a force to be reckoned with. But does it work just like conformity, or is there something different about the way minorities and majori- ties effect change? Some theorists have proposed that a single process accounts for both directions of social influence—that minority influence is like a “chip off the old block” (Latané & Wolf, 1981; Tanford & Penrod, 1984). Others have taken a dual- process approach (Moscovici, 1980; Nemeth, 1986). In this second view, majorities and minorities exert influence in different ways and for different reasons. Majorities, because they have power and control, elicit public conformity by bringing stressful normative pressures to bear on the individual. But minorities, because they are seen as seriously committed to their views, produce a deeper and more lasting form of private conformity, or conversion, by leading others to become curious and rethink their original positions.

To evaluate these single- and dual-process theories, researchers have com- pared the effects of majority and minority viewpoints on participants who are otherwise neutral on an issue in dispute. On the basis of this research, two con- clusions can be drawn. First, the relative impact of majorities and minorities depends on whether the judgment that is being made is objective or subjective, a matter of fact or opinion. In a study conducted in Italy, Anne Maass and oth- ers (1996) found that majorities have greater influence on factual questions, for which only one answer is correct (“What percentage of its raw oil does Italy import from Venezuela?”), but that minorities exert equal impact on opinion questions, for which there is a range of acceptable responses (“What percent- age of its raw oil should Italy import from Venezuela?”). People feel freer to stray from the mainstream on matters of opinion, when there is no right or wrong answer.

The second conclusion is that the relative effects of majority and minor- ity points of view depend on how and when conformity is measured. To be sure, majorities have a decisive upper hand on direct or public measures of conformity. After all, people are reluctant to oppose a group norm in a con- spicuous manner. But on more indirect or private measures of conformity, on attitude issues that are related but not focal to the point of conflict, or after the passage of time—all of which softens the extent to which majority

80220_ch07_ptg01_hr_262-307.indd 280 27/01/16 11:29 AM

281 Conformity

participants would appear deviant—minorities exert a strong impact (Clark & Maass, 1990; Crano & Seyranian, 2009; Moscovici & Personnaz, 1991). As Moscovici cogently argued, each of us is changed in a meaningful but subtle way by minority opinion.

In Rogues, Rebels and Dissent: Just Because Everyone Agrees, Doesn’t Mean They’re Right, Charlan Nemeth (2016) reviews more than 30 years of research indicating that dissenters in a group serve an invaluable purpose—regardless of whether their views are correct. Dissenters may feel like a nuisance, she notes, and an unnecessary source of conflict, forcing others to defend a position that everyone else agrees with. Yet research shows that dissent sparks innovation (De Dreu & De Vries, 2001).

Simply by their willingness to stay firmly independent, minorities can force other group members to think more carefully, more openly, in new and different ways, and more creatively about a problem, enhancing the quality of a group’s output. In one study, participants exposed to a minority viewpoint on how to solve anagram problems later found more novel solutions themselves (Nemeth & Kwan, 1987). In a second study, those exposed to a consistent minority viewpoint on how best to recall information later recalled more words from a list they were trying to memorize (Nemeth et al., 1990). In a third study, interacting groups with one dissenting confederate produced more original analyses of complex business problems (Van Dyne & Saavedra, 1996). Importantly, Nemeth and her colleagues (2001) found that to have influence over a group, lone individuals must exhibit “authentic dissent,” not merely play “devil’s advocate,” a tactic that actually bol- sters a majority’s position.

j Culture and Conformity We humans are a heterogeneous and diverse lot. As a matter of geography, some of us live in large, heavily populated cities whereas others live in small towns, af- fluent suburbs, rural farming or fishing communities, jungles, expansive deserts, high-altitude mountains, tropical islands, and vast arctic plains. Excluding dia- lects, more than 6,500 different languages are spoken. There are also hundreds of religions that people identify with—the most common being Christianity (32%), Islam (23%), Hinduism (14%), and Buddhism (8%), with Judaism (0.20%) and others claiming fewer adherents. Roughly 15% of the world’s population is not affiliated with a religion (Adherents.com, 2015).

Linked together by historical time and geographical space, each culture has its own ideology, music, fashions, foods, laws, customs, and manners of expression. As many tourists and exchange students traveling abroad have come to learn, sometimes the hard way, the social norms that influence human conduct can vary in significant ways from one part of the world to another.

In Do’s and Taboos Around the World, R. E. Axtell (1993) warns world travelers about some of these differences. Dine in an Indian home, he notes, and you should leave food on the plate to show the host that the portions were generous and you had enough to eat. Yet as a dinner guest in Bolivia, you would show your appreciation by cleaning your plate. Shop in an outdoor market in Iraq, and you should expect to negotiate the price of everything you buy. Plan an appointment in Brazil, and the person you’re scheduled to meet is likely to be late; it’s nothing personal. In North America, it is common to sit casually opposite someone with your legs outstretched. Yet in Nepal, as in many Muslim countries, it is an insult to point the bottoms of your feet at

80220_ch07_ptg01_hr_262-307.indd 281 27/01/16 11:29 AM

282 Chapter 7 Conformity

someone. Even the way we space ourselves from each other is influenced by culture. Americans, Canadians, British, and northern Europeans keep a polite distance between themselves and others and feel “crowded” by the touchier, nose-to-nose style of the French, Greeks, Arabs, Mexicans, and people of South America. In the affairs of day-to- day living, each culture operates by its own rules of conduct.

Just as cultures differ in their social norms, so too they differ in the extent to which people are expected to adhere to those norms. As we saw in Chapter 3, there are different cultural orientations toward per-

sons and their relationships to groups. Some cultures primarily value individual- ism and the virtues of independence, autonomy, and self-reliance, whereas others value collectivism and the virtues of interdependence, cooperation, and social harmony. Under a banner of individualism, personal goals take priority over group allegiances. Yet in collectivistic cultures, the person is first and foremost a loyal member of a family, city, team, company, church, and state.

What determines whether a culture becomes individualistic or collectivistic? Speculating on the origins of these orientations, Harry Triandis (1995) suggested that there are three key factors. The first is the complexity of a society. As people come to live in more complex industrialized societies (compared, for example, with a simpler life of food gathering among desert nomads), there are more groups to identify with, which means less loyalty to any one group and a greater focus on personal rather than collective goals. Second is the affluence of a soci- ety. As people prosper, they gain financial independence from others, a condition that promotes social independence as well as mobility and a focus on personal rather than collective goals. The third factor is heterogeneity. Societies that are homogeneous or “tight” (where members share the same language, religion, and social customs) tend to be rigid and intolerant of those who veer from the norm. Societies that are culturally diverse or “loose” (where two or more cultures co- exist) tend to be more permissive of dissent, thus allowing for more individual expression. According to Edward Sampson (2000), cultural orientations may also be rooted in religious ideologies, as in the link between Christianity and individualism.

Cultures differ in their unique, often colorful norms. In Nigeria it’s common to ask guests to wear color-coordinated outfits, called aso ebi, at social events, such as this wedding in Lagos (top left). In Mumbai, India, a man smears a woman’s face with brightly colored powders in the festival of Holi, which marks the beginning of spring (top right). In Spain, revelers in Pamplona hold up their bandanas before the festival of San Fermin, during which six bulls run through the crowded streets in the center of town (bottom left).

Ro bi

n H

am m

on d/

N at

io na

l G eo

gr ap

hi c

Cr ea

tiv e

A P

Im ag

es /R

aj an

is h

Ka ka

D es

m on

d Bo

yl an

/R eu

te rs

/C or

bi s

individualism A cultural orientation in which independence, autonomy, and self-reliance take priority over group allegiances.

collectivism A cultural orientation in which interdependence, cooperation, and social harmony take priority over personal goals.

80220_ch07_ptg01_hr_262-307.indd 282 27/01/16 11:29 AM

283 Compliance

compliance Changes in behavior that are elicited by direct requests.

Early research across nations showed that autonomy and indepen- dence were most highly valued in the United States, Australia, Great Britain, Canada, and the Netherlands, in that order. In contrast, other cultures value social harmony and “fitting in” for the sake of com- munity, the most collectivist people being from Venezuela, Colombia, Pakistan, Peru, Taiwan, and China (Hofstede, 1980). Although re- search shows that the differences are even more complicated, that individuals differ even within cultures, and that cultures change over time, it is clear that nations on average vary in their orientations on the dimension of individualism (Schimmack et al., 2005). The differ- ence can be measured in books and other written materials—by the contents of the stories told (Imada, 2012) and the use of the first- person singular pronouns, “I” and “me” (Hamamura & Xu, 2015; Twenge et al., 2013).

Do cultural orientations influence conformity? Among the Bantu of Zimbabwe, an African people in which deviance is scorned, 51% of participants who were placed in an Asch-like study conformed— more than the number typically seen in the United States (Whittaker & Meade, 1967). When John Berry (1979) compared participants from 17 cultures, he found that conformity rates ranged from a low of 18% among Inuit hunters of Baffin Island to a high of 60% among village- dwelling Temne farmers of West Africa. Additional analyses have shown that conformity rates are generally higher in cultures that are collectivistic rather than individualistic in orientation (Bond & Smith, 1996). Hence, many anthropologists—interested in human culture and its influence over individuals—study the processes of conformity and independence (Spradley et al., 2015).

Compliance In conformity situations, people follow implicit or explicit group norms. But an- other common form of social influence occurs when others make direct requests of us in the hope that we will comply. Situations that call for compliance take many forms. These include a friend’s plea for help, sheepishly prefaced by the question “Can you do me a favor?” They also include the pop-up ads on the Internet de- signed to lure you into a commercial site and the salesperson’s pitch for business prefaced by the dangerous words “Have I got a deal for you!” Sometimes, the request is up front and direct; what you see is what you get. At other times, it is part of a subtle and more elaborate manipulation.

How do people get others to comply with self-serving requests? How do police interrogators get crime suspects to confess? How do political parties and chari- table organizations draw millions of dollars in contributions from voters? How do you exert influence over others? Do you use threats, promises, politeness, deceit, or reason? Do you hint, coax, sulk, negotiate, trick, throw tantrums, or pull rank whenever you can? To a large extent, the compliance strategies we use depend on how well we know the person we target, our status within a relationship, our personality, our culture, and the nature of the request.

By observing the masters of influence—advertisers, fund-raisers, politicians, and business leaders—social psychologists have learned a great deal about the

Say what you think Love who you love ‘Cause you just get So many trips ‘round the sun… Follow your arrow Wherever it points.

Country musician Kacey Musgraves performs her hit song, Follow Your Arrow. Read the lyrics above and you will see that the song expresses a cultural value of “individualism.”

W en

n Lt

d/ A

la m

80220_ch07_ptg01_hr_262-307.indd 283 27/01/16 11:29 AM

284 Chapter 7 Conformity

subtle but effective strategies that are commonly used. What they have discovered is that people often get others to comply with their requests by setting subtle psy- chological traps. Once caught in these traps, the unwary victim finds it difficult to escape. Anthony Pratkanis (2007) identified 107 methods of social influence that have been researched and published. These tactics go by various colorful names, including the lure, the 1-in-5 prize technique, the dump-and-chase technique, the disrupt-then-reframe technique, and the driving toward a goal technique. In the coming pages, some of the best known approaches will be described.

j Mindlessness and Compliance Sometimes people can be disarmed by the simple phrasing of a request, regardless of its merit. Consider, for example, requests that sound reasonable but offer no real basis for compliance. Ellen Langer and her colleagues (1978) have found that words alone can sometimes trick us into submission. In their research, an experimenter approached people who were using a library copying machine and asked to cut in. Three different versions of the request were used. In one, participants were simply asked, “Excuse me. I have five pages. May I use the Xerox machine?” In a second version, the request was justified by the added phrase “because I’m in a rush.” As you would expect, more participants stepped aside when the request was justified (94%) than when it was not (60%). A third version of the request, however, sug- gests that the actual reason offered had little to do with the increase in compliance. In this case, participants heard the following: “Excuse me. I have five pages. May I use the Xerox machine because I have to make some copies?” Huh? Read this request closely and you’ll see that it really offered no reason at all. Yet 93% in this condition complied! It was as if the appearance of a reason, triggered by the word because, was all that was needed. In fact, Langer (1989) finds that the mind is often on “automatic pilot”—we respond mindlessly to words without fully processing the information they are supposed to convey. At least for requests that are small, “sweet little nothings” may be enough to win compliance.

Langer’s research shows that sometimes we process oral requests lazily, without critical thought. In these instances, words alone, if they sound good, can be used to elicit compliance. Consider, for example, words that evoke the concept of freedom. In a series of experiments, researchers found that merely by inserting the phrase “But you are free to accept or refuse this request” into a plea, they were able to increase the numbers of people who agreed to donate money, give someone a cigarette, fill out a survey, and buy pancakes. This technique increased compliance consistently in face-to-face re- quests, on the street and in shopping malls, by mail, over the phone, and on the internet (Guéguen et al., 2013). In a meta-analysis of 42 attempts involving more than 22,000 participants, mostly in France, Christopher Carpenter (2013) concluded that the technique of evok- ing freedom through words is consistently effective.

It is interesting that although a state of mindlessness can make us vulnerable to compliance, it can also have the opposite effect. For ex- ample, many city dwellers will automatically walk past panhandlers on the street looking for a handout. Perhaps the way to increase compli- ance in such situations is to disrupt this mindless refusal response by

making a request that is so unusual that it piques the target person’s interest. To test the effect of this pique technique, researchers had a confederate approach people on the street and make a request that was either typical (“Can you spare a quarter?”) or atyp- ical (“Can you spare 17 cents?”). The result: Atypical pleas elicited more comments

Con artists prosper from the tendency for people to respond mindlessly to requests that sound reasonable but offer no real basis for compliance.

Ro be

rt M

an ko

ff /C

on de

N as

t C ol

le ct

io n

80220_ch07_ptg01_hr_262-307.indd 284 27/01/16 11:29 AM

285 Compliance

and questions from those who were targeted—and produced a 60% increase in the number of people who gave money (Santos et al., 1994). In another study, researchers who went door to door selling holiday cards gained more compliance when they dis- rupted the mindless process and reframed the sales pitch. They sold more cards when they said the price was “three hundred pennies—that’s three dollars, it’s a bargain” than when they simply asked for three dollars (Davis & Knowles, 1999).

j The Norm of reciprocity One of us recently used Lyft for the first time. Lyft is a peer-to-peer ridesharing company whose service connects drivers in the area to prospective passengers via a mobile app. Founded in 2013, in San Francisco, Lyft is a taxi service that com- petes with Uber. After arriving at the GPS-set destination and completing credit card payment for the ride, this author received from Lyft a request to rate the driver on a 5-point scale (to help ensure customer satisfaction, average driver ratings can be found on the Lyft app). That was easy. He was a 5. The car was clean and comfortable; the ride was quick; the conversation was interesting. The next day, this author was notified that the driver had rated him a 5. Not only does the com- pany seek passenger ratings of drivers but also driver ratings of passengers! The author/passenger being a social psychologist, he could not resist wondering: Had he “earned the 5 rating, or was it “payback” for his prior high rating of the driver?

A simple, unstated, but powerful rule of social behavior known as the norm of reciprocity dictates that we treat others as they have treated us (Gouldner, 1960). On the negative side, this norm can be used to sanction retaliation against those who cause us harm—as captured in the expression “an eye for an eye.” On the positive side, reciprocity can lead us to feel obligated to repay others for acts of kindness. Thus, whenever we receive gifts, invitations, compliments, and free samples, we usually go out of our way to return the favor.

The norm of reciprocity contributes to the predictability and fairness of social interaction. But it can also be used to exploit us. Dennis Regan (1971) examined this possibility in the following laboratory study. Individuals were brought together with a confederate who was trained to act in a likable or unlikable manner for an experi- ment on “aesthetics.” In one condition, the confederate did the participant an unso- licited favor. He left during a break and returned with two bottles of Coca-Cola, one for himself and the other for the participant. In a second condition, he returned from the break empty-handed. In a third condition, participants were treated to a Coke, but by the experimenter, not the confederate. The confederate then told participants in all conditions that he was selling raffle tickets at 25 cents apiece and asked if they would be willing to buy any. On average, participants bought more raffle tickets when the confederate had earlier brought them a soft drink than when he had not. The norm of reciprocity was so strong that they returned the favor even when the confederate was not otherwise a likable character. In fact, participants in this condi- tion spent an average of 43 cents on raffle tickets. At a time when soft drinks cost less than a quarter, the confederate made a handsome quick profit on his investment!

It’s clear that the norm of reciprocity can be used to trap us, unwittingly, into acts of compliance. For example, research conducted in restaurants shows that wait- ers and waitresses can increase their tip percentages by writing, “Thank you” on the back of the customer’s check, by drawing a happy face on it, or by placing candy on the check tray (Rind & Strohmetz, 2001; Strohmetz et al., 2002). But does receiving a favor make us feel indebted forever or is there a time limit to the social obligation that is so quietly unleashed? In an experiment designed to answer this question, Jerry Burger and others (1997) used Regan’s soft drink favor and had the confed- erate try to “cash in” with a request either immediately or one week later. The

80220_ch07_ptg01_hr_262-307.indd 285 27/01/16 11:29 AM

286 Chapter 7 Conformity

foot-in-the-door technique A two-step compliance technique in which an influencer sets the stage for the real request by first getting a person to comply with a much smaller request.

result: Compliance levels increased in the immediate condition but not after a full week had passed. People may feel compelled to reciprocate, but that feeling—at least for small acts of kindness—is relatively short-lived.

Some people are more likely than others to trigger and exploit the reciprocity norm. According to Martin Green- berg and David Westcott (1983), individuals who use reci- procity to elicit compliance are called “creditors” because they always try to keep others in their debt so they can cash in when necessary. On a questionnaire that measures reciprocation ideology, people are identified as creditors if they endorse such statements as “If someone does you a favor, it’s good to repay that person with a greater favor.” On the receiving end, some people try more than others not to accept favors that might later set them up to be ex- ploited. On a scale that measures reciprocation wariness, people are said to be wary if they express the suspicion, for example, that “asking for another’s help gives them power over your life” (Eisenberger et al., 1987).

Cultures may also differ in terms of their reciproca- tion wariness—with interesting consequences for social behavior. Imagine that you bump into a casual friend at the airport, stop for a drink, and the friend offers to pay for it. Would you let the friend pay? Or, suppose a sales clerk in a supermarket offered you a free sample of soup to taste. Would you accept the offer? Theorizing that the norm of reciprocity operates with particular force in col- lectivist cultures that foster interdependence, Hao Shen and others (2011) conducted a series of studies in which they posed these kinds of questions to Chinese college students from Hong Kong and to European American

students from Canada. As you can see in d Figure 7.7, the students from China were consistently less willing to accept the favor. Additional questioning revealed that these participants were more likely to see the gift giver’s motives as self-serving and to feel uncomfortably indebted by the situation.

j Setting Traps: Sequential request Strategies People who raise money or sell for a living know that it often takes more than a single plea to win over a potential donor or customer. Social psychologists share this knowl- edge and have studied several compliance techniques that are based on making two or more related requests. Click! The first request sets the trap. Snap! The second cap- tures the prey. In a classic and important book entitled Influence: Science and Practice, Robert Cialdini (2009) described a number of sequential request tactics in vivid detail. Other social psychologists have continued in this tradition (Dolinski, 2016; Kenrick et al., 2012). The best known of these methods are presented in the following pages.

The Foot in the Door Folk wisdom has it that one way to get a person to com- ply with a sizable request is to start small. First devised by traveling salespeople peddling vacuum cleaners, hairbrushes, cosmetics, magazine subscriptions, and en- cyclopedias, the trick is to somehow get your “foot in the door.” The expression need not be taken literally, of course. The point of the foot-in-the-door technique

d F i g u r e 7 . 7

People from China and Canada rated the likelihood that they would accept a free sample of soup in a supermarket (left) or a free drink from a casual friend at an airport (right). Feeling a greater burden to reciprocate, Chinese participants in both cases said they were less likely to accept the gift. Shen et al., 2011

W ill

in gn

es s

to a

cc ep

t ( 1

–7 )

Chinese Canadian

Supermarket: Soup Sample?

Airport: Free Drink?

80220_ch07_ptg01_hr_262-307.indd 286 27/01/16 11:29 AM

287 Compliance

is to break the ice with a small initial request that the customer can’t easily refuse. Once that first commitment is elicited, the chances are increased that another, larger request will succeed.

Jonathan Freedman and Scott Fraser (1966) tested the impact of this tech- nique in a series of field experiments. In one, an experimenter pretending to be employed by a consumer organization telephoned a large number of female home- makers in Palo Alto, California, and asked if they would be willing to answer some questions about the household products they use. Those who consented were then asked a few quick and innocuous questions and thanked for their assistance. Three days later, the experimenter called back and made a considerable, almost outrageous, request. He asked the women if they would allow a handful of men into their homes for two hours to rummage through their drawers and cupboards so they could take an inventory of their household products.

The foot-in-the-door technique proved to be very effective. When the par- ticipants were confronted with only the very intrusive request, 22% consented. Yet the rate of agreement among those who had been surveyed earlier more than doubled, to 53%. This basic result has now been repeated over and over again. People are more likely to donate time, money, food, blood, the use of their home, and other resources once they have been induced to go along with a small initial request. Although the effect is not always as dramatic as that obtained by Freed- man and Fraser, it does appear in a wide variety of circumstances, and it increases compliance rates, on average, by about 13% (Burger, 1999).

The practical implications of the foot-in-the-door technique are obvious. But why does it work? Over the years, several explanations have been suggested. One that seems plausible is based on self-perception theory—that people infer their at- titudes by observing their own behavior. This explanation suggests that a two-step process is at work. First, by observing your own behavior in the initial situation, you come to see yourself as the kind of person who is generally cooperative when approached with a request. Second, when confronted with the more burdensome request, you seek to respond in ways that maintain this new self-image. By this logic, the foot-in-the-door technique should succeed only when you attribute an initial act of compliance to your own personal characteristics.

Based on a review of dozens of studies, Jerry Burger (1999) concluded that the research generally supports the self-perception account. Thus, if the first request is too trivial or if participants are paid for the first act of compliance, they won’t later come to view themselves as inherently cooperative. Under these conditions, the technique does not work. Likewise, the effect occurs only when people are moti- vated to be consistent with their self-images. If participants are unhappy with what the initial behavior implies about them, if they are too young to appreciate the impli- cations, or if they don’t care about behaving in ways that are personally consistent, then again the technique does not work. Other processes may be at work, but it ap- pears that the foot opens the door by altering self-perceptions, leading people who agree to the small initial request—without any compensation—to see themselves as helpful (Burger & Caldwell, 2003). In fact, this process can still occur even when a person tries to comply with the initial small request but fails. In a series of studies, Dariusz Dolinski (2000) found that when people were asked if they could find direc- tions to a bogus street address or decipher an unreadable message—small favors that they could not satisfy—they too become more compliant with the next request.

Knowing that a foot in the door increases compliance rates is both exciting and troubling—exciting for the owner of the foot, troubling for the owner of the door. As Cialdini (2009) put it, “You can use small commitments to manipulate a person’s self-image; you can use them to turn citizens into ‘public servants,’

80220_ch07_ptg01_hr_262-307.indd 287 27/01/16 11:29 AM

288 Chapter 7 Conformity

lowballing A two-step compliance technique in which the influencer secures agreement with a request but then increases the size of that request by revealing hidden costs.

prospects into ‘customers,’ prisoners into ‘collaborators.’ And once you’ve got a man’s self-image where you want it, he should comply naturally with a whole range of your requests that are consistent with this view of himself” (p. 74).

Lowballing Another two-step trap, arguably the most unscrupulous of all com- pliance techniques, is also based on the “start small” idea. Imagine yourself in the following situation. You’re at a local car dealership. After some negotiation, the salesperson offers a great price on the car of your choice. You cast aside other con- siderations and shake hands on the deal and as the salesperson goes off to “write it up,” you begin to feel the thrill of owning a new car. Absorbed in fantasy, you are suddenly interrupted by the return of the salesperson. “I’m sorry,” he says. “The manager would not approve the sale. We have to raise the price by another $450. I’m afraid that’s the best we can do.” As the victim of an all-too-common trick known as lowballing, you are now faced with a tough decision. On the one hand, you really like the car; and the more you think about it, the better it looks. On the other hand, you don’t want to pay more than you bargained for and you have an uneasy feeling in the pit of your stomach that you’re being duped. What do you do?

Salespeople who use this tactic are betting that you’ll make the purchase despite the added cost. If the way research participants behave is any indication, they are often right. In one study, experimenters phoned introductory psychol- ogy students and asked if they would be willing to participate in a study for extra credit. Some were told up front that the session would begin at the uncivilized hour of 7 a.m. Knowing that, only 31% volunteered. But other participants were lowballed. Only after they agreed to participate did the experimenter inform them of the 7 a.m. starting time. Would that be okay? Whether or not it was, the pro- cedure achieved its objective—the signup rate rose to 56% (Cialdini et al., 1978).

Disturbing as it may be, lowballing is an interesting technique. Surely, once the lowball offer has been thrown, many recipients suspect that they were misled. Yet they go along. Why? The reason appears to hinge on the psychology of com- mitment (Kiesler, 1971). Once people make a particular decision, they justify it to themselves by thinking of all its positive aspects. As they get increasingly commit- ted to a course of action, they grow more resistant to changing their mind, even if the initial reasons for the action have been changed or withdrawn entirely. In the car dealership scenario, you might very well have decided to purchase the car because of the price. But then you would have thought about its sleek appearance, the scent of the leather interior, the iPod dock, and the brand-new satellite radio. By the time you learned that the price would be more than you’d bargained for, it would be too late—you would already have been hooked.

Lowballing also produces another form of commitment. When people do not suspect duplicity, they feel a nagging sense of unfulfilled obligation to the person with whom they negotiated. Even though the salesperson was unable to complete the original deal, you might feel obligated to buy anyway, having already agreed to make the purchase. This commitment to the other person may account for why lowballing works better when the second request is made by the same person than by someone else (Burger & Petty, 1981). It may also explain why people are most vulnerable to the lowball when they make their commitment in public rather than in private (Burger & Cornelius, 2003).

The Door in the Face Although shifting from an initial small request to a larger one can be effective, as in the foot-in-the-door and lowball techniques, oddly enough the opposite is also true. Cialdini (2009) described the time he was approached by a Boy Scout and asked to buy two five-dollar tickets to an upcoming circus. Having

80220_ch07_ptg01_hr_262-307.indd 288 27/01/16 11:29 AM

289 Compliance

door-in-the-face technique A two-step compliance technique in which an influencer prefaces the real request with one that is so large that it is rejected.

better things to do with his time and money, he declined. Then the boy asked if he would be interested in buying chocolate bars at a dollar apiece. Even though he does not particularly like chocolate, Cialdini—an expert on social influence—bought two of them. After a moment’s reflection, he realized what had happened. Whether the Boy Scout planned it that way or not, Cialdini had fallen for what is known as the door-in-the-face technique.

The technique is as simple as it sounds. An individual makes an initial request that is so large it is sure to be rejected and then comes back with a second more reasonable request. Will the second request fare better after the first one has been declined? Plagued by the sight of uneaten chocolate bars, Cialdini and others (1975) tested the effectiveness of the door-in-the-face technique. They stopped college students on campus and asked if they would volunteer to work without pay at a counseling center for juvenile delinquents. The time commitment would be forbidding: roughly two hours a week for the next two years! Not surpris- ingly, everyone who was approached politely slammed the proverbial door in the experimenter’s face. But then the experimenter followed up with a more modest proposal, asking the students if they would be willing to take a group of kids on a two-hour trip to the zoo. The strategy worked like a charm. Only 17% of the students confronted with only the second request agreed. But of those who initially declined the first request, 50% said yes to the zoo trip. Importantly, the door-in-the-face technique does not elicit mere empty promises. Most research participants who comply subsequently do what they’ve agreed to do (Cialdini & Ascani, 1976).

Why is the door-in-the-face technique such an effective trap? One possibil- ity involves the principle of perceptual contrast: To the person exposed to a very large initial request, the second request “seems smaller.” Two dollars’ worth of candy bars is not bad compared with ten dollars for circus tickets. Likewise, tak- ing a group of kids to the zoo seems trivial compared with two years of volunteer work. As intuitively sensible as this explanation seems, Cialdini and others (1975) concluded that perceptual contrast is only partly responsible for the effect. When participants only heard the large request without actually having to reject it, their rate of compliance with the second request (25%) was only slightly larger than the 17% rate of compliance exhibited by those who heard only the small request.

A more compelling explanation for the effect involves the notion of recipro- cal concessions. A close cousin of the reciprocity norm, this refers to the pressure to respond to changes in a bargaining position. When an individual backs down from a large request to a smaller one, we view that move as a concession that we should match by our own compliance. Thus, the door-in-the-face technique does not work if the second request is made by a different person (Cialdini et al., 1975). Nor does it work if the first request is so extreme that it comes across as an insincere “first offer” (Schwarzwald et al., 1979). On an emotional level, refusing to help on one request may also trigger feelings of guilt, which we can reduce by complying with the second, smaller request (O’Keefe & Figge, 1997; Millar, 2002).

That’s Not All, Folks! If the notion of reciprocal concessions is correct, then a person shouldn’t actually have to refuse the initial offer in order for the shift to a smaller request to work. Indeed, another familiar sales strategy manages to use concession without first eliciting refusal. In this strategy, a product is offered at a particular price, but then, before the buyer has a chance to respond, the seller adds, “And that’s not all!” At that point, either the original price is reduced or a bonus is offered to sweeten the pot. The seller, of course, intends all along to make the so-called concession.

An effective way to get someone to do you a favor is to make a first request that is so large the person is sure to reject it.

True

80220_ch07_ptg01_hr_262-307.indd 289 27/01/16 11:29 AM

290 Chapter 7 Conformity

that’s-not-all technique A two- step compliance technique in which the influencer begins with an inflated request, then decreases its apparent size by offering a discount or bonus.

This ploy, called the that’s-not-all technique, seems transparent, right? Surely no one falls for it, right? Burger (1986) was not so sure. He predicted that people are more likely to make a purchase when a deal seems to have improved than when the same deal is offered right from the start. To test this hypothesis, Burger set up a booth at a campus fair and sold cupcakes. Some customers who approached the table were told that the cupcakes cost 75 cents each. Others were told that they cost a dollar, but then, before they could respond, the price was reduced to 75 cents. Rationally speaking, Burger’s manipulation did not affect the ultimate price, so it should not have affected sales. But it did. When customers were led to believe that the final price represented a reduction, sales increased from 44% to 73%.

At this point, let’s step back and look at the various compliance tactics de- scribed in this section. All of them are based on a two-step process that involves a shift from a request of one size to another. What differs is whether the small or large request comes first and how the transition between steps is made (see Table 7.3). Moreover, all these strategies work in subtle ways by manipulating a target person’s self-image, commitment to the product, feelings of obligation to the seller, or perceptions of the real request. It is even possible to increase compliance by first asking “How are you feeling?” (Howard, 1990) or “I hope I’m not disturbing you, am I?” (Meineri & Guéguen, 2011), or by claiming some coincidental similarity like having the same first name or birthday (Burger et al., 2004). When you consider these various traps, you have to wonder whether it’s ever possible to escape.

j Assertiveness: When People Say No Cialdini (2009) opened his book with a confession: “I can admit it freely now. All my life I’ve been a patsy.” As a past victim of compliance traps, he is not alone.

m T a b l e 7 . 3

Sequential Request Strategies

Various compliance techniques are based on a sequence of two related requests. Click! The first request sets the trap. Snap! The second captures the prey. Research has shown that the four sequential request strategies summarized in this table are all effective.

request Shifts Technique Description

From small to large Foot-in-the-door Begin with a very small request, secure agreement, then make a separate, larger request.

Lowballing Secure agreement with a request and then increase the size of that request by revealing hidden costs.

From large to small Door-in-the-face Begin with a very large request that will be rejected; then follow that up with a more modest request.

That’s-not-all Begin with a somewhat inflated request, then immediately decrease the apparent size of that request by offering a discount or bonus.

80220_ch07_ptg01_hr_262-307.indd 290 27/01/16 11:29 AM

291 Obedience

Many people find it difficult to assert themselves in social situations. Faced with an unreasonable request from a friend, spouse, or stranger, they become anxious at the mere thought of putting a foot down and refusing to comply. There are times when it is uncomfortable for anyone to say no. However, just as we can maintain our autonomy in the face of conformity pressures, we can also refuse direct requests—even clever ones. The trap may be set, but you don’t have to get caught.

According to Cialdini, being able to resist compliance pressures rests, first and foremost, on being vigilant. If a stranger hands you a gift and then launches into a sales pitch, you should recognize the tactic for what it is and not feel indebted by the norm of reciprocity. And if you strike a deal with a salesperson who later reneges on the terms, you should be aware that you’re being lowballed and react accordingly. That is exactly what happened to one of the authors of this book. After a Saturday afternoon of careful negotiation at a local car dealer, he and his wife finally came to terms on a price. Minutes later, however, the salesman returned with the news that the manager would not approve the deal. The cost of a power moonroof, which was supposed to be included, would have to be added on. Familiar with the research, the author turned to his wife and exclaimed, “It’s a trick; they’re lowballing us!” She then became furious, went straight to the man- ager, and made such a scene in front of other customers that he backed down and honored the original deal.

What happened in this instance? Why did recognizing the attempted manipu- lation spark such anger and resistance? As this story illustrates, compliance tech- niques work smoothly only if hidden from view. The problem is, these techniques are not only attempts to influence us; they are deceptive. Flattery, gifts, and other ploys often win compliance, but not if perceived as insincere (Jones, 1964) or if the target has a high level of reciprocity wariness (Eisenberger et al., 1987). Like- wise, sequential request traps are powerful only to the extent that they are subtle and cannot be seen for what they are (Schwarzwald et al., 1979). People don’t like to be hustled. In fact, feeling manipulated typically leads us to react with anger, psychological reactance, and stubborn noncompliance—unless the request is a command and the requester is a figure of authority.

“Knowledge is power, and if you know when a clever technique is being used on you, then it becomes easier to ignore.”

—Burke Leon

Obedience Allen Funt, the creator and producer of the original TV program Candid Camera (a forerunner of the show Punk’d), spent as much time observing human behav- ior in the real world as most psychologists do. When asked what he learned from all his people watching, Funt replied, “The worst thing, and I see it over and over, is how easily people can be led by any kind of authority figure, or even the most minimal signs of authority.” He cited the time he put up a road sign that read “Delaware Closed Today.” The reaction? “Motorists didn’t question it. Instead they asked, ‘Is Jersey open?’” (Zimbardo, 1985, p. 47).

Funt was right about the way we react to authority. Taught from birth that it’s important to respect legitimate forms of leadership, people think twice before de- fying parents, teachers, employers, coaches, and government officials. The prob- lem is that mere symbols of authority—titles, uniforms, badges, or the trappings of success, even without the necessary credentials—can sometimes turn ordinary people into docile servants. Leonard Bickman (1974) demonstrated this phenom- enon in a series of studies in which a male research assistant stopped passersby

80220_ch07_ptg01_hr_262-307.indd 291 27/01/16 11:29 AM

292 Chapter 7 Conformity

obedience Behavior change produced by the commands of authority.

on the street and ordered them to do something unusual. Sometimes he pointed to a paper bag on the ground and said, “Pick up this bag for me!” At other times, he pointed to an individual standing beside a parked car and said, “This fellow is overparked at the meter but doesn’t have any change. Give him a dime!” Would anyone really take this guy seriously? When he was dressed in street clothes, only a third of the people stopped followed his orders. But when he wore a security guard’s uniform, nearly 9 out of every 10 people obeyed! Even when the uni- formed assistant turned the corner and walked away after issuing his command, the vast majority of passersby followed his orders. Clearly, uniforms signify the power of authority (Bushman, 1988).

Blind obedience may seem funny, but if people are willing to take orders from a total stranger, how far will they go when it really matters? As the pages of history attest, the implications are sobering. In World War II, Nazi officials participated in the deaths of millions of Jews, as well as of Poles, Russians, gypsies, and homo-

sexuals. Yet when tried for these crimes, all of them raised the same defense: “I was following orders.”

Surely, you may be thinking, the Holocaust was a historical anomaly that says more about the Nazis as a group of bigoted, hateful, and pathologically frustrated individuals than about the situations that lead people in general to commit acts of destructive obedience. In Hitler’s Willing Executioners, historian Daniel Goldhagen (1996) argued on the basis of past records that many German officials were willing participants in the Holo- caust—not mere ordinary people forced to follow orders. Citing historical records, others have similarly argued that Nazi killers knew, believed in, and celebrated their mission (Cesarani, 2004; Haslam & Reicher, 2007; Vetle- sen, 2005).

Yet two lines of evidence suggest that laying blame on this subgroup of German people is too simple as an expla- nation of what happened. First, interviews with Nazi war criminals and doctors who worked in concentration camps suggested, at least to some, the provocative and disturb-

ing conclusion that these people were “utterly ordinary” (Arendt, 1963; Lifton, 1986; Von Lang & Sibyll, 1983). Second, the monstrous events of World War II do not stand alone in modern history. Even today, various crimes of obedience— which may include torture, suicide bombings, and public beheadings—are being committed in ruthless regimes, militaries, and terrorist organizations through- out the world (Haritos-Fatouros, 2002; Kelman & Hamilton, 1989; Victoroff & Kruglanski, 2009).

As seen in recent Wall Street scandals, crimes of obedience are also found in the corporate world, where business leaders and their subordinates “morally disengage” from fraud and other unethical actions by denying personal respon- sibility, minimizing consequences, and dehumanizing victims (Beu & Buck- ley, 2004; Moore et al., 2012). On extraordinary but rare occasions, obedience is carried to its ultimate limit. In 1978, 900 members of the People’s Temple cult obeyed a command from Reverend Jim Jones to kill themselves by drink- ing poison. In 1997, Marshall Applewhite, leader of the Heaven’s Gate cult in California, killed himself and convinced 37 followers to do the same. Fanatic cult members had committed mass suicide before, and they will likely do so again (Galanter, 1999).

Taken to the extreme, blind obedience can have devastating results. Nazi officials killed millions during World War II, and many said they did it “because I was just following orders.”

To ph

am /T

he Im

ag e

W or

“Far more, and far more hideous, crimes have been committed in the name of obedience than have ever been committed in the name of rebellion.”

—C. P. Snow

80220_ch07_ptg01_hr_262-307.indd 292 27/01/16 11:29 AM

293 Obedience

j Milgram’s research: Forces of Destructive Obedience During the time that Adolf Eichmann was being tried in Jerusalem for his Nazi war crimes, Stanley Milgram began a dramatic series of 18 experiments. The first was published in the Journal of Abnormal and Social Psychology in 1963; the rest were reported later in his 1974 book, Obedience to Authority. Milgram did not realize it at the time—and neither did his research participants—but they were about to make history in one of the most famous psychology research programs ever conducted.

For many years, the ethics of this research has been the focus of much de- bate. Those who say it was not ethical point to the potential psychological harm to which Milgram’s participants were exposed. In contrast, those who believe that these experiments met appropriate ethical standards emphasize the profound contribution it has made to our understanding of human nature and an important social problem. They conclude that on balance, the danger that destructive obedi- ence poses for all humankind justified Milgram’s unorthodox methods. Consider both sides of the debate, which were summarized in Chapter 2, and make your own judgment. Now, however, take a more personal look. Imagine yourself as one of the approximately 1,000 participants who found themselves in a situation much like the following.

The experience begins when you arrive at a Yale University laboratory and meet two men. One is the experimenter, a stern young man dressed in a gray lab coat and carrying a clipboard. The other is a middle-aged gentleman named Mr. Wallace, an accountant who is slightly overweight and average in appearance. You exchange quick introductions, and then the experimenter explains that you and your co-participant will take part in a study on the effects of punishment on learning. After lots have been drawn, it is determined that you will serve as the teacher and that Mr. Wallace will be the learner. So far, so good.

Soon, however, the situation takes on a more ominous tone. You find out that your job is to test the learner’s memory and administer electric shocks of increas- ing intensity whenever he makes a mistake. You are then escorted into another room, where the experimenter straps Mr. Wallace into a chair, rolls up his sleeves, attaches electrodes to his arms, and applies “electrode paste” to prevent blisters and burns. Mr. Wallace is concerned but the experimenter responds by reassuring him that although the shocks will be painful, the procedure will not cause “per- manent tissue damage.” In the meantime, you can personally vouch for how pain- ful the shocks are because the experimenter stings you with one that is supposed to be mild. The experimenter then takes you back to the main room, where you

This is the shock generator Milgram used (left). It still exists and can be seen in the Archives of the History of American Psychology at the University of Akron. Participants in Milgram’s studies believed they were shocking Mr. Wallace, the man being strapped into his chair (right).

Fr om

th e

fil m

O be

di en

ce b

y St

an le

y M

ilg ra

m c

op yr

ig ht

1 96

5 an

d di

st rib

ut ed

b y

Th e

Pe nn

. S ta

te U

ni ve

rs ity

A ud

io V

is ua

l S er

vi ce

80220_ch07_ptg01_hr_262-307.indd 293 27/01/16 11:29 AM

294 Chapter 7 Conformity

are seated in front of a “shock generator,” a machine with 30 switches that range from 15 volts, labeled “slight shock,” to 450 volts, labeled “XXX.”

Your role in this experiment is straightforward. First you read a list of word pairs to Mr. Wallace through a microphone. Then you test his memory with a

series of multiple-choice questions. The learner answers each question by pressing one of four switches that light up signals on the shock genera- tor. If his answer is correct, you move on to the next question. If it is incorrect, you announce the correct answer and shock him. When you press the appropriate shock switch, a red light flashes above it, relay switches click inside the machine, and you hear a loud buzzing sound go off in the learner’s room. After each wrong answer, you’re told, the intensity of the shock should be increased by 15 volts.

You aren’t aware, of course, that the experi- ment is rigged and that Mr. Wallace—who is actu- ally a confederate—is never really shocked. As far as you know, he gets zapped each time you press one of the switches. As the session proceeds, the learner makes more and more errors, leading you to work your way up the shock scale. As you reach 75, 90, and 105 volts, you hear the learner grunt in pain. At 120 volts, he begins to shout. If you’re still in it at 150 volts, you can hear the learner cry out, “Experimenter! That’s all. Get me out of here. I re- fuse to go on!” Screams of agony and protest con- tinue. At 300 volts, he says he absolutely refuses to continue. By the time you surpass 330 volts, the learner falls silent and fails to respond—not to be heard from again. Table 7.4 lists his responses in grim detail.

Somewhere along the line, you turn to the experimenter for guidance. “What should I do? Don’t you think I should stop? Shouldn’t we at least check on him?” You might even confront the experimenter head-on and refuse to continue. Yet in answer to your inquiries, the experimenter— firm in his tone and seemingly unaffected by the learner’s distress—prods you along as follows:

m T a b l e 7 . 4

The learner’s Protests in the milgram experiment

As participants administered progressively more intense shocks, they heard the learner moan, groan, protest, and complain. All participants heard the same programmed set of responses. Eventually, the learner fell silent and ceased to respond.

75 volts Ugh!

90 volts Ugh!

105 volts Ugh! (louder)

120 volts Ugh! Hey, this really hurts!

135 volts Ugh!!

150 volts Ugh!!! Experimenter! That’s all. Get me out of here. I told you I had heart trouble. My heart’s starting to bother me now. Get me out of here, please. My heart’s starting to bother me. I refuse to go on. Let me out.

165 volts Ugh! Let me out! (shouting)

180 volts Ugh! I can’t stand the pain. Let me out of here! (shouting)

195 volts Ugh! Let me out of here. Let me out of here. My heart’s bothering me. Let me out of here! You have no right to keep me here! Let me out! Let me out of here! Let me out! Let me out of here! My heart’s bothering me. Let me out! Let me out!

210 volts Ugh!! Experimenter! Get me out of here. I’ve had enough. I won’t be in the experiment any more.

225 volts Ugh!

240 volts Ugh!

255 volts Ugh! Get me out of here.

270 volts (Agonized scream) Let me out of here. Let me out of here. Let me out of here. Let me out. Do you hear? Let me out of here.

285 volts (Agonized scream)

300 volts (Agonized scream) I absolutely refuse to answer any more. Get me out of here. You can’t hold me here. Get me out. Get me out of here.

315 volts (Intensely agonized scream) I told you I refuse to answer. I’m no longer part of this experiment.

330 volts (Intense and prolonged agonized scream) Let me out of here. Let me out of here. My heart’s bothering me. Let me out, I tell you. (Hysterically) Let me out of here. Let me out of here. You have no right to hold me here. Let me out! Let me out! Let me out! Let me out of here! Let me out! Let me out!

d Please continue (or, please go on).

d The experiment requires that you continue.

d It is absolutely essential that you continue.

d You have no other choice; you must go on.

What do you do? In a situation that begins to feel more and more like a bad dream, do you follow your own conscience or obey the experimenter? Milgram described this procedure to psychiatrists, college students, and middle-class adults, and he

80220_ch07_ptg01_hr_262-307.indd 294 27/01/16 11:29 AM

295 Obedience

asked them to predict how they would behave. On average, these groups estimated that they would call it quits at the 135-volt level. Not a single person thought he or she would go all the way to 450 volts. When asked to predict the percentage of other people who would deliver the maximum shock, those interviewed gave similar estimates. The psychiatrists estimated that only one out of a thousand people would exhibit that kind of extreme obedience. They were wrong. In the study just described, involving 40 men from the New Haven area, participants exhibited an alarming degree of obedience, administering an average of 27 out of 30 possible shocks. In fact, 26 of the 40 participants—that’s 65%—delivered the ultimate punishment of 450 volts.

The Obedient Participant At first glance, you may see these results as a lesson in the psychology of cruelty and conclude that Milgram’s participants were seriously disturbed. But research does not support such a simple explanation. To begin with, those in a “control group” who were not prodded along by an experimenter refused to continue early into the shock sequence. What’s more, Milgram found that virtu- ally all participants, including those who had administered severe shocks, were tor- mented by the experience. Many of them pleaded with the experimenter to let them stop. When he refused, they continued. But in the process, they trembled, stuttered, groaned, perspired, bit their lips, and dug their fingernails into their flesh. Some burst into fits of nervous laughter. On one occasion, said Milgram, “we observed a [participant’s] seizure so violently convulsive that it was necessary to call a halt to the experiment” (1963, p. 375).

Was Milgram’s 65% “baseline” level of obedience attributable to his unique sample of male participants? Not at all. Forty women who participated in a later study exhibited precisely the same level of obedience: 65% threw the 450-volt switch. Before you jump to the conclusion that something was amiss in New Haven, consider the fact that Milgram’s basic finding has been obtained in several cultures and with children as well as college students and older adults (Blass, 2012). Obedience in the Milgram situation is so universal that it led one author to ask, “Are we all Nazis?” (Askenasy, 1978).

The answer, of course, is no. An individual’s character makes a difference, and some people, depending on the situation, are far more obedient than oth- ers. In the aftermath of World War II, a group of social scientists—searching for the root causes of prejudice and bigotry—sought to identify individuals with an authoritarian personality and developed a questionnaire known as the F-Scale to measure it (Adorno et al., 1950; Stone et al., 1993). What they found is that people who get high scores on the F-Scale (F stands for “Fascist”) are rigid, dogmatic, sexually repressed, ethnocentric, intolerant of dissent, and punitive. They are sub- missive toward figures of authority but aggressive toward subordinates. Indeed, people with high F scores are also more willing than low scorers to administer high-intensity shocks in Milgram’s obedience situation (Elms & Milgram, 1966).

Although personality characteristics may make someone vulnerable or resistant to destructive obedience, it is clear that the situation in which people find themselves has a profound effect. By carefully altering particular aspects of his basic scenario, Milgram was able to identify factors that increase and decrease the 65% baseline rate of obedience in more than 20 variations of the basic experiment (see d Figure 7.8). Three factors in particular are important: the authority figure, the proximity of the victim, and the experimental procedure (Blass, 1992; Miller, 1986).

The Authority What is perhaps most remarkable about Milgram’s findings is that an experimenter in a white lab coat is not a powerful figure of authority. Unlike

80220_ch07_ptg01_hr_262-307.indd 295 27/01/16 11:29 AM

296 Chapter 7 Conformity

military superiors, employers, coaches, or teachers, the experimenter in Milgram’s research could not ultimately enforce his commands. Still, his physical presence and apparent legitimacy played major roles in drawing obedience. When Milgram dimin- ished the experimenter’s status by moving his lab from the distinguished surround- ings of Yale University to a run-down urban office building in nearby Bridgeport, Connecticut, the rate of total obedience dropped to 48%. When the experimenter was replaced by an ordinary person—supposedly another participant—there was a sharp reduction to 20%. Similarly, Milgram found that when the experimenter was in charge but issued his commands by telephone, only 21% fully obeyed. (In fact, when the experimenter was not watching, many participants in this condition feigned obedience by pressing the 15-volt switch.) One conclusion, then, is clear. At least in the Milgram setting, destructive obedience requires the physical presence of a prestigious authority figure.

If an experimenter can exert such control over research participants, imagine the control wielded by truly powerful leaders—whether they are physically pres- ent or not. An intriguing field study examined the extent to which hospital nurses would obey unreasonable orders from a doctor. Using a fictitious name, a male physician called several female nurses on the phone and told them to administer a drug to a specific patient. His order violated hospital regulations: The drug was uncommon, the dosage was too large, and the effects could have been harmful. Yet out of the 22 nurses who were contacted, 21 had to be stopped as they pre- pared to obey the doctor’s orders (Hofling et al., 1966).

d F i g u r e 7 . 8

Factors That influence Obedience

Milgram varied many factors in his research program. Without commands from an experimenter, fewer than 3% of the participants exhibited full obedience. Yet in the standard baseline condition, 65% of male and female participants followed the orders. To identify factors that might reduce this level, Milgram varied the location of the experiment, the status of the authority, the participant’s proximity to the victim, and the presence of confederates who rebel. The effects of these variations are illustrated here. Milgram, 1974

Control — no commands

Baseline — males

Baseline — females

Of ce building

Ordinary person in charge

Experimenter in remote location

Victim in same room as participant

Participant required to touch victim

Two confederates rebel

10 20 30 40 50 60 70

Percentage of participants who exhibited full obedience

80220_ch07_ptg01_hr_262-307.indd 296 27/01/16 11:29 AM

297 Obedience

The Victim Situational characteristics of the victim are also important factors in destructive obedience. Milgram noted that Nazi war criminal Adolf Eichmann felt sick when he toured concentration camps but only had to shuffle papers from be- hind a desk to play his part in the Holocaust. Similarly the B-29 pilot who dropped the atom bomb on Hiroshima in World War II said of his mission, “I had no thoughts, except what I’m supposed to do” (Miller, 1986, p. 228). These events suggest that because Milgram’s participants were physically separated from the learner, they were able to distance themselves emotionally from the consequences of their actions.

To test the impact of a victim’s proximity on destructive obedience, Milgram seated the learner in one of his studies in the same room as the participant. Under these conditions, only 40% fully obeyed. But when participants were required to physically grasp the victim’s hand and force it against his will onto a metal shock plate, full obedience dropped to 30%. These findings represent significant reduc- tions from the 65% baseline. Still, 3 out of 10 participants were willing to use brute force in the name of obedience.

Milgram ran one last variation, never published, that was only recently dis- covered in Yale University’s library, where Milgram’s records have been archived for historical purposes. Located in a rundown office building in a nearby city, he called it the “Bring-a-Friend” condition. In this situation, 20 male participants were told to bring a friend, neighbor, family member, or coworker. When they arrived, the two participants drew straws to determine who would be the teacher and who the learner. Unbeknownst to the teacher, Milgram then coached the learner on how to play the role and react to the shocks according to script, at times pleading to the teacher by name. Would participants obey less in this version? Yes. Compared to a Wallace-as-victim condition run in the same location, where the full obedience rate was 50%, the obedience rate in the friend condition was only 15%. Following in the footsteps of some German citizens who defied the Nazi regime by protecting friends they were ordered to report, Milgram’s participants were indeed capable of resistance when they had a prior relationship with the victim (Rochat & Blass, 2014).

The Procedure Finally, there is the carefully scripted situation created by Milgram. Looking at the dilemma that confronted Milgram’s participants, Burger (2014) points to four critical aspects of the experimental procedure that contributed to the surpris- ing results. First, participants were led to feel relieved of personal responsibility for the victim’s welfare. The experimenter said up front that he was accountable. Burger notes that in a partial replication of Milgram that he more recently conducted, a “Milgram-lite” version to be described later, only 12% of participants who exhibited full obedience gave any indication during their sessions that they bore any sense of responsibility. In fact, when participants are led to believe that they were respon- sible, their levels of obedience drop considerably (Tilker, 1970).

The ramifications of felt responsibility are immense. In the military, on Wall Street, in corporations and other organizations, individuals often occupy positions in a hierarchical chain of command. Eichmann was a mid-level bu- reaucrat who received orders from Hitler and transmitted them to others for implementation. Caught between individuals who make policy and those who carry it out, how personally responsible do those in the middle feel? Wesley Kilham and Leon Mann (1974) examined this issue in an obedience study that cast participants in one of two roles: the transmitter (who took orders from the experimenter and passed them on) and the executant (who actually pressed the shock levers). As they predicted, transmitters were more obedient (54%) than executants (28%).

80220_ch07_ptg01_hr_262-307.indd 297 27/01/16 11:29 AM

298 Chapter 7 Conformity

The second feature of Milgram’s scenario that promoted obedience is the use of gradual escalation in small increments. Participants began the session by de- livering mild shocks and then only gradually escalated to voltage levels of high intensity. After all, what’s another 15 volts compared with the current level? By the time they realized the frightening implications of what they were doing—for example, choosing to align with the experimenter once the learner protested at 150 volts—it had become more difficult to stop (Gilbert, 1981; Packer, 2008). This sequence is much like the foot-in-the-door technique. In Milgram’s words, people become “integrated into a situation that carries its own momentum. The subject’s problem . . . is how to become disengaged from a situation which is moving in an altogether ugly direction” (1974, p. 73). We should point out that obedience by momentum is not unique to Milgram’s research paradigm. As reported by Am- nesty International, many countries still torture political prisoners, and those who are recruited for the dirty work are trained, in part, through an escalating series of commitments (Haritos-Fatouros, 2002).

According to Burger (2014), the third feature of Milgram’s situation that made it difficult for participants to resist is that participants found themselves in a novel situation, unimaginable, like no other they have been in before. As such, they did not know what the norms were, how others have reacted, or how they were sup- posed to respond. That is why, in a variation of Milgram’s research in which two confederates posed as co-participants who refused to continue, the full obedience rate plummeted to 10% (see Figure 7.8).

The fourth aspect of the procedure that promoted obedience is that the task was quickly paced. At the outset, Milgram’s experimenter instructed participants to work at a “brisk pace.” Indeed, as one can plainly see in a film that Milgram had produced of his sessions, those who hesitated were immediately prompted to proceed. All this gave participants no time to ponder, consider their values and their options, think about possible consequences, or make careful decisions.

j Milgram in the Twenty-First Century When Stanley Milgram published the results of his first experiment in 1963, at the age of 28, a New York Times headline read: “Sixty-Five Percent in Test Blindly Obey Order to Inflict Pain.” Milgram had pierced the public consciousness and was poised to become an important and controversial figure in psychology—and beyond. In a biography, Thomas Blass (2004) tells of how Milgram became inter- ested in obedience and the impact his studies have had on social scientists, legal scholars, the U.S. military, and popular culture around the world (Milgram’s book has been translated into 11 languages). Now, in an age filled with threats of global conflict, extremism, terrorism, economic hardship, breaches to cybersecurity and privacy, and new forms of lethal weaponry, obedience to authority is an issue of such importance that social psychologists all over the world continue to ponder its ramifications (Benjamin & Simpson, 2009; Blass, 2009; Jetten & Mols, 2014; Reicher et al., 2014). Milgram himself died in 1984 at the age of 51.

Today, a grainy black-and-white film that Milgram produced in 1965 in which a number of sessions were recorded from a hidden camera stands as visual proof of this phenomenon. It is clear from looking at this film that this experiment was conducted in another era, a pre-computer, pre-digital era in which research participants called the young experimenter “sir.” Would these results be repeated today? Would you obey the commands of Milgram’s experimenter?

In an effort to answer this question, Dutch researchers Wim Meeus and Quinten Raaijmakers (1995) created a different but analogous situation. They

in experiments on obedience, most participants who were ordered to administer severe shocks to an innocent person refused to do so.

FALSe

80220_ch07_ptg01_hr_262-307.indd 298 27/01/16 11:29 AM

299 Obedience

constructed a moral dilemma much like Milgram’s. Rather than order participants to inflict physical pain on someone, however, they ordered them to cause psy- chological harm. When participants arrived at a university laboratory, they met a confederate supposedly there to take a test as part of a job interview. If the con- federate passed the test, he’d get the job; if he failed, he would not. As part of a study of performance under stress, the experimenter told participants to distract the test-taking applicant by making an escalating series of harassing remarks. On cue, the applicant pleaded with participants to stop, became angry, faltered, and eventually fell into a state of despair and failed. As in Milgram’s research, the question was straightforward: How many participants would obey orders through the entire set of 15 stress remarks, despite the apparent harm caused to a real- life job applicant? In a control group that lacked a prodding experimenter, no one persisted. But when the experimenter ordered them to go on, 92% exhibited complete obedience despite seeing the task as unfair and distasteful. It appears that obedience is a powerful aspect of human nature brought about by the docile manner in which people relate to figures of authority—even today.

In a more recent—and even more direct—attempt to revisit Milgram, Burger (2009) conducted a “partial replication” for which he paid $50 to 70 men and women, a diverse group that ranged from 20 to 81 years old, and used the same procedure. In the original experiment, the learner first protested and asked to stop at 150 volts, at which point nearly all participants paused and indicated a reluctance to continue. Some outright refused at this point. Of those participants who did continue, however, most went all the way. On the basis of this finding, Burger followed the Milgram protocol up to 150 volts in order to esti- mate the number of participants who would have pulled the switch at 450 volts. He also added a condition in which a defiant confederate posing as another participant refused to continue. In light of post-Milgram changes in standards for research ethics, he took additional precautions; he excluded from the study individuals he feared would experience too much stress and then in- formed and reminded participants three times that they could withdraw from the study at any time without penalty.

Despite all that has changed in 45 years, the obedience rate was not appreciably lower (see d Figure 7.9). In the experiment that Burger modeled, 83% of Milgram’s participants had continued past 150 volts. In Burger’s more recent study, 70% did the same (it can be estimated, therefore, that 55% would have exhibited full 450-volt obedience in the original experiment). Two additional results proved interesting: (1) Just as Milgram had found, there were no dif- ferences between men and women, and (2) the obedience rate declined only slightly, to 63%, among participants who saw a defiant confeder- ate refuse to continue.

d F i g u r e 7 . 9

Obedience in the Twenty-First Century

In the version of the experiment that Burger modeled, 83% of Milgram’s original participants continued past 150 volts. Forty-five years later, Burger saw a slight drop to 70%. Note too that the obedience rate dropped only slightly, to 63%, among participants who saw a defiant confederate refuse to continue. These results show that obedience to authority may have declined a bit over the years, but it has by no means extinguished. Burger, 2009

Milgram’s experiment

Continued Stopped

Burger’s base condition

De�ant confederate

100

Percentages of Participants Who Continued v. Stopped at 150 Volts

80220_ch07_ptg01_hr_262-307.indd 299 27/01/16 11:29 AM

300 Chapter 7 Conformity

Obedience to authority is a timeless phenomenon. In 2004, then 18-year-old Louise Ogburn, working in a McDonald’s in Kentucky, was accused of stealing money by a prank phone caller and taken into her manager’s office. Posing as police, the caller ordered several of the girl’s coworkers to have her strip naked. For four hours, she was searched, spanked, and forced into humiliating positions on command. All but one coworker obeyed the prank caller—and the girl obeyed the coworkers. Remarkably, seventy similar incidents were reported elsewhere. As seen here, Ms. Ogburn sued McDonald’s and was awarded $6 million (left). Perhaps even more astonishing, Juan Rivera of Illinois spent 20 years in prison for a crime he did not commit after he was badgered into the ultimate act of obedience—a false confession to murder. In March of 2015, Rivera and his attorneys announced that he was awarded $20 million in a settlement against prosecutors and police who coerced this confession (right).

A P

Im ag

es /B

ria n

Bo ha

nn on

This recent replication has drawn a good deal of interest. Alan Elms (2009), a graduate student of Milgram’s in the 1960s, is cautious about comparing Burger’s “obedience lite” procedure to Milgram’s but eager to see it revitalize the research program Milgram had initiated. Arthur Miller (2009), author of The Obedience Experiments, voices the same cautious excitement. Blass (2009), author of the Milgram biography, The Man Who Shocked the World, sees Burger’s experiment as an important milestone that demonstrates the stability and resilience of obedi- ence in human social behavior. In contrast, Jean Twenge (2009), author of Gen- eration Me—a 2006 book on how Americans have become more self-centered and wholly focused on personal rights—is skeptical of that conclusion that noth- ing has changed. Making precise comparisons, Twenge notes that relative to an obedience rate of 83% among Milgram’s male participants, only 67% of Burger’s men exhibited 150-volt obedience, a decline that is statistically comparable to the alarming change in U.S. obesity rates during that same period of time. Although impressed with the power of Milgram’s situation, Twenge is hopeful that destruc- tive obedience is less prevalent today, in the twenty-first century, than in the past.

j Lingering Questions Before leaving the Milgram studies, let’s consider two important questions: Why exactly did Milgram’s participants follow orders, and what are the moral implica- tions of their behavior?

On the why question, Milgram had theorized that participants were swept up in a situational momentum he had created, leading them to exhibit obedience. What they displayed was not a “blind obedience,” as some commentators have suggested; the personal conflict they experienced was palpable—but it was obedi- ence nevertheless. In contrast, Alex Haslam, Stephen Reicher, and their colleagues recently offered an alternative account—an engaged followership explanation of the results (Reicher et al., 2012; Haslam et al., 2014). These researchers theorize that participants shocked the learner because they identified with the scientific enterprise and wanted to both help the experimenter and make a contribution.

A P

Im ag

es /C

ha rle

s Re

x A

rb og

as t

80220_ch07_ptg01_hr_262-307.indd 300 27/01/16 11:29 AM

301 Obedience

From this perspective, their behavior indicated their engagement in the science, not obedience to authority. In support of this account, they noted that Milgram’s second prod, which appealed to science (“The experiment requires that you con- tinue”) was followed by more shock to the learner than his fourth prod, which took the form of a strong command (“You have no other choice, you must go on”).

To test this hypothesis in a single new experiment, Haslam et al. (2014) cre- ated an online procedure in which they asked participants to engage in a task that involved denigrating certain groups of people in a way that became increasingly unpleasant—much like climbing a shock scale. In this case, they randomly varied which prod came first and found, as predicted, that 64% of participants completed the study after receiving Prod 2, the science prod, compared to only 44% who received Prod 4, the obedience prod. This is an interesting result. But for two rea- sons it represents only a weak test of the obedience hypothesis. First, “You have no choice” is exactly the kind of command phrase that arouses psychological reac- tance and leads people in general to refuse in order to protect their threatened free- dom (see Chapter 6). Second, the participants completed this task via computer without the physical presence of an authority figure, which Milgram had shown to be necessary. Obedience or engaged followership? It will be interesting to see what future studies will be conducted to distinguish these possible explanations.

Then there is the moral question: By providing a situational explanation for the evils of Nazi Germany or modern-day terrorism, do social psychologists unwittingly excuse the perpetrators? Does focusing on situational forces let them off the hook of responsibility? Since many of Milgram’s participants were disobedient, indicating that they were free to choose resistance, one would hope not. Andrew Monroe and Glenn Reeder (2014) note that observers integrate information about the participant, his or her behavior, and the situation, in a manner that accounts for the subtle na- ture of the participant’s motives in light of the dilemma, “as caught between want- ing to help the learner and wanting to placate the experimenter” (p. 550).

In a series of studies, Miller and others (1999) found that after people were asked to come up with explanations for acts of wrongdoing, they tended to be more forgiving of those who committed the acts and they were seen as more forgiving by others. This appearance of forgiveness was certainly not Milgram’s intent, nor is it the intent of other researchers today who seek to understand hu- man cruelty, even while continuing to condemn it. Miller and his colleagues were thus quick to caution, “To explain is not to forgive” (p. 265).

j Defiance: When People rebel History books will call it the “Arab Spring.” On December 17, 2010, a young, un- employed, frustrated Tunisian man by the name of Mohamed Bouazizi set fire to himself in protest after police confiscated the fruits and vegetables he was selling on the street because he lacked a permit. Almost immediately, men and women took to the streets in protest. Despite the government’s determination to suppress dissent, the crowds grew in size, marching, holding rallies, and demanding the resignation of the prime minister. On January 14, 2011, the government was overthrown and set off protests in other countries throughout the Middle East. It’s no wonder the regimes in power sought to censor and block the social media (Howard & Hussain, 2011).

Reading Milgram’s research, it is easy to despair in light of the impressive forces that compel people toward blind obedience. But there’s also good news. Just as social influence processes can breed subservience to authority, social in- fluence processes can also breed rebellion and defiance. The Arab Spring is a modern-day example. A similar phenomenon was seen during World War II. In Re- sistance of the Heart, historian Nathan Stoltzfus (1996) described a civil protest in

80220_ch07_ptg01_hr_262-307.indd 301 27/01/16 11:29 AM

302 Chapter 7 Conformity

Berlin in which the non-Jewish wives of 2,000 newly captured Jews congregated outside the prison. The women were there initially to seek information about their husbands. Soon they were filling the streets, chanting and refusing to leave. After eight straight days of protest, the defiant women prevailed. Fearing the negative impact on public opinion, the Nazis backed down and released the men.

In addition to the role of social networking media, recent studies indicate that for better or for worse, synchrony of behavior—for example, walking in step with others, clapping, singing, chanting, or raising arms in unison—can have a unifying effect on people, increasing the tendency to follow what others are doing. In one study, pairs of participants were seated in rocking chairs, side by side, and asked to rock in unison. Other pairs also rocked, but they could not see each other and were not in rhythm. Those in the synchrony condition were later more “in synch” when working jointly to move a steel ball through a wooden maze (Valdesolo et al., 2010). Other studies have shown that acting in unison with others can also increase our tendency to feel so- cially connected, cooperate for the common good, and even comply with a request to aggress against another person (Wiltermuth, 2012; Wiltermuth & Heath, 2010).

Are the actions of a whole group harder to control than the behavior of a sin- gle individual? Consider the following study. Pretending to be part of a marketing research firm, William Gamson and others (1982) recruited people to participate in a supposed discussion of “community standards.” Scheduled in groups of nine, participants were told that their discussions would be videotaped for a large oil company that was suing the manager of a local service station who had spoken out against higher gas prices. After receiving a summary of the case, most par- ticipants sided with the station manager. But there was a hitch. The oil company wanted evidence to win its case, said the experimenter posing as the discussion coordinator. He told each of the group members to get in front of the camera and express the company’s viewpoint. Then he told them to sign an affidavit giving the company permission to edit the tapes for use in court.

You can see how the obedience script was supposed to unfold. Actually, only 1 of 33 groups even came close to following the script. In all others, people were incensed by the coordinator’s behavior and refused to continue. Some groups were so outraged that they planned to take action. One group even threatened to blow

In Hong Kong, in September of 2014, thousands of pro-democracy protesters waved phones activated with LED lights in a show of unison and instantly sent powerful images across the world. The power of this social-media facilitated demonstration led the government of China to block the photo-sharing service Instagram.

“A little rebellion now and then is a good thing.”

—Thomas Jefferson

Co rb

is N

ew s/

Co rb

80220_ch07_ptg01_hr_262-307.indd 302 27/01/16 11:29 AM

303 The Continuum of Social Influence

the whistle on the firm by calling the local newspapers. Faced with one emotionally charged mutiny after another, the researchers had to discontinue the experiment.

Why did this study produce such active, often passionate revolt when Milgram’s revealed such utterly passive obedience? Could it reflect a change in values from the 1960s, when Milgram’s studies were run? Many college students believe that people would conform less today than in the past, but an analysis of obedience studies has revealed that there is no correlation between the year a study was conducted and the level of obedience that it produced (Blass, 1999)— right up through Burger’s (2009) recent effort. So what accounts for the con- trasting results? One key difference is that people in Milgram’s studies took part alone and those in Gamson’s were in groups. Perhaps Michael Walzer was right: “Disobedience, when it is not criminally but morally, religiously, or politically motivated, is always a collective act” (quoted in Brown, 1986, p. 17).

Our earlier discussion of conformity indicated that the mere presence of one ally in an otherwise unanimous majority gives individuals the courage to dissent. The same may hold true for obedience. Notably, Milgram typically did not have more than one participant present in the same session. But in one experiment, he did use two confederates who posed as co-teachers along with the real partici- pant. In these sessions, one confederate refused to continue at 150 volts and the second refused at 210 volts. These models of disobedience had a profound influ- ence on participants’ willingness to defy the experimenter: In their presence, only 10% delivered the maximum level of shock (see Figure 7.8).

We should add that the presence of a group is not a guaranteed safeguard against destructive obedience. Groups can trigger aggression, as we’ll see in Chap- ter 11. For example, the followers of Jim Jones were together when they collec- tively followed his command to die. And lynch mobs are just that—groups, not individuals. Clearly, there is power in sheer numbers. That power can be destruc- tive, but it can also be used for constructive purposes. Indeed, the presence and support of others often provide the extra ounce of courage that people need to resist orders they find offensive.

The Continuum of Social Influence As we have seen, social influence on behavior ranges from the implicit pressure of group norms to the traps set by direct requests to the powerful commands of au- thority. In each case, people choose whether to react with conformity or indepen- dence, with compliance or assertiveness, and with obedience or defiance. From all the research, it is tempting to conclude that the more pressure brought to bear on people, the greater the influence. Is it possible, however, that more produces less? In a series of studies, Lucian Conway and Mark Schaller (2005) cast partici- pants into a corporate decision-making task in which they were asked to choose between two business options after watching others make the same decision. Con- sistently, the participants followed the group more when its members had formed their opinions freely than when they were compelled by a leader. It appears that strong-arm tactics that force people to change their behavior may backfire when it comes to changing opinions.

At this point, let’s step back and ask two important questions. First, although different kinds of pressure influence us for different reasons, is it possible to pre- dict all effects with a single overarching principle? Second, what does all the theory and research on social influence say about human nature?

80220_ch07_ptg01_hr_262-307.indd 303 27/01/16 11:29 AM

304 Chapter 7 Conformity

social impact theory The theory that social influence depends on the strength, immediacy, and number of source persons relative to target persons.

j Social impact Theory In 1981, Bibb Latané proposed that a common bond among the different processes involved in social influence leads people toward or away from such influence. Specifically, Latané proposed social impact theory, which states that social influ- ence of any kind—the total impact of others on a target person—is a function of the others’ strength, immediacy, and number. According to Latané, social forces act on individuals in the same way that physical forces act upon objects. Consider, for example, how overhead lights illuminate a surface. The total amount of light cast on a surface depends on the strength of the bulbs, their distance from the sur- face, and their number. As illustrated in the left portion of Figure 7.10, the same factors apply to social impact.

The strength of a source is determined by his or her status, ability, or rela- tionship to a target. The stronger the source, the greater the influence. When people view the other members of a group as competent, they are more likely to conform in their judgments. When it comes to compliance, sources enhance their strength by making targets feel obligated to reciprocate a small favor. And to elicit obedience, authority figures gain strength by wearing uniforms or flaunting their prestigious affiliations.

Immediacy refers to a source’s proximity in time and space to the target. The closer the source, the greater its impact. Milgram’s research offers the best example. Obedience rates were higher when the experimenter issued commands in person rather than from a remote location, and when the victim suffered in close prox- imity to the participant, he acted as a contrary source of influence and obedience levels dropped. Consistent with this hypothesis, Latané and others (1995) asked individuals to name up to seven people in their lives and to indicate how far away those people lived and how many memorable interactions they’d had with them. In three studies, the correlation was the same: The closer others are, geographically, the more impact they have on us.

Finally, the theory predicts that as the number of sources increases, so does their influence—at least up to a point. You may recall that when Asch (1956) increased the number of live confederates in his line- judgment studies from one to four, conformity levels

rose, yet further increases had only a negligible addi- tional effect.

Social impact theory also predicts that people sometimes resist social pres- sure. According to Latané, this resistance is most likely to occur when social impact is divided among many strong and distant targets, as seen in the right part of d Figure 7.10. There should be less impact on a target who is strong and far from the source than on one who is weak and close to the source, and there should be less impact on a target who is accompanied by other target persons than on one who stands alone. Thus, we have seen that conformity is reduced by the presence of an ally and that obedience rates drop when people are in the company of rebellious peers.

Over the years, social impact theory has been challenged, defended, and re- fined on various grounds (Jackson, 1986; Mullen, 1985; Sedikides & Jackson,

According to social impact theory, this “intervention” should prove persuasive.

“I invited a few friends over who think you should see a psychiatrist.”

Fr an

k Co

th am

Th e

N ew

Y or

ke r C

ol le

ct io

n /T

he C

ar to

on B

an k

80220_ch07_ptg01_hr_262-307.indd 304 27/01/16 11:29 AM

305 The Continuum of Social Influence

1990). On the one hand, critics say that it does not enable us to explain the processes that give rise to social influence or answer why questions. On the other hand, the theory enables us to predict the emergence of social influence and determine when it will occur. Whether the topic is conformity, compliance, or obedience, this theory provides a stage for inter- esting new research in the years to come.

A number of social psycholo- gists have recently argued that social impact is a fluid, dynamic, ever-changing process (Vallacher et al., 2002). Fifteen years before the Arab Spring became a real- ity, Latané and L’Herrou (1996) refined the theory in that vein. By having large groups of par- ticipants network through e-mail and by controlling their lines of communication, they found that the individuals within the net- work formed “clusters.” Over time, neighbors (participants who were in direct e-mail contact with each other) became more similar to each other than did those who were more distant (not in direct e-mail contact) within the network. Referring to the geometry of social space, Latané and L’Herrou note that in the real world, immediacy cannot be defined strictly in terms of physical distance. Speculating on the role of technology, they noted that social impact theory has to account for the type of events that unfolded in the Middle East, in 2011, and that fact that remote social media may diminish the importance of physical proximity.

j Perspectives on Human Nature From the material presented in this chapter, what general conclusions might you draw about human nature? Granted, social influence is more likely to occur in some situations than in others. But are people generally malleable or unyielding? Is there a tilt toward accepting influence or toward putting up resistance?

There is no single universal answer to these questions. As we saw earlier, some cultures value autonomy and independence; others place greater emphasis on conformity to one’s group. Even within a given culture, values may change over time. To demonstrate the point, ask yourself: If you were a parent, what traits would you like your child to have? When this question was put to American mothers in 1924, they chose “obedience” and “loyalty,” key characteristics of conformity. Yet when mothers were asked the same question in 1978, they cited “independence” and “tolerance of others,” key characteristics of autonomy.

d F i g u r e 7 . 1 0

Social impact: Source Factors and Target Factors

According to social impact theory, the total influence of other people, or “sources,” on a target individual depends on three source factors: their strength (size of source circles), immediacy (distance from the target), and number (number of the source circles). Similarly, the total influence is diffused, or reduced, by the strength (size of target circles), immediacy (distance from source circle), and number of target persons. From Latane, B. (1981). “The Psychology of Social Impact,” American Psychologist, vol 36 (p. 344).

Target

Targets

Source

Sources

As the number of people in a group increases, so does the group’s impact on an individual.

FALSe

80220_ch07_ptg01_hr_262-307.indd 305 27/01/16 11:29 AM

306 Chapter 7 Conformity

Similar trends were found in surveys con- ducted not only in the United States but also in West Germany, Italy, England, and Japan (Alwin, 1990; Remley 1988) and in laboratory experiments, where conformity rates are somewhat lower today than in the past (Bond & Smith, 1996).

Is it possible that today’s children, tomorrow’s adults, will exhibit more re- sistance to the various forms of social influ- ence? If so, what effects will this trend have on society as a whole? And what will be the future effects of Facebook, Twitter, and other social media? Cast in a positive light, conformity, compliance, and obedience are good and necessary human responses. They promote group solidarity and agreement— qualities that keep groups from being torn

apart by dissension. Cast in a negative light, a lack of independence, assertive- ness, and defiance are undesirable behaviors that lend themselves to narrow- mindedness, cowardice, and destructive obedience, often with terrible costs. For each of us and for society as a whole, the trick is to strike a balance.

According to social impact theory, a marine sergeant will exert influence to the extent that he is strong (in a position of power), immediate (physically close), and numerous (backed by others in the institution) relative to his trainees.

Bo b

Kr is

Conformity rates vary across different cultures and from one generation to the next.

True

1. Sometimes people are influenced by others without awareness—as when we mimic each other’s nonverbal behaviors and gestures without realizing it.

2. Sherif’s (1936) classic study showed that sometimes we conform to other people’s judgments and behaviors be- cause we are uncertain of what is correct and use others for informational guidance.

3. Asch’s (1951) classic study showed that sometimes we conform to other people’s judgments and behaviors even when they are clearly incorrect because we do not want to deviate and risk rejection.

4. Conformity is increased by the size of a unanimous ma- jority, up to a point, though the presence of a single dis- senting ally empowers people to resist the pressure to conform.

5. By staking out a consistent and unwavering position, dis- senting individuals inside a group can exert “minority influ- ence” and enhance the quality of a group’s decision making.

6. Just as cultures differ in terms of their social norms, they differ in adherence to those norms—people from collec- tivist cultures conform more than do people from indi- vidualistic cultures.

7. At times, people are likely to comply with direct re- quests—for example, when taken by surprise, when the request sounds reasonable, and when they feel indebted to the requestor.

8. Over the years, social psychologists have studied two- step request approaches that lead people to gain com- pliance—for example, by using the foot-in-the-door and door-in-the-face techniques.

Top 10 Key Points in Chapter 7

Review

80220_ch07_ptg01_hr_262-307.indd 306 27/01/16 11:29 AM

307 Review

9. In Milgram’s classic, controversial, and profound research, 65% of subjects fully obeyed an experimenter’s command to administer increasingly painful electric shocks to a con- federate.

10. Even today, Milgram’s experiments continue to gener- ate controversy and inform important aspects of human nature and the question of how to minimize destructive acts of obedience in real life settings.

Putting Common Sense to the Test

Conformity

When all members of a group give an incorrect response to an easy question, most people most of the time conform to that response. F False. In Asch’s classic conformity experiments,

respondents conformed only about a third of the time.

Compliance

An effective way to get someone to do you a favor is to make a first request that is so large the person is sure to reject it. T True. This approach, known as the door-in-the-face

technique, increases compliance by making the person feel bound to make a concession.

Obedience

in experiments on obedience, most participants who were ordered to administer severe shocks to an innocent person refused to do so.

F False. In Milgram’s classic research, 65% of all participants obeyed the experimenter and administered the maximum possible shock.

The Continuum of Social influence

As the number of people in a group increases, so does the group’s impact on an individual. F False. Increasing group size boosts the impact on

an individual only up to a point, beyond which further increases have very little added effect.

Conformity rates vary across different cultures and from one generation to the next. T True. Research shows that conformity rates are higher

in cultures that are collectivistic rather than individualistic in orientation and that values change over time even within cultures.

collectivism (282) compliance (283) conformity (268) door-in-the-face technique (289) foot-in-the-door technique (286) idiosyncrasy credits (280)

individualism (282) informational influence (271) lowballing (288) minority influence (279) normative influence (271) obedience (292)

private conformity (273) public conformity (273) social impact theory (304) that’s-not-all technique (290)

Key Terms

80220_ch07_ptg01_hr_262-307.indd 307 27/01/16 11:29 AM

Group Processes

This chapter examines social influence in a group context. First, we focus on the fundamentals of groups, in which we discuss issues such as why people are drawn to groups and how groups develop. We then turn to how the behavior of individuals is affected by the presence of others. Then we focus on group performance and discuss why the whole (the group decision or performance) so often is different from the sum of its parts (the attitudes and abilities of the group members). In the final section, on conflict, we examine how groups intensify or reconcile their differences.

80220_ch08_ptg01_hr_308-349.indd 308 27/01/16 5:13 PM

Group Processes

8 Fundamentals of Groups | 311

What Is a Group? Why Join a Group?

Key Features of Groups: Roles, Norms, and Cohesiveness

Culture and Cohesiveness

Individuals in Groups: The Presence of Others | 316

Social Facilitation: When Others Arouse Us

Social Loafing: When Others Relax Us

Culture and Social Loafing

Deindividuation

Group Performance: Problems and Solutions | 325

Losses and Gains in Groups

Brainstorming

Group Polarization

Groupthink

Communicating Information and Utilizing Expertise

Goals and Plans in Groups

Training and Technology

Virtual Teams

Culture and Diversity

Collective Intelligence: Are Some Groups Smarter Than Others?

Conflict: Cooperation and Competition Within and Between Groups | 339

Mixed Motives and Social Dilemmas

Negotiation

Culture and Negotiation

Finding Common Ground

Review

309Ca rin

P er

ill ou

80220_ch08_ptg01_hr_308-349.indd 309 27/01/16 5:13 PM

310 Chapter 8 Group Processes

It was a road trip like no other. On July 15, 2015, a vehicle reached a destination that had been a long time coming, and sent photos back home. When members of the U.S. National Aeronautics and Space Administration (NASA) received the photos, celebrations erupted. Their spacecraft, New Horizons, had finally approached Pluto, which, depending on your classifica- tion scheme, is either the farthest planet in our solar system or the most beloved dwarf planet in the Kuiper belt. Either way, it’s far, far away, and this was an in- credible achievement. The historic scientific and engi- neering feat necessary to complete this nearly 10-year, 3-billion-mile journey to the far reaches of our solar system required not only the tireless work of a very large group of people, but also precise coordination among all its members. This is what groups can do at their best.

While impressive, it may not seem surprising that this super-intelligent, hardworking group of people at NASA could pull this off. After all, they really are rocket scientists. But history, and social psychology, teach us that groups of even super-smart people can make re- ally bad decisions, and there is something about the dynamics of groups that can make this happen. For ex- ample, two of the biggest tragedies in NASA’s history, the explosions of the space shuttles Challenger and Columbia, in 1986 and 2003, respectively, occurred in no small part because of bad group processes—involving

Bi ll

In ga

lls /N

A SA

/G et

ty Im

ag es

N ew

s/ G

et ty

Im ag

NASA team members celebrate in July 2015 the extraordinary feat of its space vehicle, New Horizons, flying close up to Pluto and sending back remarkable photos.

Putting Common SenSe to the Test

T F People will cheer louder when they cheer as part of a group than when they cheer alone.

T F People brainstorming as a group come up with a greater number of better ideas than the same number of people working individually.

T F Group members’ attitudes about a course of action usually become more moderate after group discussion.

T F People and groups tend to do worse when they have “do your best” goals than when they have very specific, ambitious goals.

T F Large groups are more likely than small groups to exploit a scarce resource that the members collectively depend on.

T F When people or groups negotiate with each other, the best solution is one in which both parties compromise and split the resources 50–50.

Circle Your Answer

80220_ch08_ptg01_hr_308-349.indd 310 27/01/16 5:13 PM

311 Fundamentals of Groups

group A set of individuals who interact over time and have shared fate, goals, or identity.

We begin our exploration of groups by asking the basic questions: What is a group? Why do people join groups? We then focus on three important aspects of groups: roles, norms, and cohesiveness.

j What Is a Group? Why Join a Group? What Is a Group? The question might seem quite simple, but if you step back and think about it, the answer is less obvious. For example, many students are mem- bers of a variety of groups on social media. Are these really groups? You may be part of a large social psychology class: Is this a group identity that is meaningful to you?

A group may be characterized as a set of individuals who have direct interactions with each other over a period of time and share a common fate, identity, or set of goals. A group can also consist of people who have joint membership in a social cat- egory based on sex, race, or other attributes; this characteristic is especially relevant for the issues discussed in Chapter 5 on Stereotypes, Prejudice, and Discrimination.

Groups vary in the extent to which they are seen as distinct entities, such as whether they have rigid boundaries that make them distinct from other groups. In other words, some groups seem more “groupy” (or in the unfortunate choice of word used by researchers on this issue: “entitative”) than others (Brewer, 2015; Hamilton et al., 2011). On the very low end of this dimension would be people attending a concert or working out near each other in a gym. These typically are not considered real groups. Such assemblages are sometimes called collectives— people engaging in a common activity but having little direct interaction with each other (Milgram & Toch, 1969). Much more integrated groups include tight-knit clubs, sports teams, or work teams—groups that engage in very purposeful activi- ties with a lot of interaction over time and clear boundaries of who is in and not in the group. People tend to identify more strongly with these more integrated, co- herent groups and to get more satisfaction from them (Crawford & Salaman, 2012).

Culture can also shape the nature of what makes a group a group. People in Western cultures are more likely to define and identify with groups based on what

factors such as poor communication, biased sampling of information, and pres- sure toward conformity. This, unfortunately, is also what groups can do.

In short, people are often at their best—and their worst—in groups. It is through groups that individuals form communities, pool resources, and share suc- cesses. But it is also through groups that ideas stagnate in endless discussion, selfish impulses flourish in the anonymity of a crowd, and prejudices turn into genocide and war. Clearly, it is important that we understand how groups work and how individuals influence, and are influenced by, groups. In this chapter, we first introduce the fundamentals of what groups are and how they develop, then we examine groups on several levels: At the individual level, we explore how individuals are influenced by groups; at the group level, we explore how groups perform; and at the intergroup level, we explore how groups interact with each other in cooperation and competition.

The research we report in this chapter reveals a fascinating fact: Groups can be quite different from the sum of their parts. When you think about that statement, it suggests something almost mystical or magical about groups. How can a group be better—or worse—than its individual members? The math may not seem to add up, but the theory and research discussed in this chapter will help answer this question.

“You think because you understand ‘one,’ you must understand ‘two’ because one and one make two. But you must also understand ‘and.’”

—Ancient Sufi saying

Fundamentals of Groups

80220_ch08_ptg01_hr_308-349.indd 311 27/01/16 5:13 PM

312 Chapter 8 Group Processes

members do, whereas in Eastern cultures how group members relate to each other may be more important (Yuki, 2003).

As we will discuss later in this chapter, more groups today defy our traditional view of what is a group. Particularly in the worlds of business and technology, groups often consist of people who are dispersed widely across time and space and communicate exclusively through technol- ogy, and the dynamics of these groups may be extremely different from those of traditional groups. For example, these newer, more dis- persed types of groups tend to have more shift- ing boundaries and membership, and they tend to be more self-managed or have shared lead- ership (Hackman & Katz, 2010). Understanding these unique dynamics and determining how to help these groups reach their potential is one of the more exciting challenges in contemporary social psychological research on groups.

Why Join a Group? The complexity and ambitions of human life require that we work in groups. Much of what we hope to produce and accomplish can be done only through collective action. Lone individuals cannot play symphonies or football games, build cities or industries, or run governments or universities.

At a more fundamental level, humans may have an innate need to belong to groups stemming from evolutionary pressures that increased people’s chances of survival and reproduction if they lived in groups rather than in isolation. Indeed, according to the social brain hypothesis, the unusually large size of humans’ brains evolved because of our unusually complex social worlds. As we wrote in the first paragraph of Chapter 1: We have such large brains in order to socialize (Dunbar, 2014; Spunt et al., 2015; van Vugt & Kameda, 2014).

For humans, attraction to group life serves not only to protect against threat and uncertainty in a physical sense but also to gain a greater sense of personal and social identity. According to social identity theory, which was discussed in Chapter 5, an important part of people’s feelings of self-worth comes from their identification with particular groups, and so people care a great deal about being part of groups and about how their groups are valued (Ellemers & Haslam, 2012; Greenaway et al., 2015; Hogg, 2014; Jetten et al., 2015). Beyond feelings of self- worth, our groups often give us meaning and purpose and, according to some scholars, perhaps even a symbolic sense of immortality as our most cherished groups live on beyond our lifetimes (Schimel & Greenberg, 2013). These cherished aspects of group membership are at the root of why being rejected by a group is one of life’s most painful experiences (Eisenberger, 2015).

j Key Features of Groups: Roles, Norms, and Cohesiveness

Once an individual has joined a group, a process of adjustment takes place as the individual is socialized to how things work in the group. This socialization process may be formal and explicit, such as through an initiation or orientation program, mentoring or supervision, or documentation, or it may be implicit, as

People join a group for any of several reasons, such as to affiliate with others, obtain social status, and interact with individual group members.

Sa ra

h Su

dh of

f/ Re

du x

80220_ch08_ptg01_hr_308-349.indd 312 27/01/16 5:13 PM

313 Fundamentals of Groups

newcomers observe how established members behave in the group. Effectively socializing new members can produce short-term and long-term benefits for the group as a whole (Levine & Choi, 2010; Moreland & Levine, 2002). One thing that newcomers can do to more quickly get older members to accept them and utilize their knowledge is to use more group-focused language (such as using “we” and “us” rather than “I” and “you”) with group members (Kane & Rink, 2015).

Two of the things that are especially important for newcomers to learn—and for more established members of the group to continue to understand or revise—are the roles they are expected to play in the group, and what the norms of the group are. Newcomers also learn how cohesive the group is. We focus on each of these three features of groups—roles, norms, and cohesiveness—in the following sections.

Roles People’s roles in a group—their set of expected behaviors—can be formal or informal. Formal roles are designated by titles: teacher or student in a class, vice president or account executive in a corporation. Informal roles are less obvious but still powerful. Robert Bales (1958) distinguished between two fundamental types of roles: an instrumental role to help the group achieve its tasks and an expressive role to provide emotional support and maintain morale. The same person can fill both roles, but often they are assumed by different individuals, and which of these roles is emphasized in groups may fluctuate over time, depending on the needs of the group.

One problem that can seriously harm the performance of groups is when there is a mismatch between members’ skills and what roles they occupy in the group. It is far too common in groups for members to be assigned or take on group roles in a way that is less than thoughtful and systematic. Members may be assigned roles simply based on who is available at a given point in time rather than who is best suited for the role. Groups function much better when members are assigned roles that best match their talents and personalities, and social psychologists are among the consultants often hired to help groups do this better (Woolley et al., 2007).

Group members sometimes are uncertain about exactly what their group roles are supposed to be. They may also find themselves in roles that conflict with other roles they have to play, either within the group (such as needing to be demand- ing while also being the source of emotional support) or between groups (such as between work and family). Role uncertainty, instability, and conflict are all associated with poorer job performance, as well a vari- ety of other problems, including workplace bullying or other interpersonal conflicts, emotional exhaustion and burnout, and high turnover in the group (Bowling et al., 2015; Dishon-Berkovits, 2014; Kauppila, 2014; Tubre & Collins, 2000; Wilson & Baumann, 2015).

Sometimes an opposite problem of role uncertainty or conflict develops: Group members can become so absorbed in their role that they lose themselves—and their personal beliefs and sense of morality—in their group role. Interrogators of a prisoner or suspect may get so lost in their role that they fail to recognize the ethical lines they are crossing. One of the most disturb- ing studies in the history of social psychology examined this kind of issue, in what came to be known as The Stanford Prison Experiment (Haney et al., 1973). Partici- pants in this study were randomly assigned to play the roles of prisoners or guards in a simulated prison. The guards in particular got so into their roles as protectors of the “prison” and enforcer of the rules that they soon became cruel and sadistic.

“The individuals which took the greatest pleasure in society would best escape various dangers, whilst those that cared least for their comrades, and lived solitary, would perish in greater numbers.”

—Charles Darwin The Descent of Man

When roles in a group are not distributed properly, group performance suffers.

“I’ve had to be both hunter and gatherer.”

To m

C he

ne y

/T he

N ew

Y or

ke r C

ol le

ct io

n/ Th

e Ca

rt oo

n Ba

80220_ch08_ptg01_hr_308-349.indd 313 27/01/16 5:13 PM

314 Chapter 8 Group Processes

Even the social psychologist in charge of the study was so focused on his role as “prison warden” that he ignored his ethical responsibilities as a researcher. This fascinating but troubling study is discussed in detail in Chapter 12 on Law.

Norms In addition to roles for its members, groups also establish norms, rules of conduct for members. Like roles, norms may be either formal or informal. Fraterni- ties and sororities, for example, usually have written rules for the behavior expected from their members. Informal norms are more subtle. What do I wear? How hard can I push for what I want? Who pays for this or that? What kind of language, joking, or socializing is typical? These norms provide individuals with a sense of what it means to be a good group member. Figuring out the unwritten rules of the group can take time and cause anxiety.

Groups often exert strong conformity pressures on individuals who deviate from group norms and perceive or treat these members very harshly, in part because de- viations from group norms can threaten group members’ sense of uniformity and social identity with the group (Hutchison et al., 2013; Packer, 2014). One way that low-status or new group members may try to establish a stronger position in the group is to be especially punishing of members who break group norms—especially if high-status group members would witness their harsh reaction to norm violators

(Jetten et al., 2010). On the other hand, group members who are highly identified with the group and care about its collective success may be willing to deviate from a group norm if they think that the norm is likely to harm the group (Jetten & Hornsey, 2014; Packer et al., 2014).

How tolerant groups are to violations of norms can be, itself, a kind of norm. Some groups, for example, pride themselves on how heterogeneous and free-thinking its members are. Others strongly value uniformity. Paul Hutchison and others (2011) designed an ex- periment to see if they could manipulate this sense of how uniform the group members are supposed to be. They asked some British students questions about their university that were designed to high- light the uniformity of the students’ backgrounds and attitudes at their school. For example, they were asked what percentage of the students preferred popular music over classical music, and what per- centage liked to watch movies. Because the large majority of the stu- dents would prefer popular music and would like movies, questions like these would prime the students to see their similarities. The re- searchers asked other students questions induced to make them see their school as full of diverse backgrounds and attitudes. They were asked to estimate the percentages of students who most preferred dance, rock, hip-hop, pop, or classical music, and what percentages most preferred movies featuring science fiction, love stories, comedy, martial arts, and so on. These questions would highlight the wide variety of opinions and tastes that were accepted at their university.

The students then read about a student who expressed an attitude about a war that was either typical or atypical of the students there. How would the participants react to this student? Participants in general rated the student with the typical attitude toward the war positively. If the student expressed an atypical attitude, however, they disliked him more if they were induced to see their university as very uniform than if they were induced to see their university as diverse and heterogeneous (see d Figure 8.1). In other words, a deviation from a group norm was met with more tolerance if the

d F I G u R e 8 . 1

Tolerance for Deviating From the Norm

Students evaluated a fellow student who expressed an attitude that was either typical or atypical for their university. In general, the student expressing the typical attitude was evaluated more positively than the atypical student. But if the students had first been primed to see their university as heterogeneous, they were less negative toward the atypical student than were the students who had been primed to see their university as homogenous. Hutchison et al., 2011.

Typical

Homogeneous Heterogeneous

Ev al

ua tio

Atypical

Target group member

80220_ch08_ptg01_hr_308-349.indd 314 27/01/16 5:13 PM

315 Fundamentals of Groups

group cohesiveness The extent to which forces push group members closer together, such as through feelings of intimacy, unity, and commitment to group goals.

participants saw the group as more variable than if they saw their group as typically sharing similar background and values. Indeed, when a group develops a strong norm of encouraging diverse viewpoints, it can cause members to conform to the norm of not conforming!

Culture and Norms Cultures vary in how much they tolerate behavior that deviates from the norm. Anthropolo- gist Pertii Pelto (1968) first described the difference be- tween “tight” and “loose” cultures. He proposed that tight cultures have strong norms and little tolerance for behav- ior that deviates from the norm, while loose cultures have relatively weaker norms and greater tolerance for deviant behavior. Michele Gelfand and others (2011, 2012) demon- strate that greater ecological and historical threats (such as resource scarcity or natural disasters), higher population density, and more restrictive governments and religious in- stitutions encourage the formation of tight societies. Loose societies, in contrast, are more likely to thrive in environ- ments that have fewer historical and ecological threats and are characterized by situations that have few constraints on individuals, allowing individuals to behave according to their own discretion. Analyzing the similarity between people’s values, norms, and behavior across 65 countries, Uz (2015) identified Egypt, Indonesia, and Morocco as the tightest countries and Belgium, Luxembourg, and France as the loosest countries (see Table 8.1). However, one does not have to travel to another country to experience differences in tightness and looseness. For example, within the United States, Harrington and Gelfand (2014) found that the southern states of Mississippi, Alabama, and Arkansas had the highest tightness rankings, while the western states of California, Oregon, and Washington had the highest looseness rankings.

Cohesiveness Groups whose members share similar attitudes and closely follow the groups’ norms are more likely than other groups to be cohesive. Group cohe- siveness refers to the forces exerted on a group that push its members closer together (Cartwright & Zander, 1960; Festinger, 1950). Members of cohesive groups tend to feel commitment to the group task, feel positively toward the other group members, feel group pride, and engage in many—and often intense—interactions in the group (Dion, 2000; Rosh et al., 2012).

An interesting question is whether cohesiveness makes groups perform better. It may seem obvious that it should. Indeed, groups often strive to achieve cohesiveness, and if they feel that they are not very cohesive, they may take steps to improve it, such as ordering everyone to go through bonding exercises or hang out together at a rustic retreat. In fact, however, the relationship between cohesiveness and performance is not a simple one. The causal relationship works both ways: On the one hand, when a group is cohesive, group performance often improves; on the other hand, when a group performs well, it often becomes more cohesive. Many team athletes recognize that winning creates team chemistry even more than team chemistry creates winning.

Whatever the cause and effect, several longitudinal studies and meta-analyses of hundreds of studies have provided evidence showing that group cohesion is associated with better performance, but other variables tend to be important in

Based on Uz, 2015

m T a b l e 8 . 1

Tightness–looseness of Various Countries

Cultures considered “tight” have strong and homogenous norms, values, and behaviors, and there is much less tolerance for deviance from norms. The numbers next to the countries listed in this table are an index according to one recent analysis of how tight–loose these countries are, with higher numbers indicating greater looseness.

Argentina 75 Luxembourg 114

Bangladesh 7 Mexico 75

Belgium 120 Morocco 0

Canada 84 Netherlands 79

Egypt 4 Nigeria 18

France 100 Philippines 32

Greece 58 Poland 43

India 44 Russian Fed. 57

Indonesia 3 Saudi Arabia 22

Ireland 71 South Africa 68

Japan 43 Spain 84

Jordan 5 USA 58

80220_ch08_ptg01_hr_308-349.indd 315 27/01/16 5:13 PM

316 Chapter 8 Group Processes

predicting when and to what extent this re- lationship emerges, such as how large the group is, whether the cohesiveness is pri- marily about the group’s attraction to the task (task cohesion) or to each other (inter- personal cohesion), and what type of task the group performs (Castaño et al., 2013; Mathieu et al., 2015; Picazo et al., 2015). As we will see a bit later in the chapter, group cohesiveness can also lead to conformity and narrow-mindedness that can cause major problems for groups.

j Culture and Cohesiveness Group cohesiveness can be affected in different ways as a function of cultural

differences. For example, cohesiveness in collectivist cultures may be associated more with social harmony, cooperation, and interpersonal relations than in indi- vidualist cultures, where recognizing members’ unique skills, perspectives, and job-focused efforts may be more essential for group cohesiveness (Lai et al., 2013). Cultural differences exist also in norms about how leaders are expected to behave in groups, and this has consequences for group cohesiveness. Respect and obedience to leaders are more important to people from collectivist cultures than those in individualist cultures. Consistent with this point, Hein Wendt and others (2009) found in a study of over 150,000 individuals from 615 organizations across 80 countries that having a directive, controlling leader was associated with greater group cohesiveness in collectivist societies than in individualist cultures.

The extent to which groups are comfortable with conflict and heated debate among their members also varies across culture. This was demonstrated in a study by Roger Nibler and Karen Harris (2003) involving five-person groups of strangers and friends in China and the United States. Each group had to decide how to rank 15 items to be taken aboard a lifeboat from a ship that was about to sink. This task tends to trigger a fair amount of initial disagreement among group members before a consensus can be reached. With the Chinese groups and the groups of American strangers, these kinds of disagreements tended to be perceived as troubling and inter- fered with group performance. To the groups of American friends, in contrast, these disagreements were more likely to be seen as simply part of a freewheeling debate, and the sense of freedom to exchange opinions and disagree with one another tended to improve rather than hurt performance on this task. More recently, Daan Bisseling and Filipe Sobral (2011), using surveys and interviews of 366 members of companies from the Netherlands and Brazil, found that intragroup conflict was associated with reduced satisfaction and job performance among employees in the more collectivistic culture of Brazil, but not among the more individualistic Dutch workers.

20 th

C en

tu ry

F ox

F ilm

C or

p/ Ev

er et

t C ol

le ct

io n

A scene from the classic cult movie, Office Space, depicts frustrated employees from a work group that is anything but cohesive.

Individuals in Groups: The Presence of Others When we engage in activities in groups, we are in the presence (either physically or virtually) of others. It’s an obvious point, but some of its consequences are profound and surprising. In this section we focus on three important effects that the presence of others can have on individuals: social facilitation, social loafing, and deindividuation.

80220_ch08_ptg01_hr_308-349.indd 316 27/01/16 5:13 PM

317 Individuals in Groups: The Presence of Others

j Social Facilitation: When Others Arouse us Social psychologists have long been fascinated by how the presence of others affects behavior. In Chapter 1, we reported that Norman Triplett’s article, “The Dynamogenic Factors in Pacemaking and Competition” (1897–1898), is often cited as the earliest publication in the field. Triplett began his research by studying the official bicycle records from the Racing Board of the League of American Wheelmen for the 1897 sea- son. He noticed that cyclists who competed against others performed better than those who cycled alone against the clock. After dismissing various theories of the day (our favorite is “brain worry”), he proposed his own hypothesis: The presence of another rider releases the competitive instinct, which increases nervous energy and enhances performance. To test this proposition, Triplett got 40 children to wind up fishing reels, alternating between performing alone and working in parallel. Triplett reported that children were more likely to perform better when they worked side by side than when they worked alone. (In a contemporary twist on this classic, Michael Strube (2005) used modern statistical techniques to reanalyze Triplett’s original data and found that the results were quite weak and not as straightforward as Triplett’s report suggested.)

Later research following Triplett’s studies proved mixed. Sometimes the presence of others (side by side or with an audience out front) enhanced performance; at other times, performance declined. It seemed that Triplett’s promising lead had turned into a blind alley, and social psychologists had largely abandoned this research by World War II. But years later, Robert Zajonc (1965; 1980) saw a way to reconcile the con- tradictory results by integrating research from experimental psychology with social psychological research. Zajonc offered an elegant solution: The presence of others increases arousal, which can affect performance in different ways, depending on the task at hand. Let’s see how this works.

The Zajonc Solution Zajonc proposed a three-step process:

1. The presence of others creates general physiological arousal, which energizes behavior. Based on experimental psychol- ogy research and principles of evolution, Zajonc argued that all animals, including humans, tend to become aroused when in the presence of conspecifics—that is, members of their own species.

2. Increased arousal enhances an individ- ual’s tendency to perform the dominant response. The dominant response is the reaction elicited most quickly and easily by a given stimulus.

3. The quality of an individual’s perfor- mance varies according to the type of task. On an easy task (one that is simple or well learned), the dominant response is usually correct or successful. But on a difficult task (one that is complex or un- familiar), the dominant response is often incorrect or unsuccessful.

Putting these three steps together (see d Figure 8.2) yields the following scenarios. Suppose you are playing the

d F I G u R e 8 . 2

Social Facilitation: The Zajonc Solution

According to Zajonc, the presence of others increases arousal, which strengthens the dominant response to a stimulus. On an easy task, the dominant response is usually correct, and thus the presence of others enhances performance. On a difficult task, the dominant response is often incorrect, and thus the presence of others impairs performance. © Cengage Learning®

Presence of another person or member of the same species

Increased arousal

Strengthened dominant response

Incorrect response Performance impairment

Correct response Performance enhancement

Easy Task Difficult Task

80220_ch08_ptg01_hr_308-349.indd 317 27/01/16 5:13 PM

318 Chapter 8 Group Processes

social facilitation A process whereby the presence of others enhances performance on easy tasks but impairs performance on difficult tasks.

violin. If you’re an excellent player and are performing a well-learned, familiar ar- rangement, having other people around should enhance your performance; the presence of others will increase your arousal, which will enhance your dominant response. Because this arrangement is so well learned, your dominant response will be to perform it well. However, if you are just learning to play the violin and you are unfamiliar with this arrangement, the presence of others is the last thing you’ll want. The increase in arousal should enhance the dominant response, which in this case would be unsuccessful violin playing.

Taken as a package, these two effects of the presence of others—helping per- formance on easy tasks but hurting performance on difficult tasks—are known as social facilitation. Unfortunately this term has been a prime source of confusion for countless students. The trick is to remember that the presence of others facili- tates the dominant response, not necessarily the task itself. This facilitation of the dominant response does, in effect, facilitate easy tasks, but it makes difficult tasks even more difficult.

Zajonc proposed that social facilitation is universal, occurring not only in hu- man activities but also among other animals, even insects. Have you ever wondered, for instance, how well a cockroach performs in front of other cockroaches? Neither did we—that is, we didn’t until we first learned of Zajonc’s creative research on the topic. Zajonc and his colleagues (1969) had cockroaches placed in a brightly lit start box connected to a darkened goal box. When the track was a simple one, with a straight runway between the start box and the goal box, cockroaches running in pairs ran more quickly toward the goal box than did those running alone. But in a more complex maze that required a right turn to reach the goal box, solitary cockroaches outraced pairs. In a particularly creative follow-up experiment, Zajonc and his col- leagues found that cockroaches completed the easy maze faster, and the difficult maze slower, if they raced in front of a crowd of spectator cockroaches than if they raced with no audience. How did the researchers get cockroaches to participate as spectators? The researchers placed cockroaches in plexiglass “audience boxes” along either side of the maze, and this “audience” produced social facilitation.

U.S. swimmer Michael Phelps surges to the lead on the way to one of his record-setting gold medals at the 2012 Olympics in London. Through social facilitation, some of the best athletes benefit from the presence of an audience and high pressure when performing well-learned routines.

Fr an

co is

X av

ie r M

ar it/

A FP

/G et

ty Im

ag es

80220_ch08_ptg01_hr_308-349.indd 318 27/01/16 5:13 PM

319 Individuals in Groups: The Presence of Others

mere presence The proposition that the mere presence of others is sufficient to produce social facilitation effects.

evaluation apprehension theory A theory that the presence of others will produce social facilitation effects only when those others are seen as potential evaluators.

distraction–conflict theory A theory that the presence of others will produce social facilitation effects only when those others distract from the task and create attentional conflict.

Social Facilitation Research Today Zajonc’s formulation revived interest in the issues raised by Triplett’s early research, and suddenly the inconsistent find- ings that had been reported began to make sense. The results of a meta-analysis of 241 studies were consistent with much of Zajonc’s account (Bond & Titus, 1983). And despite its long history, research today continues to demonstrate new examples of social facilitation and to test its scope and limitations. Rui-feng Yu and Xin Wu (2015), for example, found that the presence of an observer slowed down partici- pants’ screening of baggage in an X-ray security scan when the task was complex but sped up the performance when the task was simple. Social facilitation effects have been demonstrated in settings where individuals are taking driving tests (a word of advice: Don’t take your road test with another test-taker present in the car!), gambling electronically, and being given neuropsychological tests, and the effects have been found even when the “others” present were merely a photograph of a favorite TV character or a computer display of a “virtual” person (Eastvold et al., 2012; Gardner & Knowles, 2008; Park & Catrambone, 2007; Rockloff et al., 2012; Rosenbloom et al., 2007).

It is in part because of the effects of social facilitation that firefighters, police officers, military personnel, and others must train so much to be ready to make split-second decisions under highly arousing situations. They may seem to practice scenarios to the point of overtraining, but it is only through such repetition that their dominant response can be assured to typically be the correct one. When they are in the midst of a raging fire or mili- tary firefight, it is often impossible for them to think carefully, and so it’s all the more essential that their dominant response be a good one (Gladwell, 2005).

Alternative explanations for Social Facilitation Social facilitation effects have been replicated across many domains, but not all of Zajonc’s theory has received universal support. Zajonc proposed that the mere presence of others is sufficient to produce social facilitation. Some have argued, however, that a better expla- nation is the evaluation apprehension theory, which proposes that performance will be enhanced or impaired only in the presence of others who are in a position to evaluate that performance (Geen, 1991; Henchy & Glass, 1968). In other words, it’s not simply because others are around that I’m so aroused and therefore inept as I try to learn to snowboard on a crowded mountain. Rather, it’s because I worry that the others are watching and probably laughing at me, possibly uploading a video of my performance to YouTube. These concerns increase my dominant response, which, unfortunately, is falling.

Another account of social facilitation, distraction–conflict theory, points out that being distracted while we’re working on a task creates attentional conflict (Baron, 1986; Sanders, 1981). We’re torn between focusing on the task and glancing at the distracting stimulus. When we are conflicted about where to pay attention, our arousal increases.

So is one of these theories right and the others wrong? Probably not. It seems likely that all three of the basic elements described by these theories (mere pres- ence, evaluation, and attention) can contribute to the impact others have on our own performance (Uziel, 2007). But as we are about to see in the next section, there is even more to the story of how individuals are affected by the presence of others.

Fear of being evaluated by others can sometimes hurt performance.

“God, this is going to be all over YouTube.”

M ar

sh al

l H op

ki ns

/ Th

e N

ew Y

or ke

r C ol

le ct

io n/

Th e

Ca rt

oo n

Ba nk

80220_ch08_ptg01_hr_308-349.indd 319 27/01/16 5:13 PM

320 Chapter 8 Group Processes

social loafing A group-produced reduction in individual output on tasks where contributions are pooled.

j Social Loafing: When Others Relax us In the tasks relevant to social facilitation effects, an individual’s be- havior can be identified and evaluated. But on some tasks, efforts are pooled so that the specific performance of any one individual can- not be determined. It just so happens that research on these kinds of collective endeavors can also be traced back to Triplett’s era of the late nineteenth century. In Chapter 1 we mentioned the French agricultural engineer Max Ringelmann as another contender for hav- ing conducted the first social psychology research. In research con- ducted during the 1880s, Ringelmann discovered that people’s output declined when they worked together rather than alone on simple tasks like pulling a rope or pushing a cart (Kravitz & Martin, 1986; Ringelmann, 1913).

d F I G u R e 8 . 3

Social Loafing: When Many Produce Less

Social loafing is a group-produced reduction in individual output on simple tasks. In this study, college students were told to cheer or clap as loudly as they could. The noise produced by each of them decreased as the size of the group increased. Based on Latané et al., 1979.

So un

d pr

es su

re p

er p

er so

Group size

1 2 4 6

ClappingCheering

People will cheer louder when they cheer as part of a group than when they cheer alone.

FALSe

Why did individual output decline? One explanation is that the individuals exerted less effort when they acted collectively, but an- other explanation is that the individuals simply demonstrated poor coordination when working together—some pulled while others re- laxed and vice versa. How can you distinguish lack of effort from poor coordination in a task like this? Nearly 100 years after Ringelmann’s research, Alan Ingham and his colleagues (1974) answered this ques- tion by using a rope-pulling machine and blindfolding participants. In one condition, participants were led to think that they were pulling with a bunch of other participants, and in another condition the par- ticipants were informed that they were pulling alone (which, in fact, they were). The researchers told the participants to pull as hard as they could. Ingham and colleagues were able to measure exactly how hard each individual participant pulled, and they observed that the participants pulled almost 20% harder when they thought they were pulling alone than when they thought they were pulling with others.

Bibb Latané and his colleagues (1979) found that group-produced reductions in individual output, which they called social loafing, are

common in other types of tasks as well. For example, imagine being asked as part of a psychology experiment to cheer or clap as loudly as you can. Common sense might lead you to think that you would cheer and clap louder when doing this together with others in a group than when performing alone because you would be less em- barrassed and inhibited if others were doing the same thing as you. But Latané and his colleagues found that when performing collectively, students tended to loaf— they exerted less effort. The noise generated by each individual decreased as the size of the group increased (see d Figure 8.3). This social loafing occurred even among cheerleaders, who are supposed to be experts at cheering and clapping with others!

Social loafing is not restricted to simple motor tasks. Sharing responsibility with others reduces the amount of effort that people put into more complex motor tasks, such playing team sports; cognitive tasks, such as completing memory, math, or verbal tests; and important, enduring real-world behaviors, such as working collab- oratively on collective farms or team projects (Hoigaard & Ommundsen, 2007; Liden et al., 2004; Miles & Greenberg, 1993; Plaks & Higgins, 2000; Tan & Tan, 2008; Weldon et al., 2000). When others are there to pick up the slack, people slack off.

Most college students are quite familiar with having to work on group projects, and many have seen the pitfalls of social loafing in these settings. Praveen Aggarwal and Connie O’Brien (2008) studied several hundred college students to assess what factors can reduce the incidence of social loafing, and they offer these three strategies: (1) limit the scope of the project—projects that are very large and complex should

80220_ch08_ptg01_hr_308-349.indd 320 27/01/16 5:13 PM

321 Individuals in Groups: The Presence of Others

collective effort model The theory that individuals will exert effort on a collective task to the degree that they think their individual efforts will be important, relevant, and meaningful for achieving outcomes that they value.

be broken into smaller components; (2) keep the groups small; and (3) use peer evaluations. Curt Dommeyer (2012)’s research suggests also that making each group member a “manager” of different segments of the task can also help reduce loafing.

Businesses have taken note of social loafing and have applied these research findings in an effort to reduce it in the workplace. For example, one form of social loafing at the workplace has come to be known as cyberloafing, which involves personal non-work use of online technology, such as online shopping, watching videos, or messaging friends; cyberloafing can be a huge drain on workers’ pro- ductivity (Askew et al., 2014; Glassman et al., 2015; Karaoğlan et al., 2015). One result is that workers’ actions on the job are coming under increasing surveillance as the number of computer keystrokes they make per hour or the content of their calls, e-mails, or Internet browsing can be recorded electronically.

Collective effort Model Several researchers have constructed theoretical accounts to explain the findings about when social loaf- ing is more or less likely to occur. The most influential of these is the collective effort model (Karau & Williams, 2001). This model as- serts that individuals will try hard on a collective task when they think their efforts will help them achieve outcomes they person- ally value. If the outcome is important to individual members of the group and if they believe they can help achieve the desired outcome, they are less likely to socially loaf. In fact, in these cases they may even engage in social compensation by increasing their efforts on collective tasks to try to compensate for the anticipated social loafing or poor performance of other group members. The next time you work on a group project, such as a paper that you and several other students are supposed to write together, consider the factors that increase and decrease social loafing. You might want to try to change aspects of the situation so that all group members are motivated to do their share of the work.

j Culture and Social Loafing Steven Karau and Kipling Williams (1993) conducted a meta-analysis of 78 studies and found social loafing to be a reliable phenomenon that is evident across numerous tasks and in countries around the world. Despite its prevalence around the world, some group and cultural differences in tendencies to socially loaf have been found.

Karau and Williams’s meta-analysis found that social loafing was less prevalent among women than among men and less prevalent among people from East Asian, collectivist cultures (such as those in China, Japan, and Taiwan) than among people from Western, individualist cultures (such as those in Canada and the United States). With their tendencies to be more aware of their connections and mutual reliance on others, women and people from collectivist cultures may be relatively more con- cerned about the possible negative interpersonal impact of social loafing. If someone in the group does violate a norm of hard work, however, people from collectivist cul- tures may take more offense. That is the implication of research by Tsang-Kai Hung and others (2009), who found that workers in Taiwan who perceived co-workers as socially loafing were likely to be motivated to seek revenge toward their co-workers.

In an interesting twist, Ying-yi Hong and others (2008) hypothesized that there are times when people from collectivist cultures may be especially likely to socially loaf. Because people from collectivist cultures tend to be concerned with behav- ing consistently with group norms, they may be tempted to socially loaf if they are working in a group that has established a group norm of low productivity and effort.

Individuals often don’t try as hard in groups as they do alone. If they can be convinced that their efforts will pay off, however, their output can soar.

“We just haven’t been flapping them hard enough.”

Sa m

G ro

ss /

Th e

N ew

Y or

ke r C

ol le

ct io

n/ Th

e Ca

rt oo

n Ba

80220_ch08_ptg01_hr_308-349.indd 321 27/01/16 5:13 PM

322 Chapter 8 Group Processes

deindividuation The loss of a person’s sense of individuality and the reduction of normal constraints against deviant behavior.

The researchers found support for this idea in a set of studies with Chinese students. When these students engaged in a task with a group of co-workers who were not being productive, the participants reduced their own efforts if they thought their effort would be evident to their co-workers. Not wanting to publicly deviate from the group norm, these students conformed to the norm of working less hard.

j Deindividuation Being in the presence of others can lead to the relatively common effects of social facilitation and social loafing. But occasionally the presence of others can lead to extreme actions. A group of sports fans excited at their home team’s victory spills out of the arena onto the street, and soon a riot develops, complete with small fires, overturned cars, and looting. A conflict between two people late at night in a loud dance club suddenly turns to a raucous brawl, with bottles flying and punches thrown involving dozens of people. What turns an unruly crowd into a violent mob? No doubt there are many factors, including imitation of aggressive models, intense frustration, and alcohol consumption. These factors are discussed in Chapter 11 on Aggression. But there’s also deindividuation, the loss of a person’s sense of indi- viduality and the reduction of normal constraints against deviant behavior. Most investigators believe that deindividuation is a collective phenomenon that occurs primarily in the presence of others (Diener et al., 1976; Festinger et al., 1952).

Philip Zimbardo (1969) observed that arousal, anonymity, and reduced feelings of individual responsibility together contribute to deindividuation. These elements certainly seem to be in play in the destructive behavior that groups of sports fans often exhibit after a big win or loss. For example, immediately after the University of Connecticut’s men’s basketball team won the national championship tournament in April 2014, thousands of rowdy fans gathered on and around UConn’s campus, and the celebration took a destructive turn as some students set fires, flipped cars, and threw a light post through the window of a college building (Ward, 2014). The losing team that night was the University of Kentucky; when Kentucky won the champion- ship two years before, similar rioting and destruction broke out there. It’s important to note that the fans in these incidents were not frustrated and angry at a bitter loss;

A St

oc k/

Co lla

ge /C

or bi

“The sentiments and ideas of all the persons in the gathering take one and the same direction, and their conscious personality vanishes.”

—Gustave Le Bon (1895)

Here, Chinese farmers cooperate on a task in which individual contributions cannot be identified. Social loafing on such tasks occurs less often in Eastern cultures than in Western ones.

80220_ch08_ptg01_hr_308-349.indd 322 27/01/16 5:13 PM

323 Individuals in Groups: The Presence of Others

they were happy! So why would they—and the fans of many teams around the world after a big victory—act out destructively? All three elements that Zimbardo specified were present here: The fans were very aroused by their team’s victory, the thousands of celebrating fans provided the individuals with relative anonymity, and these fac- tors, quite possibly along with alcohol consumed during the game, contributed to reduced feelings of individual responsibility.

According to Steven Prentice-Dunn and Ronald Rogers (1982; 1983), two types of environmental cues—accountability cues and attentional cues—make deviant be- haviors such as this rioting more likely to occur. Accountability cues affect the indi- vidual’s cost–reward calculations. When accountability is low, those who commit deviant acts are less likely to be caught and punished, and people may deliberately choose to engage in gratifying but usually inhibited behaviors. Being in a large crowd or wearing a mask are two examples of instances when accountability may be low, and these factors are associated with more extreme and destructive behaviors.

Attentional cues focus a person’s attention away from the self. In this state, the individual attends less to internal standards of conduct, reacts more to the immediate situation, and is less sensitive to long-term consequences of behavior (Diener, 1980). Behavior slips out from the bonds of cognitive control, and people act on impulse. When you are at a party with very loud music and flashing lights, you may be swept up with the pulsating crowd and feel your individual identity slipping away.

One context ripe for deindividuation is online behavior. If you’ve ever been part of an online community where people can post comments anonymously (such as through the popular app, Yik Yak), there is a good chance that you’ve witnessed some of the nasty effects of deindividuation. Many well-intentioned sites or discussions or even the comments sections linked to online newspaper stories, videos, or celebrity gossip blogs soon devolve into a torrent of crude, hostile, and prejudiced venting and taunting that would never happen without the cloak of anonymity.

Trick or Treat: Field experiments on Halloween One particularly creative set of field experiments by Edward Diener and Arthur Beaman and their colleagues (Beaman et al., 1979; Diener et al., 1976) demonstrated how accountability and

University of Kentucky fans turn over a car as they celebrate Kentucky’s win over archrival Louisville in men’s basketball in March 2012. Two years later, when Kentucky lost to UConn in April 2014, similar scenes played out on UConn’s campus. In the midst of crowds like these, people may feel deindividuated, which can lead to deviant behavior.

A P

Im ag

es /C

hr is

tia n

Ra nd

ol ph

80220_ch08_ptg01_hr_308-349.indd 323 27/01/16 5:13 PM

324 Chapter 8 Group Processes

social identity model of deindividuation effects (SIDe) A model of group behavior that explains deindividuation effects as the result of a shift from personal identity to social identity.

attentional cues can affect behavior on a night when many otherwise well-behaved individuals act in antisocial ways: Halloween. When you think about it, Halloween is a perfect time to study deindividuation: Children often wear costumes with masks, travel in large groups at night, and are highly aroused. In one study, the researchers unobtrusively ob- served more than 1,300 children who came trick-or-treating to 27 homes spread around Seattle. At each of these homes, a researcher met and greeted the children, who were either alone or in groups. In one condition, the researcher asked the children their names and where they lived; in another condition, the researcher did not ask them any questions about their identi- ties. When asked to identify themselves, the children should have become more self-aware and more accountable for their actions. Children who were not asked to reveal their identities and were in a group should have felt relatively deindividuated, safe, and anonymous in their costumes.

The children were then invited to take one item from a bowl full of candy and were left alone with the bowl. Hid- den observers watched to see how many pieces of candy each child took. What did the observers see? As d Figure 8.4 illustrates, children who were in a group were more likely to break the rule and take extra candy than were children who were alone. Add anonymity to the presence of a group, and children were even more likely to do so. In other words, the children were most likely to take extra candy when they were the most deindividuated: when they were in a group and had not been asked to identify themselves. In another experi- ment, the researchers placed a mirror behind the candy bowl in some conditions. As noted in Chapter 3, the presence of a mirror tends to increase people’s self-awareness. Children who had been asked their names, especially older children, were much less likely to steal candy if there was a mirror present than if there wasn’t. Older children are more likely to have internal standards against stealing, and making these children self-aware made them more likely to act according to those standards.

Moving From Personal to Social Identity The loss of personal identity does not always produce antisocial behavior. In a study conducted by Robert Johnson and Leslie Downing (1979), female undergraduates donned garments resembling either robes associated with the Ku Klux Klan (a very racist hate group) or nurses’ uniforms. Half of the participants were individually identified throughout the study; the others were not. All of the participants were then given the opportunity to in- crease or decrease the intensity of electric shocks delivered to a supposed other participant (actually, a confederate in the experiment) who had previously behaved in an obnoxious manner. Participants wearing Ku Klux Klan costumes increased shock levels in both the identified and anonymous conditions. However, among those in nurse’s apparel, anonymous participants decreased shock intensity four times more frequently than did identified participants!

These findings support the social identity model of deindividuation effects (SIDe), which proposes that whether deindividuation affects people for better or

Alone

Not anonymous

% W

ho T

oo k

Ex tr

a C

an dy

Anonymous

Group

d F I G u R e 8 . 4

Trick or Treat(s)

Children trick-or-treating on Halloween were part of a field experiment. Some were alone and some were in groups with other trick-or-treaters. The experimenter who greeted them allowed a random half of the children to remain anonymous but asked the other half to indicate their names and where they lived. The experimenter instructed each child to take only one piece of candy, after which she left the children alone. The bars in this graph represent the percentage of children who took more than the one piece of candy. Consistent with predictions based on deindividuation, the children who were in a group and were anonymous were most likely to cheat by taking extra candy. Based on Diener et al., 1976.

80220_ch08_ptg01_hr_308-349.indd 324 27/01/16 5:13 PM

325 Group Performance: Problems and Solutions

process loss The reduction in group performance due to obstacles created by group processes, such as problems of coordination and motivation.

for worse reflects the characteristics and norms of the group immediately sur- rounding the individual as well as the group’s power to act according to these norms. As personal identity and internal controls are submerged, social identity emerges and conformity to the group increases. If a group defines itself in terms of prejudice and hatred against another group, for example, deindividuation can ignite an explosion of violence. If a group defines itself in terms of helping others, in contrast, deindividuation may promote prosocial, selfless actions. The conse- quences of losing your personal identity, therefore, depend on what you lose it to (Chen & Wu, 2015; Spears & Postmes, 2015).

Group Performance: Problems and Solutions Social facilitation, social loafing, and deindividuation can all affect individu- als whether they are working in real groups or are merely part of a collective or crowd. In this section, we examine processes that are specific to groups, where interaction among members is more direct and mean- ingful. We focus on how well groups perform, and in doing so, we ad- dress a fundamental question: Aren’t two (or more) heads better than one? Although eight people typically can outproduce a lone individual, do eight people working together in a group typically outperform the sum of eight people working individually? You may be surprised to learn how often and in what ways groups perform worse than their potential would suggest. We also discuss when and how groups are more likely to perform well.

j Losses and Gains in Groups When a group performs worse than its potential, it experiences what Ivan Steiner (1972) called process loss. Process loss refers to the re- duction of group productivity due to problems in the dynamics of a group. According to Steiner, some types of group tasks are more vul- nerable to process loss than others.

For instance, on an additive task, the group product is the sum of all the members’ contributions. Donating to a charity is an additive task, as is making noise at a pep rally. Of course, for these tasks, the more the merrier in terms of overall output. A group will make more noise at the rally than an individual. However, as we have seen, people often indulge in social loafing during additive tasks, which creates process loss. In other words, each member’s contribution may be less than it would be if that person worked alone. As a result the group performs less than its potential.

On a conjunctive task, the group product is determined by the individual with the poorest performance. Mountain-climbing teams or an offensive line trying to protect the quarterback on a pass play in football may be engaged in such a task; the “weakest link” will determine their success or failure. Because of this vulner- ability to the poor performance of a single group member, group performance on conjunctive tasks tends to be worse than the performance of a single average individual.

On a disjunctive task, the group product is (or can be) determined by the per- formance of the individual with the best performance. Trying to solve a problem or develop a strategy may be a disjunctive task: What the group needs is a single suc- cessful idea or answer, regardless of the number of failures. In principle, groups

Just as the strength of a chain depends on its weakest link, the group product of a conjunctive task is determined by the individual with the poorest performance. In mountain climbing, for example, if one person slips or falls, the whole team is endangered.

G or

do n

W ilt

si e/

G et

ty Im

ag es

80220_ch08_ptg01_hr_308-349.indd 325 27/01/16 5:13 PM

326 Chapter 8 Group Processes

process gain The increase in group performance so that the group outperforms the individuals who make up the group.

brainstorming A technique that attempts to increase the production of creative ideas by encouraging group members to speak freely without criticizing their own or others’ contributions.

have an edge on individuals in the performance of disjunctive tasks. The more people involved, the more likely it is that someone will make a breakthrough. In practice, however, group processes can interfere with coming up with ideas and getting them accepted, resulting in process loss. For example, groups may not real- ize which group members have the best ideas or are most expert. Unless the best solution for a particular problem is easily and clearly identifiable once it has been suggested, the group may fail to implement it; as a result, the group will perform worse than its best members (Soll & Larrick, 2009; Stasser et al., 1995; Straus et al., 2009). Have you ever had the experience of knowing you had the right idea but being unable to convince others in your group until it was too late? If so, then you have experienced firsthand the problem of process loss on a disjunctive task. A bit later in this chapter, we will discuss strategies that help groups recognize and utilize expertise.

On some kinds of tasks, groups can even show process gain, in which they outperform even the best members. Groups often perform better than the best individuals on tasks in which (1) the correct answer is clearly evident to everyone in the group once it is presented, and (2) the work on the task can be divided up so that various subgroups work on different aspects of the task (Carey & Laughlin, 2012). In the business world, process gain tends to be called synergy, and it is the ideal that business and organizational groups strive for (Mercier et al., 2015; Meslec & Curşeu, 2013; Schulz-Hardt & Mojzisch, 2012).

j Brainstorming During the 1950s, advertising executive Alex Osborn developed a technique called brainstorming that was designed to enhance the creativity and productivity of problem-solving groups. The ground rules for brainstorming call for a freewheel- ing, creative approach:

d Express all ideas that come to mind even if they sound crazy.

d The more ideas, the better.

d Don’t worry whether the ideas are good or bad, and don’t criticize anyone’s ideas; they can be evaluated later.

d All ideas belong to the group, so members should feel free to build on each other’s work.

Alex Osborn (1953) claimed that by using these procedures, groups could generate more and better ideas than could individuals working alone. The idea caught on. Brainstorming was soon a popular exercise in business, government, and education. In fact, it remains very popular today. But when the research caught up with the hype, it turned out that Osborn’s faith in the group process was unfounded. In fact, “nominal groups” (several individuals working alone) produce a greater number of better ideas than do real groups in which members interact with each other. Brainstorming can indeed be effective, but people brainstorm- ing individually produce more and higher-quality ideas than the same number of people brainstorming together. One meta-analysis concluded that brainstorm- ing groups are only about half as productive as an equal number of individuals working alone (Mullen et al., 1991). Rather than being inspired by each other and building on each other’s ideas, people brainstorming in a group underperform (Paulus et al., 2015; Stroebe et al., 2010).

The top half of Table 8.2 presents several explanations for why group brain- storming is ineffective. Recall our discussion of social loafing from earlier in the

“A committee should consist of three men, two of whom are absent.”

—Herbert Beerbohm Tree

80220_ch08_ptg01_hr_308-349.indd 326 27/01/16 5:13 PM

327 Group Performance: Problems and Solutions

chapter, for example. As you can see in Table 8.2, social loafing is one factor that contributes to process loss in group brainstorming. The first obstacle listed in the table, production blocking, seems to be especially problematic for groups.

Despite the research evidence, brainstorming is still a popular device in many organizations. People who participate in interactive brainstorming groups typi- cally think that it works wonderfully and evaluate their performance more favor- ably than do individuals in nominal groups. They also enjoy themselves more, and group brainstorming can increase group cohesiveness (Henningsen & Henningsen, 2013). But they are working under the illusion that group brainstorming is much more effective than it really is.

Fortunately, after the initial shock at how poor group brainstorming really is, researchers have demonstrated a number of strategies that improve productivity while also preserving the enjoyment that group brainstorming can produce. For example, alternating types of brainstorming sessions (such as by having members brainstorm alone and then together, or first together and then alone) has proven very effective. Other strategies that improve group performance include training people in effective brainstorming, giving the group a subset of categories to be- gin the brainstorming process, using a trained facilitator during brainstorming sessions, and giving groups more time (rather than rush them with a tight dead- line) (Baruah & Paulus, 2011; Henningsen & Henningsen, 2013; Oxley et al., 1996; Paulus et al., 2006, 2015).

Using technology to allow groups to engage in what is called electronic brain- storming can help groups brainstorm more effectively. By allowing all group mem- bers to enter ideas from their own laptops, phones, or other devices and have the ideas appear on a central screen, electronic brainstorming combines the freedom of working alone and not having to wait their turns with the stimulation of see- ing others’ ideas. The bottom half of Table 8.2 presents some of the factors that make this type of brainstorming effective. A meta-analysis of research on elec- tronic brainstorming reported encouraging results (DeRosa et al., 2007). Groups using electronic brainstorming tend to perform much better than other brain- storming groups, although the evidence is mixed concerning whether electronic

Small groups engaging in brainstorming. Despite its popularity, interactive group brainstorming suffers from several group dynamics problems. Fortunately, researchers have determined some ways to improve brainstorming.

Cy ru

s M

cC rim

m on

/ C

on tr

ib ut

or /G

et ty

Im ag

People brainstorming as a group come up with a greater number of better ideas than the same number of people working individually.

FALSe

80220_ch08_ptg01_hr_308-349.indd 327 27/01/16 5:13 PM

328 Chapter 8 Group Processes

group polarization The exaggeration of initial tendencies in the thinking of group members through group discussion.

brainstorming groups outperform nominal groups (Dornburg et al., 2009). One interesting finding is that when the group size gets relatively large, electronic brainstorming seems to be especially effective (Paulus et al., 2013).

j Group Polarization Imagine that you’re asked to recommend whether someone should behave in a risky or a cautious man- ner, such as whether a business should risk trying to expand or whether an employee in a stable but bor- ing job should quit and take a more creative job in a new but unproven tech company. Imagine after you process the relevant information and come to an ini- tial opinion on this, you and a group of other people who have also arrived at initial opinions come to- gether and discuss the case. What will be the result of the group discussion of the various opinions?

Common sense suggests two alternative pre- dictions. Perhaps the most reasonable prediction is that there will be a compromise as group members move toward the average of all the individuals’ attitudes. But common sense also suggests another prediction. Many people familiar with committees agree that forming a committee is a good way not to get something done. We think that as lone indi- viduals we are willing to take risks and implement new ideas, but we recognize that groups tend to be cautious and slow moving. Wary of leading the group toward a risky decision, people often become more cautious in their views as they discuss them with other group members. So another prediction here would be that the people in your group would become more cautious in what they advocate.

So which prediction is the correct one: movement toward the average attitude or movement toward caution? It turns out that neither commonsense notion is correct. Rather, group discussion tends to enhance or exaggerate the initial leanings of the group. Thus, if most group members initially lean toward a risky position on a partic- ular issue, the group members on average will move toward an even riskier position after the discussion. But if group members in general initially lean toward a cautious position, the group discussion leads to greater caution. This effect is called group polarization: the exaggeration through group discussion of initial tendencies in the thinking of group members (Moscovici & Zavalloni, 1969; Myers & Lamm, 1976).

Group polarization is not restricted to decisions involving risk versus caution. Any group decision can be influenced by group polarization, from serious deci- sions such as how to allocate scarce medical resources or how much to pay CEOs, to more mundane decisions such as what theme a sorority should use at its next party (Chandrashekaran et al., 1996; Furnham et al., 2000; Zhu, 2014).

What causes group polarization? According to persuasive arguments theory, the greater the number and persuasiveness of the arguments to which group members are exposed, the more extreme their attitudes become. If most group members favor a cautious decision, for example, most of the arguments discussed will favor

m T a b l e 8 . 2

brainstorming in Groups: Problems and Solutions

Factors That Reduce the effectiveness of Group Brainstorming

d Production blocking. Having to wait their turn to speak, people may forget their ideas, don’t generate additional ideas until they can speak, or simply lose interest.

d Free riding. As others contribute ideas, individuals see their own contributions as less needed or less likely to have much impact. They therefore try less hard and engage in social loafing.

d Evaluation apprehension. People may be hesitant to suggest unusual ideas for fear of looking foolish and being criticized.

d Performance matching. Group members work only as hard as they see others work. Once the other three factors above have reduced the performance of a brainstorming group, performance matching can help maintain this relatively inferior performance.

Why electronic Brainstorming Is effective

d Production blocking is reduced because members can type in ideas whenever they come to mind.

d Free riding is reduced by having the computer keep track of each member’s amount of input.

d Evaluation apprehension is reduced because group members contribute their ideas anonymously.

d Performance matching is reduced because group members spend less time focusing on the performance of others as they type in their own ideas. In addition, performance matching is less of a problem because the initial performance of groups that brainstorm electronically is likely to be high.

d Group members can benefit by seeing the ideas of others, which can inspire new ideas that they might not otherwise have considered.

80220_ch08_ptg01_hr_308-349.indd 328 27/01/16 5:13 PM

329 Group Performance: Problems and Solutions

caution, giving the members more and more reasons to think caution is the correct approach (Pavitt, 1994; Vinokur & Burnstein, 1974). Another set of explanations is based on how people compare themselves to fellow group members as well as to outgroup members. In the case of group discussions, as individuals learn that most of the other group members lean in one direction on some issue, they may adopt a more extreme attitude in this same direction. In other words, people who are members of a group that believes that X is good may be willing to state to the group that twice X is even better. By advocating for twice X, individuals can distinguish themselves in the group in a manner that is approved by the group. In addition, groups may become more extreme as a way to differentiate themselves from other groups (Lamm & Myers, 1978; McGarty et al., 1992; Suhay, 2015).

Each of these accounts spotlights particular processes, but taken together they all seem to contribute to the emergence of group polarization. Now that you know about group polarization, you should be able to see evidence of it often as you observe the groups around you. It seems that political groups today have become more and more polarized, moving to extremes rather than toward moderation and compromise. On a smaller level, observe how the culture of a team may evolve over the course of a season or follow the attitudes of a group as it prepares for a debate against another group and you’re likely to see group polarization develop through the processes we’ve just described.

j Groupthink The processes involved in group polarization may set the stage for an even greater and more dangerous bias in group decision making. This bias can be seen in ex- amples of very smart people collectively making very dumb decisions. For example, several days before the U.S. space shuttle Columbia disintegrated as it reentered the Earth’s atmosphere on its way home on February 1, 2003, a team of engineers at NASA reviewed a video of foam breaking off the shuttle during launch and hitting the area near the left wing. The group speculated about whether the impact could have damaged the heat-shielding tiles located there. One engineer, Rodney Rocha, made more than half a dozen requests of NASA managers to go outside the agency

Group members’ attitudes about a course of action usually become more moderate after group discussion.

FALSe

Scholars have identified groupthink as underlying poor group decision making in a number of contexts, including decisions involving war, the economy, and even proceeding with a climb of Mount Everest in dangerous conditions.

Ch ris

tia n

Ko be

r/ JA

I/ P

as sa

ge /C

or bi

80220_ch08_ptg01_hr_308-349.indd 329 27/01/16 5:13 PM

330 Chapter 8 Group Processes

groupthink A group decision- making style characterized by an excessive tendency among group members to seek concurrence.

d F I G u R e 8 . 5

Charting the Course of Groupthink

Irving Janis depicted groupthink as a kind of social disease, complete with antecedents and symptoms, that increases the chance of making a bad decision. Based on Janis, 1982.

• High cohesiveness • Group structure Homogeneous members Isolation Directive leadership Unsystematic procedures • Stressful situations

Antecedents

• Overestimation of the group • Close-mindedness • Increased pressures toward uniformity Mindguards and pressure on dissenters Self-censorship Illusion of unanimity • Defective decision making Incomplete survey of alternatives Incomplete survey of objectives Failure to examine risks of preferred choice Failure to reappraise initially rejected alternatives Poor information search Selective bias in processing information at hand Failure to work out contingency plans

Symptoms

High Probability of a Bad Decision

and seek images from spy satellite photos or powerful telescopes that could provide a better look at the possible damage to the Columbia while it was in space. These re- quests were ignored or rejected. One manager said that he refused to be a “Chicken Little.” The flight director e-mailed his rejection of the engineer’s request with these chilling words: “I consider it to be a dead issue” (Glanz & Schwartz, 2003). The engineer’s concerns about the tiles turned out to be justified, and all seven crew members on Columbia died in the ensuing tragedy.

Or consider one of the greatest fiascoes in U.S. history: the decision to invade Cuba in 1961. When John Kennedy became president of the United States, he assembled one of the most impressive, most highly educated groups of advisers in the history of American government. But this group devised a plan—inherited from the previous administration—to launch a small invasion of Cuba at the Bay of Pigs in order to spark a people’s revolt that would overthrow Fidel Castro’s government. Despite the intelli- gence of Kennedy and his group of advisers, their plan was hopelessly flawed. For example, once the invaders landed at the Bay of Pigs, they expected to be supported by anti- Castro forces camped in the mountains nearby. But had Kennedy and his advisers consulted a proper map, they might have noticed that the invaders were to land 80 miles away from these mountains and were separated from them by a huge swamp. This was just one of a number of baffling mistakes the planners made. Ultimately the invasion failed miserably. The invaders were quickly killed or captured, the world was outraged at the United States, and Cuba allied itself more closely with the Soviet Union—exactly the oppo- site of what Kennedy had intended. The United States was humiliated. After the fiasco, Kennedy himself wondered, “How could we have been so stupid?” (Janis, 1982).

According to Irving Janis (1982), the answer to this question lies in a particular kind of flawed group dy-

namic that he called groupthink, an excessive tendency to seek concurrence (that is, agreement or uniformity) among group members. Groupthink emerges when the need for agreement takes priority over the motivation to obtain accurate information and make appropriate decisions. d Figure 8.5 outlines the factors that contribute to groupthink, along with its symptoms and consequences.

Janis believed that three characteristics contribute to the development of groupthink:

1. Because highly cohesive groups are more likely to reject members with deviant opinions, Janis thought they would be more susceptible to groupthink.

2. Group structure is also important. Groups that are composed of people from sim- ilar backgrounds, isolated from other people, directed by a strong leader, and lacking in systematic procedures for making and reviewing decisions should be particularly likely to fall prey to groupthink.

3. Stressful situations can provoke groupthink. Under stress, urgency can overrule accuracy and the reassuring support of other group members becomes highly desirable.

80220_ch08_ptg01_hr_308-349.indd 330 27/01/16 5:13 PM

331 Group Performance: Problems and Solutions

The idea that high cohesiveness contributes to groupthink got a lot of atten- tion when Janis introduced the model because, as we indicated earlier, cohesive- ness is something that most people view as something that groups should strive for and value. However, even though high cohesiveness in a group might feel great and make for smooth and easy interactions among group members, it can foster emphasis on preserving each other’s feelings rather than an objective examina- tion of the facts or an honest discussion of differences of opinion. Such a concern with agreement over accuracy is a hallmark of groupthink. Working collectively, individuals may be focused more on mutually reinforcing each other’s judgments than arriving at the most accurate decisions.

In Janis’s formulation, groupthink is a kind of social disease, and infected groups display the behavioral symptoms indicated in the middle of d Figure 8.5. For ex- ample, NASA’s refusal to ask for outside help in obtaining images of the Columbia shuttle in space clearly illustrated the symptom of closed-mindedness. Another symp- tom, pressures toward uniformity, was evident in these examples. During the plan- ning of the Bay of Pigs invasion, the president’s brother, Robert Kennedy, served as a “mindguard” and warned dissenting members to keep quiet. An alarming example of this occurred during the debate about whether to go ahead with the launch of the space shuttle Challenger in 1986, 17 years before the Columbia disaster. When a vice president for engineering called for a delay in the launching of the Challenger because of fear that the unusually cold weather could cause the O-ring seals in the rocket boosters to fail, a manager pressured him to change his vote by telling him to “take off your engineer hat and put on your management hat.” Pressured to abandon one role and adopt the other, he changed his vote, and the tragic fate of the Chal- lenger and its crew may have been sealed at that moment, as an O-ring seal did fail and the shuttle exploded 73 seconds into the flight. In the end, this vice president’s decision helped foster an illusion of unanimity because top-level managers at NASA were unaware of all the dissent voiced by the engineers that morning.

More recently, scholars have proposed that groupthink was behind poor group decision making in a number of contexts, including in the decisions to go to war in Iraq, torture prisoners of war, and climb Mount Everest in dangerous conditions, as well as in the economic decisions that contributed to the global financial disas- ter in 2008 (Ben-Hur et al., 2012; Burnette et al., 2011; Houghton, 2008; Post, 2011).

When alternatives are not considered, the behavioral symptoms of groupthink can result in the defective decision making outlined in Figure 8.5.

Preventing Groupthink To guard against groupthink, Janis urged groups to make an active effort to process information more carefully and accurately. He rec- ommended that decision-making groups use the following strategies:

d To avoid isolation, groups should consult widely with outsiders.

d To reduce group pressures to conform, leaders should explicitly encourage criti- cism and not take a strong stand early in the group discussion.

d To establish a strong norm of critical review, subgroups should separately discuss the same issue, a member should be assigned to play devil’s advocate and ques- tion all decisions and ideas, and a “second chance” meeting should be held to reconsider the group decision before taking action.

President Kennedy appeared to arrive at similar conclusions after the Bay of Pigs disaster. In the following year, 1962, the United States and the Soviet Union were at the brink of war after U.S. military intelligence discovered that Soviet mis- siles in Cuba were aimed at the United States. During this crisis, Kennedy stayed

“A committee is a cul-de-sac down which ideas are lured and then quietly strangled.”

—Barnett Cocks

80220_ch08_ptg01_hr_308-349.indd 331 27/01/16 5:13 PM

332 Chapter 8 Group Processes

away from initial meetings about how to respond to the situation, he consulted with experts outside his inner circle of advisers, and he told his brother Robert to play “devil’s advocate” and challenge all ideas—in sharp contrast to Robert’s role of “mindguard” during the Bay of Pigs planning sessions. Unlike the Bay of Pigs invasion, the Cuban missile crisis ended exactly as Kennedy had hoped: The Soviet Union withdrew its missiles from Cuba, and a war was avoided.

Research on Groupthink: Myth or Reality? Groupthink is a rather distinctive theory in social psychology. On the one hand, its im- pact has been unusually broad: It is discussed in a variety of disciplines outside psychology including business, political science, and communi- cation, and it has spawned numerous workshops, websites, and train- ing videos. Yet on the other hand, the empirical support for the model is much less impressive than its fame would suggest. This may be partly due to the difficulty of experimentally testing such a broad set of vari- ables in high-pressure group settings. In addition, many researchers disagree with Janis about the specific conditions that make groups vul- nerable to groupthink (Baron, 2005; Hackman & Katz, 2010; Turner & Pratkanis, 1998). For example, group cohesiveness does not seem to be as much of a risk factor as Janis thought it was.

Some researchers do believe, however, that when multiple anteced- ents of groupthink are evident simultaneously, such as a strong and controlling leader and a great deal of stress, groups are indeed vulnerable to the kinds of faulty deci- sion making that Janis described. For now, the bottom line may be that even if one is skeptical about the details of Janis’s theory of groupthink, as many and perhaps most scholars are, it does seem clear that many of the factors that Janis specified as antecedents and symptoms of groupthink can contribute to very faulty group decision making, and groups would be wise to be on the lookout for these traps.

Perhaps most important, the steps that Janis advocated to avoid these traps should help most groups, regardless of the merits or problems with the theory of groupthink. Indeed, research has shown empirical support for the effectiveness of several strategies in curtailing groupthink tendencies. These include inserting some- one in the group to play the role of a “reminder” who is responsible for informing the group about the dangers of biased decision making, making individual group members believe that they will be held personally responsible for the outcome of their group’s decisions, increasing the diversity of group members, and creating a group norm that encourages critical thinking and discourages the search for concur- rence (Ben-Hur et al., 2012; Postmes et al., 2001; Schultz et al., 1995; t’Hart, 1998).

j Communicating Information and utilizing expertise One of the biggest flaws in how groups perform is that they often fail to use all the information or skills that group members have. In this section we will explore some of the dynamics that cause this problem, as well as some factors that can help groups better communicate and use important information.

Information Sharing and Biased Sampling Imagine that the chief execu- tives of a company meet to decide whether to launch a risky expansion of their business. All of them share much of the same information about the potential costs and benefits, and the weight of this evidence suggests the risk is worth it. But one of the executives is an expert on some technical aspects of the stock market and knows a reason why this move might have bad stock implications. Another

A controlling leader who discourages disagreement can promote groupthink, leading to bad decisions.

“On second thought, don’t correct me if I’m wrong.”

Ri ch

ar d

Cl in

e /T

he N

ew Y

or ke

r C ol

le ct

io n/

Th e

Ca rt

oo n

Ba nk

“Nor is the people’s judgment always true: The most may err as grossly as the few.”

—John Dryden

80220_ch08_ptg01_hr_308-349.indd 332 27/01/16 5:13 PM

333 Group Performance: Problems and Solutions

biased sampling The tendency for groups to spend more time discussing shared information (information already known by all or most group members) than unshared information (information known by only one or a few group members).

executive has special knowledge of one of their primary competitors and knows that the competitor might make a countermove that could be devastating. A third executive has a better sense than the other executives that the morale among the company’s personnel could be hurt by the expansion. If these unique pieces of infor- mation made it into the discussion, the weight of the evidence would tip the other way, toward not expanding the company yet. But too often group members fail to share their uniquely held information, and bad decisions result.

This example illustrates what Garold Stasser (1992; Stasser & Titus, 2003) termed biased sampling. Because of biased sampling, a group may fail to con- sider important information that is not common knowledge in the group. Inad- equately informed, the group may make a bad decision. There are several reasons why groups tend to exhibit this bias. For example, commonly shared information is likely to be socially validated by the group, and this in turn causes it to be more easily remembered and trusted. The validation also makes group members more confident in discussing and reiterating these pieces of information (Levine & Smith, 2013). The results of meta-analyses of hundreds of studies involving many thousands of people make clear how important sufficient information sharing is for group performance, and how biased the information sharing in groups tends to be (Lu et al., 2012; Mesmer-Magnus & DeChurch, 2009).

Sometimes biased sampling can have tragic consequences. The commission that investigated the 1986 explosion of the space shuttle Challenger concluded that inade- quate sharing of information contributed to the disaster. For example, some engineers had information indicating that it would be unsafe to launch the shuttle that morning because of the low temperature, but this information was not shared with everyone. The people who ultimately made the decision to launch therefore were not aware of all the information that was relevant for their decision. The commission concluded, “If the decision-makers had known all the facts, it is highly unlikely that they would have decided to launch” (Presidential Commission on the Space Shuttle Chal- lenger Accident, 1986, p. 82). Lessons learned after that tragedy appeared to have been forgotten 17 years later: Inadequate sharing of vital information also appeared to have contributed to the Columbia disaster in 2003.

Part of the problem in the NASA disasters was that the communication network, which defines who can speak with whom based on a group’s structure, made it difficult for information to be distributed to all of the decision makers. In many organizations, in- formation is passed up a chain of command through layers of middle management and only some of this information makes it all the way up to the executives who make the final deci- sions. This was the case at NASA. The engineers who were most familiar with the physical details of the shuttle could not communicate directly with the NASA of- ficials at the top of the chain of command. According to Rodney Rocha, who was one of the engineers who most emphatically warned about the possible damage to the Columbia, “Engineers were often told not to send messages much higher than their own rung in the ladder” (Glanz & Schwartz, 2003). Because warnings were suppressed as the flow of information moved up the chain of command, the peo- ple at the top did not know the extent of the concerns of those lower in the chain.

Fortunately, researchers have discovered several conditions under which bi- ased sampling is less likely to occur. For example, leaders who encourage a lot of group participation and a thorough exchange and critical discussion of ideas are

The U.S. space shuttle Challenger explodes shortly after its launch on January 28, 1986, killing all seven crew members. Inadequate sharing of information and a flawed communication network were among the group dynamics problems that contributed to NASA’s flawed, and ultimately fatal, decision to launch the shuttle that cold morning.

A P

Im ag

es /B

ru ce

W ea

ve r

80220_ch08_ptg01_hr_308-349.indd 333 27/01/16 5:13 PM

334 Chapter 8 Group Processes

more likely to elicit unshared (as well as shared) information during group discus- sions than are leaders who are more controlling or who encourage group members to find common ground (Larson et al., 1998; Lee et al., 2010; van Ginkel & van Knippenberg, 2008, 2012).

J. Lukas Thürmer and others (2015) found that information sharing and group decisions were much better when group members had to make a plan of when and how they would review alternatives before settling on their ultimate decision. In one of their experiments, for example, three-person groups had to make a series of deci- sions, such as which job applicant should be hired. All of the group members were given a set of identical information relevant for their decision, but each member also received information unique to them that was crucial for the overall decision. If only commonly shared information was discussed, one of the job candidates would seem the best. But if all of the information, including the unshared information, was discussed, then the best decision would be to hire a different candidate. This hidden- profile technique is commonly used in studies of group decision making.

All of the groups in this study were given financial incentives to reach the best decisions, all knew that each group member had a slightly different (but all true) set of information, and all were encouraged to consider alternatives before reaching their final decision. The only difference between conditions was that half of the groups were given a specific plan about how and when to consider alterna- tives: “When we finally take the decision sheet to note our preferred alternative, then we will go over the advantages of the non-preferred alternatives again.” Hav- ing a specific plan made a big difference. As can be seen in d Figure 8.6, groups

d F I G u R e 8 . 6

Improving Information Sharing

Small groups received sets of information to be used to make a series of group decisions. Some information was known to all group members, but each group member also had some unique information. If only commonly shared information was discussed, the group would make bad decisions. Groups that were given a specific plan about how and when to consider alternatives to the group decision discussed more of the unique information (left) and made better decisions (right) than did groups just given the more general goal of considering alternatives. Based on Thürmer et al., 2015

General Goal

D is

cu ss

io n

of U

ni qu

e In

fo rm

at io

Speci c Plan

Condition

0 General Goal

D ec

is io

n Q

ua lit

Speci c Plan

Condition

80220_ch08_ptg01_hr_308-349.indd 334 27/01/16 5:13 PM

335 Group Performance: Problems and Solutions

transactive memory A shared system for remembering information that enables multiple people to remember information together more efficiently than they could do so alone.

given this plan were more likely to discuss the unshared informa- tion and more likely to reach the correct decision.

Information Processing and Transactive Memory In gen- eral, groups are susceptible to the same information-processing biases as individuals, only more so. In reviewing the research on group information processing, Verlin Hinsz and others (1997) concluded, “If some bias, error, or tendency predisposes individuals to process information in a particular way, then groups exaggerate this tendency. However, if this bias, error, or tendency is unlikely among individu- als processing the information (e.g., less than half the sample), then groups are even less likely to process information in this fashion” (pp. 49–50). Vivid examples, for example, often affect individuals’ judgments more than do statistics covering a much larger sample of information. Hinsz and others (2008) have found that groups are even more susceptible to this bias than individuals are.

One key advantage of groups, though, is that they can divide a large body of information into smaller portions and delegate different members to remember these more manageable portions. Have you and a friend ever divided up a list of things to remember, such as which grocery items to remember to buy at the store or which sections of a textbook chapter each of you should be responsible for remembering before you study together? If so, you have tried to take advantage of a shared process known as transactive memory, which helps groups re- member more information more efficiently than individuals (Heavey & Simsek, 2015; Liao et al., 2015; Wegner et al., 1991). This is one of the great benefits of working in groups. But process loss can occur in this domain as well. Social loafing may occur, for example, when group members don’t do their share of the work while expecting others to pick up the slack. A particularly important problem is that groups may not distribute the tasks and roles among group members in a rational or efficient man- ner. They may, for example, fail to match individuals to tasks based on their skills, expertise, and preferences.

Groups that develop good transactive memory systems have enormous advan- tages over other groups (Heavey & Simsek, 2015; Huang et al., 2013; Seong et al., 2015). Effective transactive memory systems involve a few key elements. First, the group must develop a division of knowledge, and the group members must be able to communicate and remember this information in the group; everyone must know who knows what. Second, the group members must be able to trust each other’s specialized knowledge. And third, the group members need to coordinate their efforts so that they can work together on a task smoothly and efficiently. Julija Mell and others (2014) found that having a central group member who is the expert at the “metaknowledge” of being able to identify which group members know which information or skills can also be a big advantage.

j Goals and Plans in Groups You’ve probably worked in many groups for which the goal was simply to “do your best.” Despite the popularity of such goals, the research clearly shows that they are not as effective as specific goals. As Edwin Locke and Gary Latham (2002) concluded from their review of 35 years’ worth of studies, “When people are asked to do their best, they do not do so” (p. 706). People are indeed capable of better than their vaguely defined “best.” Groups, like individuals, tend to perform better on a task when they have specific, challenging, and reachable goals, particularly if the group members are committed to the goals and believe they have the ability to

A clear example of process loss.

"Thank you for neglecting your work long enough to listen to my thoughts on efficiency."

People and groups tend to do worse when they have “do your best” goals than when they have very specific, ambitious goals.

TRue

Co nd

e N

as te

80220_ch08_ptg01_hr_308-349.indd 335 27/01/16 5:13 PM

336 Chapter 8 Group Processes

achieve them. In addition to having challenging and spe- cific goals, groups are most likely to benefit when there are incentives in place for achieving these goals.

One benefit of groups over individuals is that group members can hold each other accountable and encour- age each other to keep trying to achieve a goal. For ex- ample, Andrew Prestwich and others (2012) recruited a sample of British individuals who were interested in par- ticipating in a study on increasing their physical activity. Some of the participants were instructed to set specific plans for how to achieve their goal of increasing physical activity, including how they would respond in particular situations (if we’re in situation X, then we’ll do Y), and some were not asked to make these plans. In addition, some of these participants worked alone, and others worked with a partner. In a follow-up six months later, results indicated that participants who both had these plans and worked with a partner significantly increased their physical activity and lost significantly more weight than if they had no specific plans or if they had these plans but did not have a partner (see d Figure 8.7).

If a group doesn’t make a good, specific plan, it can fail to utilize the expertise that various group members have. In an experiment by Anita Williams Woolley and others (2008), some four-person groups had two mem- bers who were particularly expert at some of the skills needed to solve a complex task, and other groups in- cluded no experts. If the groups weren’t instructed to

make a specific plan of how to proceed before they began working on the task, the groups with experts did not do better than the group without; in fact, the group with experts did a bit worse! Without the specific planning, the individuals’ ex- pertise was not used properly. The other groups were required to discuss—before

working on the task—who would be responsible for which type of evidence and plan how they would integrate the various types of evidence they were given. Only the groups that both had experts and went through this planning did well on the task.

d F I G u R e 8 . 7

Sticking to a Plan: exercise and Weight Loss

Participants wanting to exercise more and lose weight worked either alone or with another person, and were either instructed to set specific plans for how to achieve their goal (including how they would respond in particular situations) or were not instructed to do so. Six months later, the participants who both set a specific plan and worked with a partner lost significantly more weight than did the participants in any of the other conditions. Based on Prestwich et al., 2012

3.5

2.5

1.5

0.5

0 No Plan/

Alone Plan/Alone

W ei

gh t L

os t (

kg )

No Plan/ Partner

Plan/Partner

j Training and Technology We have discussed a variety of factors that can enhance group processes. Table 8.3 presents a set of the conditions that Ruth Wageman and her col- leagues (2009) suggest are best for team effective- ness. These suggestions are consistent with the principles that we have addressed throughout this chapter, and they help illustrate the tremendous value that understanding the social psychology of groups can have in the business world or wher- ever group performance is essential. Social psy- chologists often are hired to train businesses and

m T a b l e 8 . 3

Conditions for Team effectiveness

Ruth Wageman and her colleagues (2009) reviewed the research on what makes teams most effective. Below is a list of some of the conditions they emphasize.

d Teams should be interdependent for some common purpose and have some stability of membership.

d The team’s overall purpose should be challenging, clear, and consequential.

d Teams should be as small as possible and have clear norms that specify what behaviors are valued or are unacceptable.

d A reward system should provide positive consequences for excellent team performance.

d Technical assistance and training should be available to the team.

80220_ch08_ptg01_hr_308-349.indd 336 27/01/16 5:13 PM

337 Group Performance: Problems and Solutions

group support systems Specialized interactive computer programs that are used to guide group meetings, collaborative work, and decision-making processes.

organizations to use these kinds of principles, and a great deal of research supports the value of training in improving group performance (Driskell & Salas, 2014; Salas et al., 2007, 2012).

Some of the obstacles that get in the way of good group discussion and decision making can be re- duced through the use of interactive computer pro- grams. Often referred to as group support systems (or group decision support systems), these programs help remove communication barriers and provide structure and incentives for group discussions and decisions. Compared to groups that use more con- ventional face-to-face modes of discussion, groups that use these systems often do a better job of sam- pling information, communicating, avoiding group- think, and arriving at good decisions (Adla et al., 2011; Bergey & King, 2014; Bose, 2015; Lim & Guo, 2008; Tegarden et al., 2015). Table 8.4 lists some of the ways that computerized group support systems can help groups avoid groupthink.

j Virtual Teams Teams consisting of people dispersed widely across the globe are a relatively recent and rapidly growing part of the business world. Recent estimates indi- cate that a majority of professional workers today spend time working in virtual teams, and the numbers are expected to continue to grow (Gilson et al., 2015; Mesmer-Magnus et al., 2011). Virtual teams, sometimes called dispersed teams, are “groups of people who work interdependently with shared purpose across space, time, and organization boundaries using technology to communicate and collaborate” (Kirkman et al., 2002, p. 67). Due to globalization and a variety of related factors, virtual teams will be increasingly important in businesses and organizations.

By using virtual teams dispersed across time and space, businesses and or- ganizations can have much greater reach, diversity of perspectives and skills, and understanding of broader markets and groups of people than can traditional groups. But virtual groups can be especially vulnerable to some of the factors that harm traditional groups. Given the physical distances between members and how little interaction they may have with each other, virtual groups may have a harder time building cohesiveness, keeping membership stable, socializing new members, keeping roles clear, sharing information, and developing transactive memory systems that enable the members to recognize or recall who has what knowledge or expertise in the group. Special attention must be paid to virtual groups, therefore, to offset these problems. For example, directories should be available and updated to allow members to access information about who knows what across the virtual team. Frequent teleconferencing sessions and occasional short visits to allow dispersed group members to spend some time together can also help (Beyerlein et al., 2015; Hackman & Katz, 2010; Purvanova, 2014). The lack of cohesiveness in virtual groups can have some advantages, though, such as in helping these groups avoid some of the conformity pressures associated with groupthink-like problems.

Based on Miranda, 1994.

m T a b l e 8 . 4

How Computerized Group Support Systems Help Groups avoid Groupthink

1. Allow group members to raise their concerns anonymously through the computer interface, enabling them to risk challenging group consensus without fear of direct attacks.

2. Reduce the directive role of the leader.

3. Enable group members to provide input simultaneously, so they don’t have to wait for a chance to raise their ideas.

4. Allow the least assertive group members to state their ideas as easily as the most dominating.

5. Provide a systematic agenda of information gathering and decision making.

6. Keep the focus in the group meetings on the ideas themselves rather than on the people and relationships within the group.

80220_ch08_ptg01_hr_308-349.indd 337 27/01/16 5:13 PM

338 Chapter 8 Group Processes

Because groups have become increasingly diverse in many settings, it is more important than ever that groups learn how to utilize the great benefits and minimize the costs of diversity in group processes.

Jo n

Fe in

ge rs

h/ G

et ty

Im ag

j Culture and Diversity One outgrowth of the surging use of groups dispersed around the world is the creation of greater diversity among members of the groups. Even in face-to-face groups, though, groups are becoming increasingly diverse, most obviously in terms of gender, race, ethnicity, and cultural background.

How does diversity affect group performance? How can a group best use di- versity to its advantage? The answers to such questions—and even the meaning of diversity—are likely to change as society changes in terms of its demographics and attitudes. In addition, diversity is not restricted to demographic differences among group members but can also mean differences in attitudes, personalities, skill levels, and so on. Thus, the issues surrounding diversity are particularly complex.

The evidence from empirical research concerning the effects of diversity on group performance is decidedly mixed (Hackman & Katz, 2010; Paulus & van der Zee, 2015). On the one hand, diversity often is associated with negative group dynamics. Miscommunications and misunderstandings are more likely to arise among heterogeneous group members, causing frustration and resentment and damaging group performance by weakening coordination, morale, and commit- ment to the group. Cliques and conflicts often form in diverse groups, causing some group members to feel alienated.

On the other hand, research has also demonstrated positive effects of diver- sity, such as on patterns of socialization, creativity, and the complexity and in- clusiveness of group discussion (Antonio et al., 2004; Juvonen et al., 2006; Toosi, Sommers, & Ambady, 2012). Samuel Sommers (2006), for example, found that racially diverse juries exchanged a wider range of information, cited more facts about the case being decided, and made fewer errors in their deliberations than racially homogenous juries, at least when the defendant was black. Sommers and others (2008) also found that merely anticipating being in a racially mixed group made white individuals process information relevant to race more thoroughly.

As more organizations try to attract customers and investors from diverse cultures, diversity in personnel should offer more and more advantages. Ced- ric Herring (2009) analyzed data from over 1,000 work establishments in the United States from 1996 to 1997 and found that racial diversity was associated

with greater profits and market share. Sangeeta Badal and James Harter (2014) found similar results regarding gender diversity. On the other hand, Andrea Garnero and others (2014) found that whether diversity predicted greater profits varied depending on what types of diversity (e.g., age vs. gender) and industries were studied.

Although the overall picture remains mixed, more prom- ising is research focused on understanding specific factors that can help groups achieve the benefits of diversity while avoiding or reducing its problems (Hebl & Avery, 2013; Paulus & van der Zee, 2015). One set of recent findings is that multicultural groups perform better if their members or leaders have relatively high awareness of their own and others’ cultural assumptions—what is sometimes called cultural metacognition (Chua et al., 2012; Moon, 2013; Mor et al., 2013). People exhibit cultural metacognition to the extent that they often think about and check the accuracy of their cultural knowledge, particularly in cross-cultural interactions. William Maddux and others (2014) similarly

80220_ch08_ptg01_hr_308-349.indd 338 27/01/16 5:13 PM

339 Conflict: Cooperation and Competition Within and Between Groups

found that students in an international business school program who most ex- hibited “multicultural engagement”—that is, they adapted to and learned about new cultures—were the most successful students on the job market following the program. These researchers also found that students and professionals who had experienced diversity by living abroad were more creative in tasks and more successful in their careers when they identified with both their home and abroad cultures (Tadmor, Galinsky, & Maddux, 2012).

In addition, culturally diverse groups whose members have a positive attitude toward learning new information tend to benefit from the diversity of their groups more than other groups (Nederveen Pieterse et al., 2013). Finally, diverse groups whose members are encouraged or trained to take each other’s perspectives may outperform homogenous groups or diverse groups not given special instructions to promote perspective taking (Hoever et al., 2012; Lindsey et al., 2015).

j Collective Intelligence: Are Some Groups Smarter Than Others?

We have seen throughout this chapter that groups consisting of very intelligent people, or even people who are expert at a task very relevant to what the group is working on, are far from immune to making really dumb decisions or falling short of their potential. But are some groups—as a collective working together— particularly smart and effective at, well, group stuff? Anita Williams Woolley and others (2010) tried to answer this kind of question. They had 699 people work in small groups on a wide variety of group tasks. Across these various group tasks, some groups did tend to outperform the other groups. The researchers considered these groups that tended to do well on a wide variety of group tasks as being high in “collective intelligence.” What predicts which groups are like this? The average level of intelligence of the individual members of the group did not correlate much with the group’s collective intelligence, nor did the maximum level of intelligence of any particular group member. Three factors did, however, predict collective intelligence of the groups: (1) the average social sensitivity of the group members, (2) the tendency to allow the various group members to take turns participating in the discussion (rather than having one or a few people dominate), and (3) a higher proportion of women (who tended to be higher than the men in social sensitivity).

Many of the most crucial issues confronting our world today involve conflicts between groups or between individuals and their groups. The desire of some indi- viduals to consume valuable resources conflicts with the need to protect the envi- ronment for the greater good. A nation’s claim to important territory or the right to nuclear arms conflicts with another nation’s national security. In this section, we describe some of the dilemmas groups often must confront and what factors influence whether individuals and groups act cooperatively or competitively in dealing with them. We also look at an important mechanism for resolving group conflicts: negotiation.

Conflict: Cooperation and Competition Within and Between Groups

80220_ch08_ptg01_hr_308-349.indd 339 27/01/16 5:13 PM

340 Chapter 8 Group Processes

social dilemma A situation in which a self-interested choice by everyone will create the worst outcome for everyone.

j Mixed Motives and Social Dilemmas Imagine that you have to choose between cooperating with others in your group versus pursuing your own self-interests, which can hurt the others. Examples of these mixed-motive situations are everywhere. An actor in a play may be moti- vated to try to “steal” a scene, a basketball player may be inclined to hog the ball, an executive may want to keep more of the company’s profits, a family member may want to eat more than his fair share of the leftover birthday cake, and a citizen of the Earth may want to use more than her fair share of finite valuable resources.

In each case, the individual can gain something by pursuing his or her self- interests, but if everyone in the group pursues self-interests, all of the group mem- bers will ultimately be worse off than if they had cooperated with each other. Each option, therefore, has possible benefits as well as potential costs. When you are in a situation like this, you may feel torn between wanting to cooperate and wanting to compete, and these mixed motives create a difficult dilemma. The notion that the pursuit of self-interest can sometimes be self-destructive is the basis for what is called a social dilemma. In a social dilemma, what is good for one is bad for all. If everyone makes the most self-rewarding choice, everyone suffers the greatest loss. This section examines how people resolve the tension between their coopera- tive and competitive inclinations in social dilemmas.

The Prisoner’s Dilemma We begin with a crime story. Two partners in crime are picked up by the police for questioning. Although the police believe they have com- mitted a major offense, there is only enough evidence to convict them on a minor charge. In order to get a conviction for the more serious crime, the police will have to convince one of the criminals to testify against the other. Separated during question- ing, the criminals weigh their alternatives (see d Figure 8.8). If neither confesses,

d F I G u R e 8 . 8

The Prisoner’s Dilemma

Each of two criminals is offered immunity from prosecution in exchange for a confession. If both stay silent, both get off with a light sentence on a minor charge (upper left). If both confess, both receive a moderate sentence (lower right). But if one confesses while the other stays mum, the confessing criminal goes free and the silent one spends a long time in jail. © Cengage Learning®

Prisoner A

No Confession (Cooperates with Prisoner B)

Confession (Competes with Prisoner B)

B gets 1 year

A gets 1 year

B gets 10 years

A gets 0 years

B gets 0 years

A gets 10 years

B gets 5 years

A gets 5 years

Pr is

on er

No Confession (Cooperates

with Prisoner A)

Confession (Competes with

Prisoner A)

80220_ch08_ptg01_hr_308-349.indd 340 27/01/16 5:13 PM

341 Conflict: Cooperation and Competition Within and Between Groups

prisoner’s dilemma A type of dilemma in which one party must make either cooperative or competitive moves in relation to another party. The dilemma is typically designed so that the competitive move appears to be in one’s self- interest, but if both sides make this move, they both suffer more than if they had both cooperated.

resource dilemmas Social dilemmas involving how two or more people will share a limited resource.

they will both get light sentences on the minor charge. If both confess and plead guilty, they will both receive somewhat longer sentences. But if one confesses and the other stays silent, the confessing criminal will go free while the silent criminal will pay the maximum penalty.

This situation forms the basis for the research paradigm known as the pris- oner’s dilemma. In the two-person prisoner’s dilemma, participants are given a series of choices that give them the option of cooperating or competing with each other, but either option has potential costs. Imagine that you’re Prisoner A in d Figure 8.8. You should see that no matter what Prisoner B does, you’re better off if you compete with B and confess. (You’d get 0 years rather than 1 year if B cooperates with you by not confessing; you’d get 5 rather than 10 years if B also confesses.) But here’s the dilemma: If you both confess, each of you gets 5 years. If neither of you confesses, each of you gets only 1 year. In other words, each in- dividual is better off selling out his or her partner, but if both individuals do this, they are worse off than if neither did. It’s really a perplexing situation. What do you think you would do?

This kind of social dilemma is not limited to situations involving only two in- dividuals at a time. Imagine, for example, being in a burning building or a sinking ship. Each individual might want to race for the exit or the lifeboats as quickly as possible and push others out of the way, but if everyone does that, more people will die in the panic. More lives will be saved if people leave in an orderly fash- ion. Soldiers engaged in combat may be better off individually if they take no chances and duck for cover, but if their comrades do the same thing, they all will be slaughtered by the enemy. Nations face such dilemmas as well. Two countries locked in an arms race would be better off if they stopped spending money and resources on weapons of mass destruction, but neither country wants to risk fall- ing behind the other (Dawes, 1980).

Resource Dilemmas One particular category of social dilemmas that people face frequently and in vitally important contexts is called resource dilemmas, which con- cern how two or more people share a limited resource. As with prisoner’s dilemmas, attempts to gain a personal advantage will backfire if the other party also makes the competitive choice. Resource dilemmas come in two basic types: (1) commons dilemmas and (2) public goods dilemmas.

In the commons dilemma, if people take as much as they want of a limited resource that does not replenish itself, nothing will be left for anyone. One popu- lar example of this dilemma is known as the “tragedy of the commons” (Hardin, 1968). In earlier times, people would let their animals graze on the town’s lush grassy commons. But if all the animals grazed to their hearts’ content (and to their owners’ benefit), the commons would be stripped, the animals’ food sup- ply would be diminished, and the owners’ welfare would be threatened. Today, the tragedy of the commons is a clear danger on a global scale. Deforestation, air pollution, carbon emission, ocean dumping, massive irrigation, overfishing, com- mercial development of wilderness areas, overpopulation, and overconsumption all pit individual self-interest against the common good.

To teach his students about the tragedy of the commons, in June 2015 a social psychologist named Dylan Selterman gave his class a chance to earn extra credit points during their exam. Each student had to indicate whether they’d prefer 2 or 6 points of extra credit—the catch was that if more than 10% of the class chose 6 points, no one would get any extra credit. In the end, about 20% chose the 6 points, and therefore nobody earned any extra credit—and this is the typical result Selterman and others have found whenever they present this dilemma to their

Want to participate in an online prisoner’s dilemma game? Simply search online with the phrase “Play Prisoner’s Dilemma” and you should find several websites that offer a chance to do this.

80220_ch08_ptg01_hr_308-349.indd 341 27/01/16 5:13 PM

342 Chapter 8 Group Processes

students. It’s simply too tempting for too many individuals to try to maximize their own points, at the risk of the greater good. A student tweeted a picture of this exam question, and it went viral; as a result Selterman was soon on the national news explaining social dilemmas to the nation.

In public goods dilemmas, all of the individuals are supposed to contribute resources to a common pool. Examples of these public goods include the blood supply, public broadcasting, schools, libraries, roads, and parks. If no one gives, the service can’t continue, and all will suffer. Again, self-interest conflicts with the public good.

Responding to Social Dilemmas: Groups and Individuals Social dilem- mas pose serious threats to people’s quality of life and even to life itself. How do people try to solve them? What factors make them more or less cooperative when faced with these dilemmas?

Fear and greed are two critically important factors in determining reactions to these dilemmas—the fear of being exploited by others and the greedy desire to maximize one’s own outcomes. Trust, therefore, is essential in promoting cooperation because it reduces the fear of being exploited. Similarly, a sense of belongingness and identity with the greater group also promotes cooperation, in part because this perspective can reduce fear and greed (Klapwijk & Van Lange, 2009; McLeish & Oxoby, 2011; Swann et al., 2012). In an interesting twist on this idea, Carsten De Dreu (2012) found that individuals who tend to feel that they can’t depend on others were significantly more likely to cooperate on a social dilemma if they received a dose of oxytocin—a neurochemical that can increase people’s feeling of bonding with another person and reduce their concerns of betrayal.

Cultural and individual differences also play a role. Some research has found that collectivists tend to cooperate more when dealing with friends or ingroup members but compete more aggressively when dealing with strangers or outgroup members; individualists’ behavior does not vary as much as a function of who the other party is (De Dreu et al., 2007; Oyserman et al., 2002; Wong & Hong, 2005).

Individuals’ social value orientation also matters (Murphy & Ackermann, 2014). People with a prosocial, cooperative orientation seek to maximize joint gains or achieve equal outcomes. Those with an individualist orientation seek to maximize their own gain. And people with a competitive orientation seek to maximize their own gain relative to that of others.

These are just a few of the variables that help de- termine competition or cooperation in social dilemmas. Take a look at the factors listed in Table 8.5. Each of these has been shown to facilitate the best solutions