Assignment: Cognitive Aspects of Criminal Behavior

Social Cognition

Learning Objectives At the end of this chapter, you should be able to: • describe the field of cognitive psychology and explain social cognition;

• explain how low moral reasoning contributes to criminal behaviour;

• explain the various mechanisms of moral disengagement employed by people to avoid the negative feelings associated with violating their moral standards;

• identify and describe different thinking patterns that contribute to criminal behaviour;

• explain how affect and cognition can interact in ways that lead to violent

behaviour;

• explain how groups and authority figures influence individual decisions to en­ gage in criminal and violent behaviour;

• identify different social information processing errors and expla in how they contribute to criminal and violent behaviour; and

• describe the general aggression model (GAM).

The Psychology of Criminal and Violent Behaviour

Umar Farouk Abdulmutallab Detroit, Michigan-"Hey dude, your pants are on fire," ex­

claimed the passenger sitting

beside 25-year-old Umar Farouk

Abdulmutallab, who became

known as the Underwear Bomb­

er (Harris, 2011, para. 15). Notic­

ing the smoke and flames, fellow

passengers aboard Northwest

Airlines Flight 253, flying to De­

troit on Christmas Day 2009,

rushed to Abdulmutallab's aid

and set about extinguishing

the flames (Lewis, 2012). Once

the fire was doused and Abdul­

mutallab's pants were stripped

away, it became apparent that there was something odd about

his underwear. As one person

explained, "All I know is they

Islam at a young age through the

Internet and became convinced

that he should join the jihad,

or holy war, against the West.

While completing his masters'

degree in engineering, he broke

off communicat ion with his

family and made his way to Ye­

men, where he tried to contact

Al-Qaeda operatives. The group

was initially skeptical of him, but

his persistence convinced them

to train him. He was eventually

instructed to blow himself up in a plane flying over a densely

populated area of the United

Photo 8.1 This photo of Umar Farouk States so as to maximize the loss

Abdulmutallab was released by the of life.

US Marshals Service in December The agents who interviewed

2009, after Abdulmutallab's arrest. Abdulmutallab described him

AP Photo/ US Marshals Service · as "different because of his ab­

were bulky and they were burning" (Gardham, 2011 ,

p. BS). The bulkiness was from a bomb that, when it

failed to denotate, caught fire . The FBI later learned

Abdulmutallab believed that, if he blew up the

plane, he would become a martyr and go to heav­

en (Rosenfield, 2011). According to one agent who

interviewed him, " He did not bat an eye in saying if

that plane was going to go down, it was going to go

down . That was God's call. .. . His responsibility was

just to detonate the bomb, and whatever happened,

happened" (Lewis, 2012, para. 10).

The FBl's subsequent investigation of

Abdulmuttalab's background revealed that he came

from a wealthy family who cared deeply about him

(Lewis, 2012). Although he was raised in the Islam­

ic faith, his family did not share his fanati ca l beliefs

and were shocked to hear of his bombing attempt.

His mother reported that he was exposed to radical

solute devotion to Jihad. The

_ degree of conviction that he

had to the cause is unparalleled" (Lewis, 2012, para .

49). He reported wearing the underwear bomb for

the three weeks leading up to the incident to en­

sure that he was comfortable with the device and

could get it through airport security, something au­

thorities think may have contributed to the bomb's

failure to detonate (Ferran, 2012). During his vari­

ous court proceedings, Abd ulmutallab called the

Un ited States "a cancer" (Harris, 2011, para . 10) and

described the bomb he was carry ing as a "bless­

ed weapon to save the lives of innocent Muslims"

(White, 2012 , para . 21). He claimed to have acted

out of a religious duty and that attacks like the one

he launched were "the most virtuous of deeds"

(Rose nfield, 2011, para. 14). Even at the end of his

sente nci ng hearing, he was unrepe ntant, declar­

ing that he was "p roud to kill in the name of God"

(Ferran, 2012, para. 5).

8 Social Cognition

Introduction

Many of Abdulmutallab's statements reveal the deeply held beliefs and engrained think­ ing patterns that supported and justified his attempt to blow up Flight 253. The study of how we perceive, think, remember, and make decisions is called cognitive psychology. Within this field, a distinction is sometimes drawn between nonsocial and social cogni­ tion. Nonsocial cognition focuses on how information about the physical world, such as time, space, and action, is perceived, processed, stored, and utilized in decision-making (Ross & Fabiano, 1981). Social cognition is concerned with how we understand ourselves and other people (Fiske & Taylor, 1991). It involves our abilities to perceive what others are thinking and feeling, understand their perspectives, appreciate their perceptions of us, and solve interpersonal problems (Ross & Fabiano, 1981). This chapter examines social cognition and how it contributes to offending.

Moral Reasoning Morality is the ability to discern right and wrong. The dominant model of moral develop­ ment continues to be the one put forward by Lawrence Kohlberg halfa century ago. Kohlberg (1976) was interested in moral reasoning, the analytical or logical process that people use to make judgments about what is right and wrong. To investigate moral reasoning, he created a series of hypothetical moral dilemmas, the best-known of which involves Heinz, who stole the drug his wife desperately needed to survive (Kohlberg, 1969). Participants presented with this dilemma were asked if they agreed with Heinz's actions. Kohlberg was not really interested in what participants decided; he wanted to know why they approved or disap­ proved of stealing the drug and the logic they used to arrive at their judgment.

Based on this research, Kohlberg (1976) developed his theory of moral development. The resulting model identifies three major levels of moral reasoning, each with two stages (see Table 8.1). The first is the preconventional level, which encompasses the earliest forms of moral reasoning that are typical of young children, who assess right and wrong with reference to external circumstances such as avoiding punishment or gaining rewards from people with authority over them. You may recognize this type of reasoning if you have ever heard a child tell a friend that they should not do something because they might get in trouble. Individuals operating at the second, or conventional, level have internalized society's rules, which they see as important for maintaining order. They determine what is right and wrong according to the expectations of people who are important to them (e.g. family, close friends) or out of a sense of obligation toward society as a whole. The average adolescent and most adults exhibit this type of moral reasoning. The third and highest level is the postconventional. Individuals at this level do not accept society's rules uncritically and instead develop their own set of moral principles based on equality, dig­ nity, and a respect for human rights. They recognize that their principles will occasionally clash with society's rules and laws but believe they should stand by their own beliefs. Ac­ cording to Kohlberg, only a small minority of adults ever attain this level.

As a stage-based theory, Kohlberg's model makes several assumptions. First, it presup­ poses that individual moral reasoning operates only one stage at a time. Second, changes in moral development always occur in an upwards direction. Once you achieve a higher stage, you never regress to a lower one. Third, each stage is a prerequisite for the next one and

cognitive psychology The study of the mental processes involved in human percep­ tion, thought, memory, and decision-making.

nonsocial cognition The study of how we understand the physical world and other matters unrelated to people and their social interactions.

social cognition The study of how we understand our­ selves and other people.

morality The ability to distinguish between right and wrong.

moral reasoning The analyt­ ical process used to arrive at decisions about what is right and wrong.

preconventional moral reasoning The lowest of Lawrence Kohlberg's levels of moral development, whereby right and wrong are deter­ mined by the anticipated rewards or punishments.

conventional moral reasoning The middle of Lawrence Kohlberg's levels of moral development, whereby right and wrong are deter­ mined by the expectations of other significant people, such as family members, close friends, or society at large.

postconventional moral reasoning The highest of Lawrence Kohlberg's levels of moral development, whereby right and wrong are deter­ mined by an individual's own principles of equality, justice, and respect for human rights.

The Psychology of Criminal and Violent Behaviour

Table 8.1 Kohlberg's six hierarchical stages of moral development

Level Stage

3. Postconventional

2. Conventional

1. Preconventional

Stage 6: Universal ethical principle orientation Right is determined by one's conscience, according to principles such as justice, eq uality, and respect for human dignity.

Stage 5: Social contract, legalistic orientation Right is determined by society's rules, which should be upheld with some exceptions.

Stage 4 : "Law and order" orientation Right is determined by society's rules, which must be upheld for the system to work.

Stage 3: "Good boy-nice girl" orientation Right is determined by the expectations of those who are close or important to you.

Stage 2: Instrumental relativist orientation Right is determined by what will be rewarded.

Stage 1: Punishment and obedience orientation Right is determined by what will be punished.

A minority of adults

The average adolescent and adult

Children less than 9-11

Source: Adapted from Kohlberg, L. (1978). The cognitive developmental approach to behavior disorders: A study of the deve lopment of moral reasoning in delinquents. In G. Serban (Ed.), Cognitive defects in the devel­ opment of mental illness (pp. 207-219). New York, NY: Brunner/Maze!.

therefore movement through the stages always follows the same sequence and no stage is ever skipped. Although the sequence of development is the same for everyone, the rate of progress and the highest stage attained show individual variation. Jennings, Kilkenny, and Kohlberg (1983) speculate that interacting with others operating at higher levels of moral reasoning re­ veals shortcomings in our own judgments, which in turn spurs further moral growth. Thus, the speed of moral development and the stage at which development peaks are both highly dependent on a person's individual circumstances. People with extensive social interactions dominated by high moral reasoners are expected to develop more quickly and attain higher levels of moral reasoning than individuals with smaller and morally immature social circles.

Moral Development and Offending

Investigations of delinquent youths typically find that they are dominated by preconven­ tional moral reasoning, whereas nondelinquent groups exhibit a much higher proportion of conventional moral reasoning (see Table 8.2 for a summary of selected studies). A recent meta-analysis confirms that delinquents exhibit significantly lower moral reasoning than their nondelinquent counterparts (Starns et al., 2006). Moreover, this difference persists after controlling for other variables such as socio-economic status, cultural heritage, age, intelligence, and gender. Studies also show that morally immature offenders are much more likely to recidivate than morally mature ones (Van Vugt et al., 2011) and that treat­ ment programs designed to boost moral maturity can significantly reduce run-ins with the law (e.g. Wilson, Bouffard, & MacKenzie, 2005).

While Kohlberg's model of moral development was not originally expounded as a theo­ ry of crime, it has subsequently been elaborated on and used to explain offending behaviour.

8 Social Cognition

Table 8.2 Percentage of preconventional and conventional moral reasoning usage

among delinquent and nondelinquent samples in selected studies ·

Study Preconventional (%) Conventional (%)

Koh Iberg (1958a)

Delinquents 80 20 Controls 25 75

Hickey (1972)

Delinquents 53 47

Controls 7 93

Kohlberg (1958b)

Delinquents 92 8

Controls 58 42

Critchley (1961)

Delinquents 87 13

Controls 15 85

Hudgins & Prentice (1973) 1

Delinquents 80 20

Controls 30 70

Jurkovic & Prentice (1977)

Delinquents 71 29

Controls 42 58

Lee & Prentice (1988) 2

Delinquents 71 29

Controls 49 51

Basinger et al. (1995) 3

Delinquents 71 29

Controls 33 67

Palmer & Hollin (1998) 1.4

Delinquents 50 50

Controls 17 83

Trevethan & Walker (1989) 1

Delinquents 93 7

Controls 47 53

Notes 1. Figures reflect the percentage of youth operating at each level rather than the percentage of usage. 2. Figures reflect a subgroup of "psychopathic" delinquents. 3. Transitions between levels were grouped according to the lowest level. 4. Percentage is based on conversion of scores on the Sociomoral Reflection Measure-Short Form (Gibbs,

Basinger & Fuller, 1992).

Source: Republished with permission of Simon and Schuster, from Jennings, W. S., Kilkenny, R., & Kohlberg, L. (1983). Moral-development theory and practice for youthful and adult offenders. In W . S. Laufer & J.M. Day (Eds.), Personality theory, moral development, and criminal behavior (pp. 281-355). Lexington, MA: Lexington Books.

Kohlberg and colleagues do not propose that low moral reasoning is a causal factor in of­ fending, but they do believe that it can promote self-serving behaviour (Jennings et al., 1983). Remember, preconventional moral reasoners determine how they should act based primarily

The Psychology of Criminal and Violent Behaviour

on the probable consequences of the behaviour. This sensitivity to their immediate surround­ ings means that they are likely to offend when the chances of being caught are slim or the ex­ pected payoffs are high. This does not pose much of a problem with young children because their parents or other responsible adults are usually close at hand to punish or reward their

behaviour as appropriate. It presents more of a problem among morally immature adoles­ cents (Palmer, 2003). Parental supervision and control are loosened for adolescents, which means that they increasingly encounter opportunities where the perceived costs of offending are low (i.e. likelihood of getting caught) and the anticipated benefits are high (e.g. material rewards). Adolescents who are still operating at the preconventional level will be strongly influenced by these situational factors and thus likely to offend (Jennings et al., 1983).

High moral reasoners are better equipped to withstand opportunities to commit crime because they determine what is right and wrong based on their internal sense of social obligation and moral principles. This is why Kohlberg theorizes that high moral reasoning is essential for consistent and stable prosocial behaviour (Jennings et al., 1983). In this sense, preconventional moral reasoning does not cause offending, but conventional and postconventional moral reasoning help to prevent it. While it may be true that high moral reasoning increases our resistance to situational factors, it does not immunize us from committing crime. Table 8.2 shows that anywhere from 10 to 50 per cent of the moral reasoning in the delinquent groups was at the conventional level or higher. One explana­ tion for this apparent contradiction is that offending behaviour does not always involve moral judgments. Some offences flow from disturbed thinking processes associated with mental disorders (see Chapter 9) or from impulsive acts committed without time to en­ gage in moral reasoning (Jennings et al., 1983).

Moral development theory also highlights a major pitfall of criminal justice system policies that promote incarceration. Because institutions are dominated by low moral reasoners, prison offers few opportunities for moral growth through exposure to higher levels of moral reasoning (Jennings et al., 1983). While imprisonment does not preclude offenders from attaining conventional moral reasoning, their progress will probably be much slower than if they remained in the community and were surrounded by peers op­ erating at higher levels. The situation for conventional moral reasoners is even worse. Be­ cause they will encounter few, if any, inmates who are morally more mature, their moral development will stagnate. Incarcerating offenders, then, impedes moral development and leaves them vulnerable to situational factors that promote criminal behaviour when they are released back into the community.

Moral Disengagement At one time or another, most of us have done something we knew was wrong. Perhaps you stayed silent when a cashier undercharged you for an item or gave you too much change. Maybe you accidently damaged someone's parked vehicle but did not leave a note for the owner or you ran a red light despite knowing you could stop safely. Put bluntly, we sometimes do things we know are wrong. One of the criticisms levelled against Kohlberg's theory is that it does not consider how the cognitive processes surrounding morality relate to actual behaviour (Lytton, Maunula, & Watts, 1987). As we saw in Chapter 7, human behaviour is, at least initially, governed by the presence of external rewards and punish­ ments. Albert Bandura (1999) suggested that, over time, internal moral standards develop, a shift that resembles the change from preconventional to conventional moral reason­ ing. He argues that people apply consequences to themselves depending on the degree to

8 Social Cognition

which they live up to their internal moral standards. People feel good about themselves when they adhere to their morals, and this serves as motivation for behaving morally in the future. When they fall short of their moral expectations, they feel badly, which deters them from engaging in amoral conduct again. In this sense, our morals serve as an inter­ nal compass guiding our behaviour.

Bandura (1999) speculated that people can turn off this self-regulatory mechanism

through moral disengagement, which refers to a variety of processes that people em­ ploy to avoid the negative psychological consequences of violating their moral standards. Although this theory can be applied to minor incidents such as those described at the beginning of this section, Bandura's main purpose was to explain the senseless, cruel, and inhumane acts that are universally regarded as wrong yet occur with shocking regu­ larity around the world. Disengagement processes operate in four ways: to minimize the reprehensible nature of the behaviour, to minimize responsibility for the behaviour, to minimize the consequences of the behaviour, or to minimize the worth of the victims or shift blame on to them. As we shall see shortly, Umar Farouq Abdulmutallab employed many of these mechanisms.

Mechanisms of Moral Disengagement

Minimize the Reprehensible Nature of the Behaviour One group of moral disengagement processes work by recasting the perpetrator's be­ haviour in a more favourable moral light. This shift can be accomplished by euphemistic labelling, which makes the behaviour sound less offensive (Bandura, 1999). For example, Abdulmutallab called himself a martyr instead of a suicide bomber and described his acts a jihad rather than terrorism. Euphemistic labelling is not restricted to terrorists. Gov­ ernment and military officials routinely do the same thing. They refer to civilians who are wounded or killed during military operations as collateral damage and waterboarding has been called an enhanced coercive interrogation technique (Henley, 2007). A second cog­ nitive strategy is moral justification, whereby otherwise objectionable behaviour is con­ strued as acceptable because of the important goal it is designed to achieve-something Abdulmutallab did when he justified blowing up the plane as a way to save Muslim lives (Bandura, 1999). Finally, people sometimes engage in advantageous comparisons to make their conduct appear less offensive and to demonstrate their moral superiority by pointing out that what they have done is, at least in their eyes, less egregious than other people's acts.

Minimize Responsibility for the Behaviour Another group of disengagement mechanisms operate by diminishing the perpetrator's per­ ceived responsibility for the objectionable conduct. This is achieved by displacing or diffusing responsibility (see Figure 8.1). Displacement of responsibility occurs when people attri­ bute responsibility for their immoral acts to people in positions of control and authority. In Abdulmutallab's case, this authority was God. Abdulmutallab thought that his task was merely to detonate the bomb and that it was up to God to determine if the plane would actually blow up. Psychologically, displacement relieves offenders of personal responsibility by attributing their immoral acts to a higher authority. Stanley Milgram's research on obedi­ ence, discussed later in this chapter, attests to the powerful effect of displacement on human aggression. Diffusion ofresponsibility is similar, except that the person spreads the respon­ sibility across a group of people who are jointly culpable for the harm. As Bandura (1999) explains, "When everyone is responsible, no one really feels responsible" (p. 198).

moral disengagement

A process, described by Albert Bandura, in which

people employ a variety of psychological mechanisms to avoid the negative feelings they might otherwise expe­ rience as a result of violating their moral standards.

euphemistic labelling A moral disengagement mechanism whereby people apply a positive description to their immoral behaviour to make it sound less offensive.

moral justification A moral disengagement mechanism whereby people rationalize their immoral behaviour on the grounds that it was neces­

sary to achieve an important goal or avoid a more serious, harmful consequence.

advantageous comparisons A moral disengagement mechanism whereby people construe their immoral behaviour as Jess offensive by judging it against another's more egregious acts.

displacement of responsibility A moral disengagement mechanism whereby people avoid re­ sponsibility for their immoral acts by attributing them to people in positions of control and authority.

diffusion of responsibility A moral disengagement mechanism whereby people minimize their responsibil­ ity for immoral acts on the grounds that they are merely one among many involved.

The Psychology of Criminal and Violent Behaviour

dehumanization A moral disengagement mechanism that loosens the perpetrators' connection to their victims by viewing them as devoid of human qualities or as objects deserving mistreatment.

attribution of blame A moral disengagement mechanism whereby the vic­ tims are blamed for provoking the perpetrators and leaving them with no alternative but to respond as they did.

Moral justification Minimizing, ignoring , Dehumanization Advantageous comparison or misconstruing Attribution

Euphemistic labelling the consequences of blame

I I I Reprehensible Detrimental Victim

conduct effects

Displacement of responsibility Diffusion of responsibility

Figure 8.1 Mechanisms of Moral Disengagement. Source: A. Bandura, Moral disengagement in the perpetration of inhumanities (Personality and Social Psychology Review 3(3)) pp. 193-209, copyright© 1999 by SAGE. Reprinted by Permission of SAGE Publications, Inc.

Minimize the Consequences ofthe Behaviour Disengagement mechanisms also minimize the harmful effects of the offender's behaviour (Bandura, 1999). For example, an offender might lessen the perceived harm of stealing by explaining that the homeowners can buy new products with the insurance money while he or she is stuck with their old junk. Research shows that, the more removed perpetrators are from their victims and the consequences, the easier it is for them to ignore or mini­ mize the harm they cause (Milgram, 1965).

Minimize the Worth of the Victim and Blame the Victim The remaining disengagement processes loosen any connection the offender feels toward the victim by characterizing him or her as different or deserving of mistreatment. One such process is dehumanization, which equates victims to monsters, subhuman creatures, or even inanimate objects (Bandura, 1999). Recall that Abdulmutallab referred to the Unit­ ed States as cancer, as though it were a disease to be eradicated. In a similar vein, serial killer Gary Ridgway referred to his victims as "garbage" (Doughton, 2003, para. 4). Psy­ chologically, it is easier to hurt something inhuman rather than human. The attribution of blame focuses on the victim's conduct, which is seen as an extreme provocation that left the offender no choice but to respond as he or she did.

Individual Development and Operation of Moral Disengagement

Most people endorse similar moral standards. Where they tend to vary is in how bad­ ly they feel about violating their moral standards and their reliance on disengagement processes. Even the same person can employ different disengagement mechanisms from one situation to the next. Bandura (1999) believes that this is why people who exhibit the same level of moral reasoning do not always exhibit the same behaviour. Furthermore, there is evidence that the combined effect of several disengagement processes is greater than any individual mechanism (e.g. Bandura, Underwood, & Fromson, 1975). A series of studies conducted by Bandura and colleagues (1975) found that participants exposed to both diffusion of responsibility and dehumanization of the victim applied higher-intensity shocks to the victim than did participants exposed to only one of these conditions (see the "Researching" box). Given the enormity of the violence attempted by Abdulmutallab, it is not surprising to see numerous disengagement processes.

_o--­

8 Social Cognition

Researching Criminal and Violent Behaviour

Moral Disengagement in Action

Bandura and colleagues (1975) investigated

the impact of dehumanization and diffusion of

responsibility on aggression in a sample of col­

lege students. Participants were told that the

study was examining the effect of punishment

on decision-making and that they would be su­

pervising a three-member team situated in an­

other room. The team had to complete a series

of tasks. If any of the proposed solutions were

unsatisfactory, the supervisor was to select and

administer a mild to painful shock to the team. In

reality, there was no team and therefore no one

was shocked; however, the researchers used the

shock level selected to measure the participants'

aggression.

The study included several conditions. The

researchers told participants in the personal re­

sponsibility condition that they each supervised

a single team member and that any shock they

administered applied to that person alone. Par­

ticipants in the diffusion of responsibility con­

dition were told that they were one of three

supervisors and their shocks would be averaged

before being administered to the whole team.

The researchers also varied the humanization

and dehumanization of the team members by

making comments about them in front of the

participants. In the humanization condition, they

were described as understanding, perceptive,

and having other human qualities; in the dehu­

manization condition, they were characterized as

rotten and animalistic. No comments were made

in the neutral condition.

Figures 8.2 and 8.3 show the results of the

study. Participants in the diffuse responsibility

condition applied significantly greater shocks than

those in the personal responsibility condition.

They also applied much stronger shocks when

the team members were dehumanized than when

they were humanized or not described .

6

5

p ·u; c 4 ~ c

..:.:. u 0

..c (/) 3 c Q) "' 2

2

Figure 8.2 Average shock intensity applied when operat­ ing under conditions of diffused responsibility or individual

responsibility.

8

7

6 p ·u; c 5 ~ c

_,,,, u 4 0

..c (/)

c 3 Q) "' 2

2

Figure 8.3 Average shock intensity applied to persons represented as humanized, neutral, or dehumanized. Source: Reprinted from Journal of Research in Personality, 9(4), Albert Bandura, Bill Underwood, and Michael E Fromson, "Disinhibition of aggres­ sion through diffusion of responsibility and dehumanization of victims," 253-269, Copyright (1975), with permission from Elsevier.

..o-- -0- ­

-

o- - <> Individua l responsibi lity

2 3 4 5 6 7 8 9 10 Trials

r·······························...............................

..... /,./

___ .a.. ... - . ~ ... o----0 ... _ -- 0----o'

• ········•

,o,, ,,o ,,'' ....... 0 ......

_o'

2 3 4 5 6 7 8 9 Trials

- Dehumanized •····• Neutra l 0---0 Humanized

0, ,' ........... 0

I I

I I

I

- -0-- --o- ---0

Group responsibility

10

EJ The Psychology of Criminal and Violent Behaviour

criminal thinking Cognitive processes and content that facilitate the initiation and

continuation of offending behaviour.

Criminal Thinking

Several theories focusing on criminal thinking have emerged over the past several de­ cades. Criminal thinking refers to cognitive processes and content that facilitate the ini­ tiation and maintenance of offending behaviour (Walters, 2006). The underlying premise of these theories is that our thinking determines our behaviour. By logical extension, the only way to understand criminal behaviour is to gain an appreciation of how offenders think (Walters, 1990).

Yochelson and Samenow's Criminal Personality

Samuel Yochelson and Stanton Samenow's The Criminal Personality (1976) summarized nearly 15 years of their clinical work at a secure psychiatric hospital in Washington, DC. During this time, they conducted countless interviews with 240 offenders-most of whom were male-found not guilty by reason of insanity. Based on these interviews, Yochelson and Samenow concluded that offenders think differently than other people. They de­ scribed 52 common criminal thinking errors and behavioural patterns that permeated all aspects of the offenders' cognitions, including what they thought before, during, and after committing a crime. The researchers claimed that a criminal act, like anything else a person might do, is simply the result of the offender's thought processes.

Yochelson and Samenow's work represented an important step forward because it highlighted cognition's role in criminal behaviour; however, their research has not es­ caped criticism. One difficulty is that they studied a fairly unusual group of offenders, raising questions over whether the thinking errors identified might be unique to that sam­ ple (Ross & Fabiano, 1981). In addition, the absence of a control group makes it impossible to assess the extent to which these same thought processes might exist among the general public. Critics also argued that many of the thinking errors were poorly operationalized and therefore difficult to validate scientifically (Walters, 2006). These weaknesses aside, Yochelson and Samenow's clinical descriptions formed the foundation for subsequent work in this area.

Criminal Thinking Styles

Combining Yochelson and Samenow's research with their own clinical experience at the federal penitentiary in Leavenworth, Kansas, Glenn Walters and Thomas White (1989) distilled eight cognitive errors that they believed were associated with criminality. They labelled these cognitive errors, or criminal thinking styles, mollification, cutoff, enti­ tlement, power orientation, sentimentality, superoptimism, cognitive indolence, and discontinuity (see Table 8.3). Walters (1990) argues that the presence of these thinking styles perpetuates the criminal lifestyle. To be clear, he is not trying to explain isolated law-breaking but the persistent and serious offending of people firmly entrenched in a criminal lifestyle.

The central tenet of Walters's (1990) theory is that the choices and early life experienc­ es of people who engage in lifestyle criminality produce a belief system that consists of the eight criminal thinking styles. Once established, this belief system promotes, justifies, and rationalizes continued criminal conduct. Walters (1990) acknowledges that surrounding social conditions such as family upbringing and neighbourhood environment impact

8 Social Cognition

Table 8.3 Criminal thinking styles

1. Mollification The process of rationalizing criminal behaviour and avoiding responsibility by blaming it on other external factors or perceived injustices.

2 . Cutoff A psychological mechanism for ignoring the fear and anxiety associated with offending that might otherwise serve as a deterrent. A cutoff can be internal, such as a stated phrase (e .g . "Nothing risked, nothing gained"), or external, such as the use of an intoxicating substance prior to committing an offence.

3. Entitlement The belief that the usual rules do not apply and that one has the right to do or take anything.

4. Power orientation The view that the world can be divided into people who exert power and control over others and those who are powerless and submissive.

5. Sentimentality An emphasis on one's positive attributes and actions as a means of overshadowing and justifying the negative and harmful things one has done.

6. Superoptimism An overly confident and optimistic view of one's abilities, especially the ability to avoid the undesirable consequences of one's law-breaking.

7. Cognitive indolence Lazy and uncritical thought processes and the tendency to be easily bored.

8. Discontinuity A lack of commitment and general failure to follow through on things one starts.

Sources: Adapted from Walters, G. D., & White, T. W. (1989). The thinking criminal: A cognitive model of lifestyle criminality. Criminal Justice Research Bulletin, 4, 1-10; Walters, G.D. (1990). The criminal lifestyle: Patterns of serious criminal conduct. Newbury Park, CA: Sage Publications.

people's opportunities and likelihood of criminal behaviour, but he strongly rejects the general sociological perspective that some people are essentially compelled into a life of crime by their social situation. According to Walters, the choices people make ultimately set them on a path toward or away from a criminal lifestyle.

To facilitate criminal lifestyle theory research, Walters (1995) created the Psychologi­ cal Inventory of Criminal Thinking Styles (PICTS), a self-report instrument that measures all eight styles. Studies show that the presence of criminal thinking, as measured by the PICTS, is associated with criminal reoffending among men and women (Walters, 2002, 2012; Walters & Elliot, 1999). In one study, women prisoners scored significantly higher on the PICTS than their male counterparts (Walters, Elliot, & Miscoll, 1998). Walters and colleagues (1998) speculate that women may be more open and candid in their responding and, as a result, receive higher scores. Alternatively, criminal behaviour is generally seen as less socially acceptable for women than for men; therefore, the women who engage in this lifestyle and end up incarcerated may have more severe procriminal thinking pat­ terns. On a more positive note, there is some evidence that cognitive-based treatments can reduce criminal thinking styles, but more research is needed to show that these changes can lower reoffending (Walters, 2002; Walters, 2012).

Hostile Thoughts and Affect The discussion so far has focused on the role of cognition in offending behaviour, largely to the exclusion of affect or emotion. Yet it would be nai:ve to think that cognition and affect do not influence one another. We know that our feelings shape our thinking. For ex­ ample, we tend to express more negative and pessimistic views when we are sad or dejected than when we are happy. Likewise, what people think impacts how they feel. Receiving a compliment on our looks evokes quite different feelings depending on whether we think

frustration In the context of the frustration-aggression hy­ pothesis, the state produced when a person is blocked from attaining an expected goal .

The Psychology of Criminal and Violent Behaviour

the person is flirting with us or about to ask us for money. The excitation transfer theory and the frustration-aggression hypothesis highlight different ways that cognition and af­ fect may interact to produce aggressive and violent behaviour.

Excitation Transfer Theory

Dolf Zillman's excitation transfer theory is based on the idea that emotional arousal can carry over from one situation to another (Zillman, Katcher, & Milavsky, 1972). The theo­

ry assumes that the physiological arousal of people who are excited dissipates gradually. Eventually, the excitement subsides to a level where people are no longer aware of their arousal despite its lingering presence. In this state, they tend to mistakenly attribute any remaining physiological arousal to their current environment. This remaining arousal could also drive their behaviour. Imagine that someone walking across a narrow cable bridge suspended high above a canyon encounters a good-looking stranger. In this con­ text, the arousal generated by the thrill (or fear) of walking on the bridge might be misla­ belled as an intense attraction to the stranger and could even fuel efforts to contact him or her later on (Dutton & Aron, 1974).

In a carefully controlled investigation of this effect, Zillman and colleagues (1972) showed that participants who were angered and then physiologically aroused by riding an exercise bike administered significantly more intense electric shocks to an antagonist than did participants who were not angered or physiologically aroused by exercise. Pre­ sumably, the arousal was misidentified as anger and energized their aggressive behaviour. To borrow Donald Hebb 's (1955) analogy, physiological arousal is like the engine that propels us, but it is cognition and our ability to recognize and process information from our surroundings that serves as the steering wheel that guides the way we label our emo­ tions and behave. Although the excitation transfer effect has not been consistently demon­ strated in research (e.g. Bornewasser & Mummendey, 1982), it may help us understand incidents such as the one involving French international soccer star Zinedine Zidane, who headbutted Marco Materazzi in the chest after the Italian defender allegedly insulted him during the 2006 World Cup final ('Tm sorry but no regrets," 2006). The theory suggests that insults or provocations are more likely to evoke intensely angry and even violent re­ sponses during highly arousing events such as close sports matches.

Berkowitz's Cognitive Neoassociation Model

Frustration-Aggression Hypothesis The frustration-aggression hypothesis is the foundation for Berkowitz's (1989, 1990) cog­ nitive neoassociation model of aggression. Originally articulated by John Dollard and col­ leagues at Yale University in 1939, the hypothesis consists of two basic propositions: every aggressive act is proceeded by a frustration and every frustration produces aggression (Dollard, Doob, Miller, Mowrer, & Sears, 1939). Their definition of frustration is steeped in the language of behaviourism that prevailed at the time, but Berkowitz (1989), using more contemporary language, describes it as being unexpectedly blocked from attaining an anticipated goal. The Dollard group noted that frustration does not always result in overt acts of physical violence but can lead to verbal threats or no visible signs of aggres­ sion. Nevertheless, they were firm in the belief that every aggressive act can be traced to an underlying frustration .

8 Social Cognition

The frustration-aggression hypothesis has been attacked on a number of grounds over the years, but its greatest weakness is this claim of universality (Berkowitz, 1978). For in­ stance, contract killers commit violence to fulfill a goal, not as a response to a blocked goal. As this example illustrates, the hypothesis has little application to instrumental violence.

The Role of Negative Affect

Berkowitz (1989) recognized that a body of research demonstrates that people are more likely to exhibit aggression when faced with a host of unpleasant experiences-such as personal insults, pain, foul odours-that do not constitute frustrations according to the theory (Berkowitz, Cochran, & Embree, 1981; Geen, 1968; Rotton, Frey, Barry, Milli­ gan, & Fitzpatrick, 1979). To accommodate these findings, he suggested that all aversive events, not just those that are frustrating , are capable of sparking aggression . According to Berkowitz (1989, 1978), an aversive event generates negative affect, or feelings of unpleas­ antness and discomfort. Not every aversive event is equally unpleasant, but in some cases the negative affect generated is sufficiently intense that it evokes an aggressive response. For example, a goal that is deliberately and unfairly thwarted by someone has an added unpleasantness, which explains why this sort of frustration is more likely to result in vio­ lence than goals frustrated accidently or for legitimate reasons (Berkowitz, 1978).

The Role of Cognition The cognitive neoassociation model of aggression-Berkowitz's (1989, 1990, 1993, 2003) most recent theoretical work-expands on what happens after negative affect is generated, including the influence of cognition. The central role played by negative affect means the model is best suited to explaining reactive violence and, like the frustration-aggression hypothesis, it offers little insight into instrumental acts of violence.

The cognitive neoassociation model incorporates two distinct phases, each governed by different processes. The initial phase is dominated by associative processes that oc­ cur in a largely automatic fashion and with little cognitive processing. The negative af­ fect automatically and simultaneously activates the flight-or-flight response, discussed in Chapter 3 (Berkowitz, 1989). The theory assumes that this response involves networks of associated feelings , thoughts , physiological responses, and expressive motor reactions such as facial expressions and other body language. Activating one part of a network leads to a spreading activation of other parts. When the fight response tendency is activated, an array of interconnected ideas (e.g. hostile thoughts), feelings (arousal), physical expres­ sions (clenched teeth, fists) associated with aggression are stimulated and are collectively experienced as a rudimentary form of anger. In the flight response tendency, the activation of this associative network is experienced as fear. The response tendency that dominates on any given occasion primarily hinges on the person (e.g. biological influences, previous learning) and situational factors (Berkowitz, 1989).

As we are interested in antisocial and violent behaviour, we will assume that the aggressive response prevails. The fight associative network includes links to the motor actions or behaviours relating to violence as well as the inclination to attack and injure. When it is strongly stimulated by unpleasant events, people feel a powerful urge to lash out violently and hurt someone (Berkowitz, 1993, 2003). This phase unfolds automati­ cally and without cognitive input, which explains the impulsive acts that characterize reactive violence.

aversive events Incidents that produce pain, frustration , and unpleasant feelings.

negative affect A feeling of discomfort or unpleasantness .

spreading activation A process that occurs when activation of one part of an in­ t e rconnected neural ne twork leads to the activation of the other parts.

The Psychology of Criminal and Violent Behaviour

The mere fact that the fight associative network is activated more strongly than the

one for flight does not mean that physical acts of violence are inevitable. Situational factors

are very influential-the threat of certain punishment usually stops people from acting

violently unless the instigation to attack is extremely powerful (Berkowitz, 1993). On the

other hand, the presence of an obvious target for the violence, especially one that is the source of the aversive event, facilitates aggression. Berkowitz (1969) believes that aggres­

sive cues in the environment (e.g. weapons, violent media images) also increase the chance of violence occurring (see the "Researching" box).

Researching Criminal and Violent Behaviour

The Influence of Aggressive Cues on Behaviour

Berkowitz and LePage (1967) conducted a now classic study on how aggressive cues in the environment in­

fluence behaviour. The participants were male univer­ sity students who believed that they were partaking in a study on physiological responses to stress. The students were paired up and put in separate rooms. They were told that they would take turns working on an assignment and giving feedback to their partner. Depending on the quality of their work, they would re­

ceive from 1 (good) to 10 (poor) mild electric shocks . Unbeknownst to the participants, their partner was a confederate of the researchers. The na'lve participants always completed the assignment first and, when it was done, they received either one or seven shocks. The number administered was predetermined and independent of the quality of their work.

The researchers expected that the partici­ pants given seven shocks would be angry at their poor evaluation .

At this point, the participant was brought to the control room to judge his or her part­ ner's work and observed the device for admin­

istering the shocks. Some of the participants also saw either two guns or some badminton racquets situated on a nearby table. The re­ searchers found that the angry students vi­ sually exposed to the firearms administered significantly more shocks than the other par­ ticipants (see Figure 8.4). They interpreted this finding as an indication that aggressive cues can stimulate aggressive behaviour, which Ber­ kowitz (1968) later summarized by stating, "The

II> -"' v 0

..<: II>

'+­ 0

(ii .D E :J z

finger pulls the trigger, but the trigger may also be pulling the finger" (p. 22).

Critics of the study argue that the results simply reflect demand characteristics, which inadvertently en­ courage participants to respond in the predicted man­ ner (Page & Scheidt, 1971). A follow-up study by Turner and Simons (1974) casts doubt on this assertion. They found that, the more participants believed the study was about aggressive responses, the fewer shocks

they delivered. Comprehensive reviews conducted since this time confirm that the presence of weapons tends to increase aggression (Bettencourt & Kernahan, 1997; Carlson, Marcus-Newhall, & Miller, 1990).

7

6

5

4

3

2

• Firearms

• No objects

Badminton racquets

Not angry (one shock) Angry (seven shocks)

Study condition

Figure 8.4 Mean number of shocks delivered with and without

firearms present. Sou rce: Based on a figure in Berkowitz, L. (1993). Aggression: Its causes, conse­ quen ce s, and control. Philadelphia, PA: Temple Universit y Press .

8 Social Cognition EJ The second phase, during which deeper cognitive processing occurs, begins imme­

diately after the initial phase is underway. Now people start to think about and evalu­ ate the causes of the event, the motives of the other people involved, and their emotions (Berkowitz, 2003). They will likely form conclusions about whether their goals were acci­ dently or intentionally thwarted, which will diminish or intensify their anger. Their rudi­ mentary anger will become more clearly differentiated into such feelings as irritation or rage. Berkowitz (1990) notes that, if they have not acted already, people are apt to become aware of their feelings and weigh their actions more carefully during this phase. Although this stage does not eliminate the possibility of violence, it increases the likelihood of pro­ social responses, especially among people with appropriate prior learning experiences.

Social Interactions Decisions to act criminally or violently do not occur in a vacuum. As the preceding discus­ sion highlights, what happens to and around people, such as experiencing aversive events, affect their behaviour. The features and events in someone's immediate surroundings are known as situational (or contextual) factors. One of the most powerful situational factors is human social interaction. Deindividuation and obedience, which are discussed next, exemplify this type of situational factor.

Deindividuation

Deindividuation, a term coined by Festinger, Pepitone, and Newcomb (1952), is a psy­ chological state marked by a decreased sense of personal identity, self-awareness, and self-evaluation, which can result from being submerged in a group (Diener, 1977). Evidence indicates that, the larger the group, the greater the deindividuated effect (Postmes & Spears, 1998). The worry is that people in this state are more prone to violating social norms and offending (Festinger et al., 1952; Zimbardo, 1970). A study by Diener, Fraser, Beaman, and Kelem (1976) illustrates the possible consequences of deindividuation on trick-or-treaters. The study was conducted in more than two dozen homes around Seattle, Washington. When the door was answered at the pre-selected locations, trick-or-treaters saw two bowls on a table: a large one containing candy and a smaller one containing coins. Children in the anonymous condition were not asked anything by the adult occupant who answered the door, but those in the non-anonymous condition were asked their name and where they lived. After directing the children to take one candy from the bowl, the adult left them alone and went to another room while a hidden observer recorded the size of the group and the children's behaviour. The researchers reported that children in conditions conducive to deindividuation (i.e. trick-or-treating in a group) were more likely than the other children to violate social expectations by taking multiple candies or money (see Figure 8.5).

The study's results point to another factor-anonymity-that also contributes to deindividuation. It is no coincidence that the children who did not disclose their names or addresses were more likely to violate the situation's social expectations. Keep in mind that all the children in the study were presumably attired in Halloween costumes, which added to their feeling of anonymity. In fact, anything that makes us less identifiable, such as masks, dark glasses, similar uniforms, or even darkness, is thought to promote dein­ dividuation (Zimbardo, 1970). For instance, fifth graders randomly assigned to teams and issued the same colour t-shirts engaged in significantly more aggressive play than

situational (or contextual)

factors Features, events, or social interactions that char­ acterize a person's surround­ ing circumstances.

deindividuation A psycho­ logical state characterized by a loss of individual identity,

self-awareness, and self­ evaluation, which is often associated with being im­

mersed in large groups.

The Psychology of Criminal and Violent Behaviour

70

60 <I>

0 50.... Ill

0 ..r: ~ 40 <I>

.... ~ 30 c: <I> u (j; 20

c...

10

0

• Non-anonymous

• Anonymous

I Alone

Study condition

In a group

Figure 8.5 Impact of deindividuation on stealing by Halloween trick-or-treaters. Source: Copyright© 1976 by the American Psychological Association. Reproduced with permission. E. Diener et al. (1976). Effects of deindividuation variables on stealing among Halloween trick-or-treaters. Journal of Personality and Social Psychology, 33, 178-183. The use of APA information does not imply endorsement by APA.

the members of the opposing team, who were playing in their regular clothes (Rehm, Steinleitner, & Lilli, 1987). The role anonymity plays in deindividuation may explain the finding of an ethnographic study that observed people of cultures who adorn themselves in masks or face paint prior to going into battle tend to inflict more extreme violence than people whose faces are not obscured (Watson, 1973).

The effect of deindividuation is frequently implicated in the results of the well-known Stanford prison experiment (Haney, Banks, & Zimbardo, 1973; Zimbardo, Maslach, & Haney, 2000). Despite the absence of some basic methodological elements required for a true scientific experiment, the study is still relevant because of its realism. In this study, two dozen volunteers were assigned the role of guard or prisoner in a simulated prison constructed in the basement of Stanford University's psychology department. Before the study formally began, the guards drafted rules governing prisoner behaviour and were supplied with military-style uniforms and mirrored sunglasses, items thought to foster the sense of anonymity (Zimbardo et al., 2000; Haney et al., 1973). The prisoners wore an­ kle chains and were dressed in smocks that had a number sewn on the front and back. The prisoners were confined to the simulated prison around the clock, with teams of guards working in shifts and maintaining a constant watch over them. As the study went on, the researchers witnessed the guards becoming increasingly aggressive, punitive, and degrad­ ing toward the prisoners. In fact, the guards' behaviour and the detrimental effects it had on the prisoners eventually forced the researchers to terminate the study after six days instead of the two weeks originally scheduled (Zimbardo et al., 2000). The outcome of this study hints at the possible deindividuating effect of prison settings and may explain events such as those at Abu Ghraib (see the following case study).

Different ideas exist about why deindividuation leads to social transgressions. The classic view is that it reduces the anxiety and fear of being singled out and subjected to so­ cial disapproval and, as a consequence, people engage in behaviours they would otherwise inhibit (Festinger et al., 1952; Zimbardo, 1970). In short, deindividuation promotes anti­ social behaviour by diminishing social controls. The social identity model of deindividu­ ation effects (SIDE; Reicher, Spears, & Postmes, 1995; Spears, Postmes, Lea, & Watt, 2001)

8 I Social Cognition EJ

Lynndie England and US Military Personnel at Abu Ghraib

Bagdad, Iraq-During the Iraq War, Abu Ghraib pris­

on served as a detention facility for the US military. In

2004 d isturbing photos of various army personnel abus­

ing Iraqi prisoners came to light. Some of the photos

show Private Lynndie England smoking a ciga rette while

pointing at the genitals of a hooded naked prisoner

being forced to masturbate, holding a leash attached

to a crawling naked prisoner wearing a dog collar, and

standing behind a pyramid of naked prisoners, giving

a thumbs-up. England and 10 other members of the

372nd Military Police Company were eventually con­

victed of offences stemming from the abusive conduct

(" Lynndie England still haunted," 2009).

England and many of the other convicted per­

sonnel adm itted that prison e rs were stripped naked,

forced to wear w o men's cl oth ing and crawl across

the floor, denied access to toilets, deprived of sleep,

and subjected to other abuses; however, they main­

tained that these pra ct ices were establish ed prior

to their arrival. En g land claimed : "Whe n we first got

there, we were like, what's g o ing on? Then you see

staff se rg eants w alking around not saying anything

[about the abuse]. You think, OK, obviously it's nor­

mal" (Brockes, 2009, para. 29). Many of the accused

Photo 8.2 This image, showing Private Lynndie En­ gland holding a naked Iraqi detainee by a leash, was

just one of the many photos revealing prisoner ab use

at Abu Ghraib .

argued at trial that they were simply followin g o rders

and that the abusive treatment was actively encour­

aged to break the will of the prisoners before being

inte rrogated (Brockes, 2009; Hersh, 2004). The pros­

e cutor at Eng land's t rial soundly rejected this asser­

tio n, pointing out th at none of the priso ne rs in th e

ph otos with England was eve r questioned ("Lynndie

England still haunted, " 2009).

takes a very different stance. SIDE postulates that people in groups shift from identifying themselves as individuals to identifying as a member of the group and, as a result, begin to adopt the behavioural norm of those around them (Reicher et al., 1995; Spears et al., 2001). According to SIDE, deindividuation facilitates both antisocial and prosocial behaviour- it just depends on the norm of the particular group (Reicher et al., 1995). It also suggests that the beh aviour of people in groups becomes increasingly "socially regulated" rather than "deregulated," as the trad itional view suggests (Spears et al., 2001, p. 337). How do anonymity and crowds contribute to this process? With greater anonymity and larger crowds, an individual has less and less sense of his or her own identity; at the same time, the identity of the group grows in prominence (Spears et al., 2001).

SIDE aligns deindividuation theory with the well-established body of social psy­ chological research on conformity, which shows that individuals usually ad apt their

The Psychology of Criminal and Violent Behaviour

Photo 8.3 A lone American fan stands in a crowd of Canadians celebrating at the 2010 Winter Olympics.

Photo 8 .4 A crowd watches a car burn during the Van­ couver riots of 2011, which began after the Canucks lost game 7 of the Stanley Cup finals to the Boston Bruins.

behaviour to be consistent with those around them.

Deindividuation studies also appear to support SIDE. A meta-analysis of deindividuation research found that participant behaviour tends to follow the situational norm of the group, be it good or bad, as opposed to always acting contrary to the norms of society (Postmes & Spears, 1998). Finally, the theory appears better able to account for the wide range of behaviour displayed by different crowds, a point aptly illustrated by two events occurring a short time apart in Vancouver, British Columbia. In 2010 the city host­ ed the Winter Olympics, during which thousands of people thronged through the streets in a jovial and festive atmosphere (Sin, Cooper, & Mercer, 2010). Sixteen months later, a crowd in the same area of the city and probably involving some of the same indi­ viduals rioted, causing injuries to 150 people and mil­ lions of dollars in property damage, after the home team Canucks lost the Stanley Cup to the Boston Bruins (Chan, 2011). Many of the people eventually identified as participants in the riot claimed that they were "swept up" in the moment (e.g. Coleburn, 2011; Dhillon, 2012). While the traditional view of deindi­ viduation accounts for the rioters' actions, it does not fit the positive behaviour seen throughout the Winter Olympics. SIDE theory, on the other hand, can ex­ plain both events. It suggests that people adopted the emerging normative behaviour of the crowd, which happened to be good-natured and prosocial during the Olympics but decidedly antisocial and destructive following the Canucks' defeat.

Obedience

As mentioned in the Abu Ghraib case study, many of the convicted military personnel claimed that they were simply following orders. Similar defences were advanced by peo­ ple tried for their role in the Holocaust, a matter of deep interest to Stanley Milgram, an American psychologist with a strong Jewish identity who had grown up during World War II (Milgram, 2000). In a well-known series of studies, he investigated the degree to which ordinary Americans complied with commands to inflict pain on another person.

obedi ence The action of All Milgram's (1963, 1965) studies of obedience employed the same basic design. complying with the directions Community members were recruited through newspaper ads to participate in a study they of a higher authority.

believed was exploring the effect of punishment on learning. Pairs of participants drew to determine whether they would be the teacher or the learner. In reality, the draw was rigged so that a confederate of the researcher was always the learner and the na"ive participant was always the teacher. The teacher administered a task to the learner, who was in a separate

8 Social Cognition

room and hooked up to a shock generator. Whenever the learner made a mistake, the teacher was to deliver a shock, the strength of which increased with each incorrect answer to a maximum of 450 volts. Participants who expressed an unwillingness to continue at

any point were given up to four verbal prompts by the experimenter, beginning with the relatively mild "Please continue" and escalating to the strong command of "You have no choice, you must go on."

When Milgram (1965) asked a group of psychiatrists how far participants would go, they predicted that less than 1 per cent would remain obedient to the very end. The ac­ tual results were dramatically different. Despite obvious signs that many participants felt uneasy at what they were doing, 65 per cent of them delivered the maximum shock, even

though the learner pounded on the wall and stopped answering after the twentieth shock (300 volts).

Milgram (1965) conducted a series of follow-up studies using the same basic design and involving close to a thousand participants. He discovered that various situational fac­ tors, including the learner's proximity, affected people's obedience. In the most remote study condition, the teacher could not see or hear the learner except for the protest occur­ ring after the 300-volt shock. Proximity was increased in the second condition by adding a series of audible protests, beginning with grunts (75 V), followed by statements such as "I can't stand the pain" (180 V) and ending with agonizing screams of pain. The third condition included the verbal protests but situated the learner only 1.5 feet away from the teacher. Finally, in the fourth condition, the teacher had to physically hold the learner's hand against a shockplate. Milgram observed an inverse relationship between obedience and victim proximity. Participants were most obedient when the victim was farthest re­ moved and least obedient when the victim was closest (see Figure 8.6).

Milgram (1965) also examined the effect of changing the proximity of the authority figure. Putting the experimenter only a few feet away from the teacher was associated with the greatest obedience, whereas removing the experimenter from the room and issuing in­ structions to the teacher via a phone or recording was associated with the least obedience.

70

c: 60 w >

~ 50 u 0

~ 40 E ~

E 30 ·x "'E c: 20 <lJ"'

::::?: 10

0 Unseen and

unheard Unseen but

heard Seen and

heard Seen, heard,

and hand forced

Victim proximity

Figure 8.6 Effect of victim proximity on obedience to authority. Source: Adapted from Stanley Milgram, Human Relations 18(57), pp. 57-76, copyright© 1965 by SAGE. Reprinted by Permission of SAGE Publications, Ltd.

EJ The Psychology of Criminal and Violent Behaviour

social cues The words, gesture s, and actions of other p eo ple that p rovi d e clues about their fe e ling s, thoughts, and motives .

encoding Th e p ro ce ss o f perceiving and organizing incoming stimuli such as so cial cues .

In short, proximity to the victim and authority work in opposite ways. Obedience in­ creases the closer the individual is to the source of authority and decreases the closer the individual is to the victim.

One reason so many participants continued giving the shocks may be the task's in­ cremental nature. The intensity of the shocks increased in relatively small steps. Once participants decided to apply the shock at one level, it became difficult to justify not going just a little bit further and applying the next shock. While it is impossible to know with

certainty what happened in Abu Ghraib, similar processes were probably at work . There is evidence that the low-ranking military personnel convicted of perpetrating the abus­ es were acting under orders, although who gave them is not clear (Hersh, 2004). As we learned in our discussion of moral disengagement, acting on the authority of someone else serves to minimize personal responsibility. Moreover, the abuse may have started as rela­ tively minor and isolated incidents that gradually progressed into more egregious abuses and human rights violations.

Social Information Processing Theories Social information processing (SIP) theories liken the human mind to a computer that gen­ erates output based on the data going in and how they are processed (Huesmann, 1998). The "data" consist of the stimuli or social cues people perceive, such as the words and actions of others, which give them clues about their social situation; the "output" is the behaviour generated in response to the information processed (Dodge, 1986; Huesmann, 1998). Human behaviour, then, is the product of the underlying sequence of mental opera­ tions used to process social cues (Dodge & Schwartz, 1997). Do not let the computer anal­ ogy mislead you into thinking that these theories assume everyone is destined to behave exactly the same way. Human behaviour is highly complex and variable because every situation presents a unique set of social cues and everyone possesses different "hardware" and "software," which they use to process this information (Huesmann, 1998).

Since the 1980s, several models based on SIP theory have been proposed to explain the occurrence of antisocial and, more particularly, aggressive behaviour (Anderson & Bushman, 2002; Crick & Dodge, 1994; Huesmann, 1988). While these models diverge from one another in a number of ways, they share four key processing stages: encoding social cues, interpreting social cues, searching for behavioural responses, and evaluating and selecting a response (Anderson & Huesmann, 2003; Huesmann, 1998). Theorists speculate that information deficiencies and errors in the cognitive processes at one or more of these stages lead to poor decision-making and inappropriate behaviour. The following section summarizes the basic information processing stages of the models proposed by Kenneth Dodge (Crick & Dodge, 1994; Dodge, 1986) and Rowell Huesmann (1988, 1998).

Encoding Process

To respond effectively in any social situation, we must first understand it. Situational aware ­ ness begins when people perceive and organize incoming social cues, a process known as encoding. This process depends on our perceptual abilities (e.g. seeing and hearing) as well as where we direct our attention and what we remember (Dodge & Schwartz, 1997). Encoding is akin to the acquisition stage of observational learning discussed in Chapter 7. Central to SIP theories is the notion that everyone differs in their perception, attention,

8 Social Cognition

and memory; therefore, even people experiencing the same event do not acquire or retain exactly the same version of it. This appears to be particularly true for antisocial and ag­ gressive individuals. Studies show that youngsters who engage in aggressive behaviour are more likely than other children to seek out, focus on, and remember aggressive social cues in their environment (Dodge & Frame, 1982; Gouze, 1987). Not surprisingly, distorted or deficient information of this sort creates difficulties when it comes to accurately interpret­ ing the social situation (Dodge & Schwartz, 1997).

Interpretation Process

Interpretation involves integrating the available encoded social cues and other information to form a mental picture or understanding of the situation (Dodge , 1986). The interpreta­ tion generated is affected by the available social cues as well as the person's mood, arousal, memories of similar past experiences, and attributions (Crick & Dodge, 1994; Huesmann, 1998). Attribution is the process of making causal judgments about behaviour and events (Fiske & Taylor, 1991). For example, imagine that a child is hit with a ball thrown by some­ one playing with other children. In most cases, without more information, the child will assume it was simply an accident and not attribute any intention on the part of the ball thrower. We all use attributions on a daily basis because they help us make sense of others' actions and inform our decisions about how we should respond (Fiske & Taylor, 1991).

Unfortunately, the attributions we make are prone to errors such as hostile attribution bias, which is the tendency to interpret the ambiguous actions of others as signs of aggres­ sive intent or behaviour and is one of the major information processing errors identified by Dodge 's model (Crick & Dodge, 1994; Dodge, 1986). Returning to our example, some­ one exhibiting hostile attribution bias would assume that he or she was hit with the ball on purpose, despite the equivocal nature of the surrounding circumstances. This error in information processing can lead to problems because people tend to retaliate when they believe others are acting aggressively against them (Dodge, 1980).

Dodge (1980) demonstrated the presence of hostile attribution bias among aggres­ sive children who discovered the puzzle they were assembling was destroyed by a peer on purpose (i.e. hostile intent), by accident (i.e. benign intent), or in unclear circumstances (i.e. ambiguous intent). When participants were given an opportunity to act against the puzzle-wrecker, differences only emerged in the ambiguous intent condition. Boys known to be aggressive who were in the ambiguous condition displayed the same level of aggres­ sion as they did in the hostile intent condition. The non-aggressive boys' level of aggres­ sion in the ambiguous intent condition matched the level they displayed in the benign intent condition. These results suggest that the aggressive boys faced with ambiguous so­ cial cues were biased toward making hostile attributions, whereas the non-aggressive boys made non-hostile attributions. Other studies support this general finding (Nasby, Hayden, & DePaulo, 1980; Steinberg & Dodge, 1983).

Response Search Process

Once they interpret a social situation, people search their long-term memory for suitable behavioural responses (Crick & Dodge, 1994; Huesmann, 1998). Central to Huesmann's model is the idea that the behavioural responses retrieved from memory take the form of scripts. A script is an organized unit of knowledge or mental template that lays out

attribution The process of making causal judgm e nts about people's behaviour and events.

script An organized unit of knowledge or mental tem­ plate that lays out the e xpect­ ed sequence of behaviour for a particular social situation as well as the likely outco me of that behaviour.

The Psychology of Criminal and Violent Behaviour

priming The process that occurs when recent exposure to stimuli increases the acces­ sibility of associated mental structures such as scripts, raising the possibility of their subsequent acti vation .

response evaluation One of the co nsiderations used to evaluate whether to enact a particular script based on how closely it fits one's values and moral beliefs.

outcome expectations One of the considerations used to evaluate whether to enact a particular script based on a cost-benefit analysis of the possible positive and nega­ ti ve outcomes.

self-efficacy One of the con­ siderations used to e valuate whether to enact a particular script based on the likelihood that it can be successfully performed .

the expected sequence of behaviour for a particular social situation as well as the likely

outcome of that behaviour (Crick & Dodge, 1994; Huesmann, 1988). For instance, most

children have a well-defined script for Halloween: ring the bell, yell "Trick or Treat" when

the door is opened, and hold out the treat bag. The expected outcome of this script is that the person answering the door will put candy in the child's bag. Thus, scripts serve as guidelines for behaviour that help people navigate their daily social interactions and social problems (Huesmann, 1998).

People maintain countless scripts in their memory, which are based on their experi­ ences and observations of others (Huesmann, 1998). The particular scripts retrieved on any given occasion depend heavily on the interpretation assigned to the situation and past information processing patterns. Scripts that are frequently accessed, rehearsed, and uti­ lized are more likely to be retrieved over ones that are not (Huesmann, 1988, 1998). Script retrieval is also influenced by priming, which occurs when recent exposure to stimuli increases the accessibility of associated mental structures such as scripts and, as a result, increases the chance of subsequent activation (Huesmann, 1998). In other words, recent experiences involving particular objects, media, or emotions may "ready" certain scripts so they are more accessible and likely to be retrieved in the near future . In Berkowitz and LePage's (1967) study, discussed in the "Researching" box on aggressive cues, visual ex­ posure to guns was accompanied by higher levels of aggression. Presumably, the sight of the weapons primed aggressive scripts in the study participants, which they subsequently retrieved and enacted (Carlson et al., 1990).

Less socially competent people manifest a number of difficulties during the response search process. Research indicates that they tend to retrieve fewer scripts, especially as young children, and those they do retrieve are often less effective or inappropriate for the situation (Gouze, 1987; Richard & Dodge, 1982). Furthermore, scripts become increas­ ingly rigid and automatic with more use. This practice does not pose a problem if scripts are prosocially oriented, but it means that frequently repeated maladaptive processing patterns and behaviours eventually become habitual and resistant to change (Crick & Dodge, 1994). This is one reason why recurrent fantasies involving acts of violence are so troubling.

Response Evaluation Process

A script will be activated only if it is perceived to be the most appropriate choice among those retrieved (Crick & Dodge, 1994; Huesmann, 1998). The process of evaluating scripts and selecting one for enactment is theorized to encompass three main considerations. One is what Dodge calls response evaluation and involves assessing the script's con­ tent against the individual's values and moral beliefs (Fontaine & Dodge, 2006; Crick & Dodge, 1994; Huesmann, 1998). Scripts that fit, or at least do not conflict with, a person's values are preferred for enactment. Keep in mind that individual beliefs do not always re­ flect those held by society at large and, consequently, people with antisocial attitudes may view a violent or otherwise procriminal script as perfectly acceptable (Huesmann, 1998). A second consideration concerns outcome expectations (Crick & Dodge, 1994). Here, the script's anticipated positive outcomes are weighed against the possible negative outcomes (Fontaine & Dodge, 2006). The last consideration, self-efficacy, focuses on the likelihood that the person can successfully perform the response in the current situation. Research indicates that maladaptive scripts may be selected when people feel they lack the requisite

8 Social Cognition

social skills to execute more prosocial ones. In one study, aggressive children expressed significantly greater confidence in their ability to enact aggressive behaviours and signifi­ cantly less in their ability to perform more conciliatory responses than non-aggressive

children (Crick & Dodge, 1989). The evaluation process ends with the selection of a script, which is immediately en­

acted (Crick & Dodge, 1994). Fontaine and Dodge (2006) acknowledge that people do not always engage in a careful evaluation process before they act. They suggest that impulsive behaviour is an indication that the evaluation process has been skipped altogether. This may occur when something in the situation brings forth an especially salient script that is performed at once and "preempts" further processing (p. 617).

Enactment of Antisocial and Violent Behaviours The decision to enact antisocial or violent behaviour does not appear to hinge on any par­ ticular error in SIP. In cognitive terms, it is the scope and depth of the processing deficien­ cies that seem to separate people who regularly resort to antisocial and violent behaviour from the rest of us. It appears that chronically aggressive people habitually make multi­ ple errors affecting more than one processing stage (Yoon, Hughes, Cavell, & Thompson, 2000). Studies also demonstrate that, as the deficiencies in information processing be­ come more serious, the associated behavioural problems increase in severity (Lochman & Dodge, 1994).

Nonetheless, certain patterns of deficient information processing are thought to pro­ mote specific types of aggression (Dodge, 1991). Studies find that errors in the initial stag­ es of processing, such as hypervigilance to aggressive social cues and hostile attribution bias, correlate most strongly to reactive violence (Dodge & Coie, 1987; Dodge, Lochman, Harnish, Bates, & Pettit, 1997). These types of processing errors explain the emotional responses to (mis)perceived insults and provocations that are characteristic of reactive violence. The roots of these problems may lie in early childhood abuse, which fosters the development of hypervigilance to signs of aggression as well as instilling fear and rage reactions (Dodge et al., 1997). In contrast, people who perceive and interpret social cues accurately but possess scripts dominated by aggressive responses and believe this type of behaviour will be rewarded tend to engage in instrumental violence (Dodge et al., 1997; Schwartz et al., 1998). The origin of the processing deficiencies associated with instrumen­ tal violence is thought to be quite different. Dodge (1991) hypothesizes that exposure to violent role models at home or in the media is responsible for expanding a person's inven­ tory of aggressive scripts. Moreover, when people learn through observation or personal experience that these violent responses are rewarding, it increases the probability that they will retrieve these scripts from memory and enact them (Dodge, 1991).

While SIP theories were developed in an effort to explain chronically aggressive be­ haviour, the theories can also be applied to explain isolated incidents of violence, such as the case of George Zimmerman, a neighbourhood watch coordinator whose misplaced suspicions led to a lethal confrontation with unarmed teen Trayvon Martin (see the case study). It is not difficult to see how misattributions and other social information processing errors might have contributed to the tragic sequence of events in this case. Theorists are also beginning to extend the maladaptive patterns of information processing identified in SIP theories-especially those surrounding response evaluation-to other antisocial behaviours unrelated to aggression, such as cheating, theft, and drug use (e.g. Fontaine, 2006).

The Psychology of Criminal and Violent Behaviour

George Zimmerman and Trayvon Martin

Sanford, Florida-On 26 February 2012 George Zim­ merman shot and killed Trayvon Martin in the gated

community where both men were staying . Zimmerman

was the coordinator of the neighbourhood watch,

which he had helped organize after a number of

homes in the complex were burglarized . He took up

the role with evident vigour, contacting police dozens

of times to report disturbances, open windows, break­

ins, and suspicious people (Robles, 2012). Zimmerman

had ambitions of becoming a law enforcement offi­

cer and, several years before the shooting, he had

even pursued and helped apprehend a shoplifter ("In

Trayvon Martin shooting," 2012).

The evening of the shooting was dark and rainy.

Zimmerman was running errands in his truck when

he spotted Martin. The 17-year-old was walking

back to a home in the complex owned by his father's

fiancee. He was wearing a gray hoodie, with the hood

pulled over his head. Zimmerman called 911 and told

the dispatcher, "We've had some break-ins in my

neighbourhood. And there's a real suspicious guy"

(para. 58). He described Martin as being in his late

teens and black. After asking when the police would

arrive, Zimmerman remarked, "These assholes, they

always get away" (para. 62).

Meanwhile, Martin was talking to his girlfriend on

his cellphone and told her that someone was watch­

ing him. When she told him to run, he replied that he

would walk quickly. Zimmerman noticed the change in

pace and informed the dispatcher that Martin was now

Photo 8.5 George Zimmerman testifies at his bond hearing on 20 April 2012.

running. Despite police recommendations that neigh­

bourhood watch volunteers be unarmed, Zimmerman

took his gun from his truck and began trailing the teen­

ager on foot. When the dispatcher became aware he

had left his vehicle, Zimmerman was advised that they

did not need him to follow Martin and arrangements

were made for him to meet police. What happened

next is unclear, but the two men encountered each oth­

er and a violent confrontation ensued. Witnesses re­

ported seeing the two men wrestle and someone was

heard screaming for help numerous times, followed by

the sound of a single gunshot which killed Martin.

At the trial, the jury accepted Zimmerman's claim

that he shot Martin in self-defence and found him not

guilty of both second-degree murder and the lesser

charge of manslaughter (Alvarez & Buckley, 2013).

Source (except where otherwise noted): Barry, D., Kovaleski, S. F., Robertson, C., & Alvarez, L. (2012, April 1). Race, tragedy and out­ rage collide after a shot in Florida. The New York Times. Retrieved at http://www. nyt i mes .co m/2012/04/02/u s/trayvo n- ma rt in-shooting­ prom pts-a-review-of-id ea ls. htm I.

distal factors Factors that indirectly increase the risk of future violent behaviour by enhancing proximate factors that encourage violence or diminishing proximate factors that disinhibit violence.

Putting It All Together: The General Aggression Model Craig Anderson and Brad Bushman's (2002) general aggression model (GAM) comb~nes many of the theories discussed in this chapter, including social information processmg, cognitive neoassociation, and excitation transfer, as well as elements of the learning theories covered in Chapter 7. The GAM is intended to help us understand individual episodes of

violence that, according to the model, are products of both distal and proximate factors (An­ derson & Carnagey, 2004). Distal factors are typically biological (e.g. executive functioning

8 Social Cognition

problems, ADHD) or environmental (e.g. high-risk neighbourhood, antisocial peers) in na­ ture. These factors operate in the background and either increase proximate factors that encourage violence or reduce proximate factors that inhibit it.

Proximate factors relate to the characteristics of the immediate episode and exert a direct influence on behaviour. Person-based proximate factors pertain to the character­ istics of the individual. For example, we saw in Chapter 5 that people vary in the degree to which they exhibit self-control, narcissism, sensation-seeking, and other personality traits conducive to violence. Individual characteristics are important because they influ­

ence how people perceive and interpret their situation and this affects how they will react (DeWall & Anderson, 2011). GAM views cognition as critically important, especially those aspects that are the subject of frequent exposure and practice. As outlined in our dis­ cussion of SIP theories, when cognitive structures such as scripts are repeatedly accessed and rehearsed, decision-making becomes more automatic, which leads people to act with little or no active processing of their situation or their response to it (De Wall & Anderson, 20ll). Situation-based factors, mentioned earlier in the chapter, relate to features of the surrounding circumstances. Many of the situational factors identified in GAM incorpo­ rate aspects of other theories, such as frustrations, provocations, aggressive stimuli, and physiological arousal. Alcohol, which is discussed in Chapter 10, is also identified as a situational factor because of its disinhibiting role.

The GAM proposes that person and situation factors affect arousal, affect, and cog­ nition. Like some of the theories already discussed, these three are thought to be highly interconnected and therefore impact one another through the process of spreading acti­ vation. Collectively, they represent a person's present internal state, which in turn impacts his or her behaviour. In this sense, they serve as routes through which person and situa­ tion factors influence behaviour.

The arousal, affect, and cognitions generated by person and situation factors are con­ stantly appraised. Much of this evaluation occurs automatically without our awareness, something that is dubbed immediate appraisal. Immediate appraisal can lead to impul­ sive and unconsidered acts, including violence, but this is highly dependent on the char­ acteristics of the individual involved. Where the immediate appraisal is associated with an undesirable outcome and people have sufficient time and cognitive abilities, they will enter into a reappraisal. During the reappraisal, people look for further social cues about the event, comb their minds for relevant memories, and consider alternative explanations of what happened. The cycle of forming a possible explanation of the situation, evaluat­ ing it, and then discarding it may be repeated numerous times. Once the person settles on a satisfactory explanation, he or she will act. The more extensive cognitive processing undertaken during reappraisal means the behavioural responses that follow are more pur­ poseful and considered than acts flowing out of immediate appraisals. An overview of the GAM is presented in Figure 8.7.

The GAM model can be applied to the Abdulmutallab case. Few distal factors seem to be present because Abdulmutallab is a bright person raised in a warm, loving, and stable family. Although his family did not support violence, he became preoccupied with rad­ ical Islamic views that appear to have promulgated ideas of hatred and violence directed against the West. Essentially, he immersed himself in a social network that supported violence, a distal factor that likely fostered the development of person-based proximate factors such as normative beliefs about aggression and retaliation, attitudes about vio­ lence, and cultural stereotypes. It is evident from Abdulmutallab's statements that other

proximate factors Factors that relate to the characteris­ tics of the person or situation

and directly impact the risk of

violence.

immediate appraisal

A largely automatic, subcon­ scious process in which arous­

al, affect, and cognition are

constantly evaluated to make

inferences about a situation.

Immediate appraisals may

lead to impulsive behavioural

responses or a reappraisal.

reappraisal The process of

re-evaluating arousal, affect, and cognition to make infer­

ences about a situation when

the initial or immediate ap­

praisal was unsatisfactory and

time and cognitive resources

permit a further evaluation.

The Psychology of Criminal and Violent Behaviour

Distal causes and processes

Proximate causes and processes

Biological modifiers

..

internal state

Environmental modifiers

Apprai'sal dedsitm

be havi o ur

Figure 8.7 The general aggression model Source: Republished with permission of Guilford Publications, from The social psychology of good and evil, A.G. Miller (ed), 2004; permission conveyed through Copyright Clearance Center, Inc.

person-based proximate factors existed, including moral justification for violence, dis­ placement of responsibility, and dehumanization. His training with Al-Qaeda made nu­ merous aggressive scripts cognitively available and, by wearing the bomb for three weeks prior to boarding the plane, he was repeatedly rehearsing mentally and, to some extent physically, the violent act he was going to perform. The bomb represents a situation-based proximate factor. Ultimately, his extreme and violent ideology, the presence of the bomb, and the situational opportunity he was afforded aboard the plane, came together in his decision to carry out his plan.

Summary 1. Cognitive psychology is the study of the mental processes involved in perceiving,

thinking, remembering, and decision-making. The subfield of social cognition is con­ cerned with how people understand themselves and others. This chapter focused on theories of criminal and violent behaviour relating to social cognition.

2. Moral reasoning is the analytic process people use to make decisions about what is right and wrong. Lawrence Kohlberg proposed that moral reasoning passes through three levels of development: preconventional, conventional, and postconventional. Research and theory indicate that people operating at the lowest, or preconventional level, are the most likely to offend. Low moral reasoners decide what is right or wrong based on what will be punished or rewarded by those in authority. This environmen­ tal sensitivity explains the high rate of offending of preconventional reasoners be­ cause situations where crime is highly rewarding or the risk of punishment is very low arise frequently in modern society.

3. Albert Bandura proposed the process of moral disengagement to account for occa­ sions when people do not abide by their own morals. Moral disengagement involves

8 Social Cognition

the use of psychological mechanisms to avoid self-censure or feeling bad about acting immorally. These mechanisms centre on reinterpreting objectionable behaviour to make it less grievous, avoiding responsibility by shifting the blame elsewhere, min­ imizing the harm caused to victims, or dehumanizing victims so they appear less human or even deserving of harm.

4. Samuel Yochelson and Stanton Samenow identified more than 50 different thinking errors and behavioural patterns that facilitated criminal behaviour based on their clinical work with offenders. Glenn Walters and Thomas White extended this work by distilling it into eight cognitive errors or criminal thinking styles labelled mollifica­ tion, cutoff, entitlement, power orientation, sentimentality, superoptimism, cognitive indolence, and discontinuity. Importantly, Walters overcame one of the major short­ comings of Yochelson and Samenow's work by showing that the presence of these criminal thinking styles relates to reoffending.

5. Some theories covered in the chapter explicitly recognize that cognition and affect in­ teract with one another in ways that can produce aggression. Dolf Zillman's excitation transfer theory assumes that arousal dissipates slowly, which may lead people to mis­ takenly attribute the residual arousal to their current environment. Studies show that, in the wrong circumstances, pre-existing arousal can be mislabelled as anger and fuel aggressive behaviour. Affect also lies at the heart of the cognitive neoassociation model. It theorizes that aversive events, such as having one's expected goals blocked, create unpleasant feelings. This negative affect automatically activates two different networks of associated thoughts, feelings, physiological responses, and motor reac­ tions. One network centres on the tendency to attack the aversive source and the other one on avoiding it. The dominating behavioural tendency varies with the person and situation. This initial phase unfolds automatically; however, Berkowitz believes that people begin to cognitively process their feelings and consider their actions within moments, which increases the chances that they will restrain any urge to lash out violently.

6. Research shows that the presence of other people can exert a strong influence on what we think and how we act. Being immersed in large crowds can produce deindividua­ tion, a state characterized by a decreased sense of personal identity. Anonymity also promotes deindividuation. The traditional view of deindividuation is that it encour­ ages social transgressions; however, a more recent theory suggests that it may simply influence people to adopt the behavioural norm of the crowd, which could be either prosocial or antisocial. In addition, Stanley Milgram showed that the presence of an authority figure can profoundly influence people's behaviour. He observed that the closer people are to authority, the more likely they are to obey commands, even if it means hurting other people.

7. Social information processing theories suggest that inappropriate behaviour is due to deficiencies in one or more of the following stages of cognitive processing: encoding so­ cial cues, interpreting social cues, searching for behavioural responses, and evaluating and selecting a response. One type of processing error identified by Kenneth Dodge is hostile attribution bias, the tendency to interpret others' ambiguous actions as aggres­ sive, which in turn can lead to unjustified acts of retaliation. Rowell Huesmann empha­ sizes the role played by cognitive scripts or mental blueprints that set out the sequence of expected behaviours in a situation. He believes that aggressive people are more likely to maintain, access, and enact scripts involving aggressive behaviour.

The Psychology of Criminal and Violent Behaviour

8. The general model of aggression ( GAM) integrates many of the theories and research

findings discussed in this chapter into a comprehensive model that explains aggressive

episodes. According to the model, distal factors relating to a person's biology and en­

vironment enhance or diminish the presence of proximate factors. Proximate factors

relate to the characteristics of the immediate situation, such as the presence ofprovoca­

tion or weapons, as well as the characteristics of the person, such as his or her person­

ality. Collectively, these proximate factors directly impact the person's arousal, affect,

and cognition, which are appraised and, if necessary and time permits, reappraised

until the person settles on what seems to be an appropriate behavioural response.

Review Questions 1. Explain why Kohlberg's theory of moral development suggests that incarcerating

young offenders is counterproductive.

2. Define moral disengagement and describe two different mechanisms that people may employ.

3. Define the term priming and use this concept to explain what Berkowitz (1968) meant when he stated, "The finger pulls the trigger, but the trigger may also be pulling the finger" (p. 22).

4. Define deindividuation and provide two different theories for how it may contribute to offending.

5. Using SIP theory, identify possible processing errors and biases that might have con­

tributed to George Zimmerman's killing ofTrayvon Martin.

Additional Readings Anderson, C. A., & Carnagey, N . L. (2004). Violent evil and the general aggression model.

In A.G. Miller (Ed.), The social psychology ofgood and evil (pp. 168-192). New York, NY: Guildford Press.

Bandura, A. (1999). Moral disengagement in the perpetration of inhumanities. Personality

and Social Psychology Review, 3, 193-209.

Berkowitz, L. (1989). Frustration-aggression hypothesis: Examination and reformulation. Psychological Bulletin, 106, 59-72.

Blass, T. (2000). Obedience to authority: Current perspectives on the Milgram paradigm. Mahwah, NJ: Erlbaum.

Dodge, K. A., & Schwartz, D. (1997). Social information processing mechanisms in ag­ gressive behavior. In D. M. Stoff, J. Breiling, & J. D. Maser (Eds.), Handbook of antiso­

cial behaviour (pp. 171-180). New York, NY: Wiley.

Huesmann, L. R. (1998). The role of social information processing and cognitive schema in the acquisition and maintenance of habitual aggressive behavior. In R. G. Geen & E. Donnerstein (Eds.), Human aggression: Theories, research, and implications for social policy (pp. 73-109). New York, NY: Academic Press.

Palmer, E. J. (2003). An overview of the relationship between moral reasoning and offend­ ing. Australian Psychologist, 38, 165-174.

Walters, G . D. (1990). The criminal lifestyle: Patterns ofserious criminal conduct. Newbury

Park, CA: Sage.