ECE 354 Assessment & Intervention During Early Childhood / week 5 discussion 1 and 2, Final projects
What is Special Education? 1
iStockphoto/Thinkstock
Pre-Test
1. 1. You can use the terms disability and handicap interchangeably. T/F 2. 2. The history of special education began in Europe. T/F 3. 3. The first American legislation that protected students with disabilities was passed in the 1950s.
T/F 4. 4. All students with disabilities should be educated in special education classrooms. T/F 5. 5. Special education law is constantly reinterpreted. T/F
6. Answers can be found at the end of the chapter.
10Intelligence and Individual Differences
Voyagerix/iStock/Thinkstock
Learning Objectives By the end of this chapter, you should be able to:
• Analyze the qualities that determine IQ tests’ usefulness and adequacy for measuring intellectual differences in children.
• Identify and relate the hierarchy of intellectual abilities posited by the psychometric approach to intelligence.
• Explain how nature and nurture interact to influence intellectual differences in children.
• Differentiate stability, change, and modifiability in intelligence.
• Evaluate evidence that IQ scores can be improved in children through intervention.
Pretest Questions
Pretest Questions
1. The modern intelligence test was first developed to identify gifted children so that they could receive advanced instruction. T/F
2. Children’s performance on measurements of nonverbal intelligence (for example, spatial reasoning) is positively related to verbal intelligence (for example, vocabulary). T/F
3. Because genetic differences influence intellectual differences between children, the impact of the environment is minimal. T/F
4. By age 6 or 7, a child’s IQ score is a fairly good predictor of subsequent IQ scores throughout the child’s development. T/F
5. Generally speaking, if an intervention boosts a young child’s IQ score, the improvement is permanent even after the intervention ends. T/F
Michael is a 6-year-old who is struggling both academically and socially. His parents are recently divorced, and he is experiencing significant psychosocial stress. Michael had some issues prior to the divorce, but his symptoms have been exacerbated. Michael’s teacher is unsure if the child’s difficulties in school stem from problems coping with the divorce or from cognitive delays that make it difficult to keep pace with the other children. As part of the assessment procedure, the school psychologist, Dr. Williams, administers an intelligence test to measure Michael’s cognitive functioning. Intelligence test performance is a factor in determining whether a child is eligible for specialized services (Individuals with Disabilities Education Improvement Act, 2004).
Questions on the intelligence test ask Michael to remember a string of randomly arranged digits (“Repeat after me: ‘3-1’”). Others require him to arrange puzzle pieces or solve addition and sub- traction problems. During testing, the 6-year-old seems tired or perhaps unmotivated. However, Williams avoids providing minor hints and suggestions to prompt Michael and help him out a bit. When the test is over, Williams tallies the child’s correct answers and determines how his scores compare with a large sample of same-age children. He will look for patterns that indicate areas of relative weakness and strength.
At the end of the assessment, Williams faces a dilemma. Michael’s outcome on the intelligence test finds him right on the borderline for placement in a special education program.
Questions to Think About
1. Should Williams have provided some hints to help the child during testing? Explain your reasoning.
2. Do you believe an intelligence test would provide enough information to make a decision about Michael’s placement? Why or why not? If not, what other information should Williams gather? Explain your answer.
3. Michael is a very young child—how should his age factor into the decision to place him in a special education program?
4. Would the divorce impact Michael’s cognitive functioning? If so, how?
Introduction
Introduction Differences in children’s intelligence predict later income, occupational status, and educa- tional attainment in adulthood (Firkowska-Mankiewicz, 2011). Children’s intelligence is even related to later health and length of life (Batty, Deary, & Gottfredson, 2007)! This broad influence raises a number of questions. What is intelligence? How is it measured? Why do children differ in intelligence? What steps, if any, can be taken to increase intelligence? These questions are heavily researched, and we will address them and other related questions in this chapter.
Although there is no single, universally agreed-upon definition, many contemporary psychol- ogists would agree with the description of intelligence as: “The ability to reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly and learn from experi- ence” (Gottfredson, 1997, p. 13; see also Kranzler & Floyd, 2013). Another definition is that intelligence is the “ability to understand complex ideas, to adapt effectively to the environ- ment, to learn from experience, to engage in various forms of reasoning, to overcome obsta- cles by taking thought” (Neisser et al., 1996, p. 77).
Notice that these definitions include cognitive processes that are broadly applicable, like reasoning, abstract thinking, and learning from experiences. Processes like reasoning and learning from experiences are evident every day in multiple settings. As you reread the defi- nitions, notice also how the cognitive processes relate to schooling. As we will see, from the beginnings of modern intelligence testing, the concept of intelligence has been closely tied to school performance.
Intelligence can also be defined as an ability or set of abilities that extend beyond those found in the previous definitions, and we will discuss various theories of the nature of intel- ligence in Section 10.2. For now, however, we will work with a conception of intelligence that is tied to the definitions we have just presented. These definitions are widely accepted and underlie most tests of intelligence and contemporary research on the subject (Nisbett et al., 2012).
Core Themes, Intelligence, and Individual Differences The nature and measurement of intelligence is a broad, expansive topic. The four overarch- ing themes in our text help us organize our understanding of the wide-ranging material. The themes also place the study of intelligence into context by relating the material to topics and ideas we have encountered in previous chapters.
Nature and nurture. The extent to which intellectual differences are related to genes or envi- ronment has generated considerable theorizing and large-scale investigations over the years. Consequently, the nature–nurture theme is evident throughout this chapter. Recall from Chap- ter 1 that the nature–nurture theme is often raised in the context of why people differ from one another. This concern with individual differences, idiographic development, is a focus as we explore why children differ from one another in intelligence.
Section 10.1Intelligence Tests
Performance and competence. Second, intelligence testing is intended to accurately measure differences in children’s intellectual competencies. We will address whether intelligence tests are considered fair and accurate. Such concerns are related to the performance–competence theme of our book. Are the test scores really measuring intelligence (competence), or do the scores reflect, to some extent, performance limitations (for example, low motivation or cul- tural misunderstandings)?
Continuity and discontinuity. Generally, intelligence testing is concerned with stable dif- ferences among individuals, rather than stage-like cognitive changes that occur within an individual (Miller, 2012a). Consequently, in this chapter we primarily focus on the question of whether there is continuity in intellectual differences. We address whether intellectual differences are relatively stable and continuous during childhood, whether children’s intel- ligence fluctuates during development, and whether childhood intelligence predicts later outcomes.
Domain general and domain specific. Our fourth theme is concerned with whether cognition is better characterized as domain general or domain specific. In this chapter, we will see evi- dence indicating that intelligence is a general ability that cuts across a wide variety of cogni- tive domains. We will also encounter a theory that suggests there are many intelligences, each somewhat specific to a particular domain.
10.1 Intelligence Tests The results of intelligence tests can have a powerful impact on a child’s educational future, as we saw in the case study. Results can influence whether a child is placed in special education classes. Sometimes children are tested for placement into accelerated education programs. Intelligence tests are also used for clinical assessments; for instance, the results of a test can help determine if a child who recently suffered a head injury is experiencing cognitive impair- ments as a result of the injury.
The influence of intelligence testing naturally raises two questions: How is intelligence mea- sured, and how do we know if an intelligence test is dependable and accurate? Our focus on testing provides a foundation for subsequent sections in which we discuss theories of intel- ligence, its stability and change, and the factors that influence individual differences in intel- ligence. First, however, we will describe how intelligence tests developed and how they are used today.
The Development of Intelligence Testing The French Third Republic passed laws in the early 1880s making school mandatory and free for all 6- to 13-year-old children (Nadeau & Barlow, 2010). The inclusiveness of the French education system meant that classrooms were filled with children of varying backgrounds
Section 10.1Intelligence Tests
and skill levels. Soon government officials were faced with the dilemma of how to effectively handle children who had difficulty keeping pace with their peers.
An instrument that could accurately identify at-risk children was sought, particularly when concerns arose over the accuracy and impartiality of a school official’s or parent’s opinion. Alfred Binet was a French psychologist who, along with his collaborator Theodore Simon, set out to resolve the problem by creating an effective measurement of children’s intellec- tual capabilities (Binet & Simon, 1916; Nicolas, Andrieu, Croizet, Santioso, & Burman, 2013; Wolf, 1973).
Binet’s test ultimately contained three distinctive features still found in many intelligence tests (Siegler, 1992). First, it measured high-level cognitive abilities like memory, vocabulary, and reasoning that clearly related to the child’s schoolwork (Binet & Simon, 1916). These abilities are closely related to the definitions of intelligence presented in the previous section.
Second, a single composite score summed across different cognitive subtests was used to estimate a child’s intelligence (Siegler, 1992). As we will see, this feature of intelligence test- ing remains influential. It also brings to the forefront the question of whether intelligence is really a singular entity. We discuss theories that address whether intelligence is made up of one or numerous entities in Section 10.2.
Third, test items were arranged in sequence from less to more difficult. This arrangement allowed test administrators to determine how far children could progress before reaching their ceiling. A ceiling is typically established when a child misses a specified number of items. Once the child’s ceiling was established, the tester could then compare it with the average ceiling reached by same-age peers. In our case study at the beginning of the chapter, Michael’s score is low because he missed items that other children his age typically answer correctly. In other words, Michael did not progress through as many items as his peers.
IQ Tests Today The conventional scoring method used to express intelligence is known as the intelligence quotient (IQ). Conceptually, an IQ score is derived by comparing an individual’s performance (called a “mental age”) against the typical performance of those who are the same chronologi- cal age. In this formulation, IQ = mental age ÷ chronological age (Stern, 1914). For instance, if an 8-year-old’s performance is at the level of a typical 10-year-old, she would have a score of 1.25 (that is, 10 ÷ 8). Conventionally, average IQ scores are set at 100 (Kranzler & Floyd, 2013). Thus, 10 ÷ 8, (or 1.25) multiplied by 100 results in an IQ score of 125.
Figure 10.1 illustrates the types of items found on commonly used intelligence tests. These tests generally have (a) subtests that measure cognitive abilities (for example, math, vocabu- lary, math, and reasoning) and (b) a full-scale composite score that is derived from subtest performance.
Section 10.1Intelligence Tests
Figure 10.1: Sample items found on IQ tests
Examples of questions found on commonly administered intelligence tests. Note that items assess various abilities such as verbal comprehension (Information; Vocabulary) and reasoning (e.g., Block Design, Picture Completion).
Matrix Reasoning
Find the missing piece from the six pictured below.
Typical IQ Subtests
Information
Arithmetic
Vocabulary
Comprehension
Picture Completion
Block Design
Object Assembly
Coding
Picture Arrangement
Similarities
On what continent is Argentina?
If four toys cost six dollars, how much do seven cost?
What does “debilitating” mean?
Why are streets usually numbered in order?
Indicate the missing part from an incomplete picture.
Use blocks to replicate a two-color design.
Assemble puzzles depicting common objects.
Using a key, match symbols with shapes or numbers.
Reorder a set of scrambled picture cards to tell a story.
In what way are dogs and rabbits alike?
?
Source: Flynn, J. R. (2007). Solving the IQ puzzle. Scientific American Mind, 18(5), 24–31.
Some commonly administered intelligence tests are summarized in Table 10.1.
Table 10.1: Representative IQ tests administered to children
Test Age range
Features
Stanford-Binet Intelligence Scales, fifth edition
2 to 85+ • Ten subtests assess cognitive processes such as working memory, general knowledge, quantitative and spatial reasoning, and vocabulary.
• Subtests are administered in both verbal and nonverbal forms. • All 10 subtests are scored. The full-scale IQ score combines all 10
subtests.
(continued)
Section 10.1Intelligence Tests
Test Age range
Features
Wechsler Preschool and Primary Scale of Intelligence, fourth edition (WPPSI-IV)
2 to 7 • Ages 2 to 3 assessed by three subtest areas of verbal comprehension, spatial knowledge, and working memory; testing for ages 4 to 7 also includes processing speed and reasoning subtest.
• Subtest scores and full-scale IQ. • Ancillary scores enhance utility for special cases, such as the
assessment of school readiness or suspected language delays. Ancillary scores include a cognitive proficiency score (working memory and processing speed) indicating information- processing efficiency.
Wechsler Intelligence Scale for Children, fifth edition (WISC-V)
6 to 16 • Five subtests assess verbal comprehension, spatial knowledge, working memory, processing speed, and reasoning.
• Subtest IQ scores and full-scale IQ • Ancillary index scores provide additional information relevant to
clinical situations (such as a nonverbal index score for children with autism) and school achievement (such as a quantitative reasoning index).
Kaufman Brief Intelligence Test, second edition (KBIT-2)
4 to 90 • Provides a quick screening measure for identifying children at academic risk or those eligible for enriched educational programs.
• Three subtests. Two measure verbal knowledge such as vocabulary, and one measures nonverbal knowledge related to reasoning about novel problems (that is, choosing which of five pictures best matches the concept in a target picture).
• Subjects receive verbal, nonverbal, and IQ composite scores.
Sources: Bain & Jaspers, 2010; Kaufman & Kaufman, 2004; Raiford & Coalson, 2014; Roid, 2003; Wechsler, 2014.
IQ Scores As noted earlier, a child’s IQ score is determined by comparing his or her performance with same-age peers. The comparison group is called the normative sample. A normative sample is a group of individuals who are representative of the larger population. In order to be repre- sentative, the sample must be inclusive and proportionately reflect the different backgrounds and ethnicities within a population. To illustrate, the normative sample for the Stanford-Binet (fifth edition) totaled nearly 5,000 individuals and was chosen to match demographic infor- mation derived from the 2000 U.S. Census (Roid, 2003).
IQ scores in normative samples generally form a normal distribution (Gottfredson, 2008). This means scores within the sample most frequently occur in the middle—the overall aver- age—and then occur less and less frequently as scores deviate further and further from the average (see Figure 10.2). Recall that most IQ tests are quantified so that the average score is 100 (Kranzler & Floyd, 2013). Thus, when a child is said to have “above average” intelligence, this means, in a literal sense, that the score is higher than 100, the average score of children the same age.
Table 10.1: Representative IQ tests administered to children (continued)
Section 10.1Intelligence Tests
Figure 10.2: Normal distribution
Scores within a normal distribution tend to cluster toward the middle (average) in their frequency. As scores deviate from the average, they occur with less and less frequency. IQ scores typically follow a normal distribution.
0 1 2 3–1–2–3
100 115 130 145857065
0.1% 2%
13% 13%
34%34%
2% 0.1%
Standard Deviation IQ Score
A standard deviation measures the extent that scores vary from the average. Conventionally, IQ scores change by 15 points for every standard deviation above or below the average (Kranzler & Floyd, 2013). In other words, IQ scores that are one standard deviation above the average are 115, while scores one standard deviation below the average are 85. When scores are normally dis- tributed, 68% of scores fall within one standard deviation of the mean, and 95% of the scores fall within two standard deviations (see Figure 10.2). Thus, 95% of IQ scores typically fall between 70 and 130.
Scores that are two standard deviations below the mean (scores of 70 or lower) indicate an intellectual disability (Prifitera, Saklofske, & Weiss, 2008). An intellectual disability is char- acterized by “significant limitations both in intellectual functioning and in adaptive behavior” that originate before age 18 (Schalock, Luckasson, & Shogren, 2007, p. 118).
The transition to school is difficult for children diagnosed with intellectual disability—the cognitive and behavioral expectations of schooling can often exceed the child’s readiness. Social skills (for example, appropriate expression of feelings) and self-regulation skills can positively contribute to the ability of 5- to 6-year-old children with an intellectual disability to adapt to school (McIntyre, Blacher, & Baker, 2006). Consequently, early interventions that focus on social and behavioral competencies may be useful in smoothing the transition to schooling for children who have an intellectual disability.
Children identified as gifted (also referred to as “talented”) exhibit extraordinary intellectual ability, promise, creativity, and motivation (McClain & Pfeiffer, 2012). Children are generally identified as “gifted” or “talented” in intelligence as their IQ scores approach 130 (McClain & Pfeiffer, 2012). Other measures, such as teacher ratings for motivation, are also typically incorporated into the process of determining giftedness.
Section 10.1Intelligence Tests
The pace is often accelerated in educational settings for children identified as gifted (Subot- nik, Olszewski-Kubilius, & Worrell, 2011). Also, educational opportunities generally provide an enriched curriculum that features topics (for example, robotics) not found in typical cur- ricula (Subotnik et al., 2011). Longitudinal research has linked participation in enriched edu- cational opportunities—such as taking college courses while in high school or participating in science fair/math competitions—to later accomplishments in STEM (science, technology, engineering, and math disciplines) for adolescents who scored exceptionally high on math assessments (Wai, Lubinski, Benbow, & Steiger, 2010). Such findings illustrate the role of nur- ture (that is, enriching experiences) in facilitating later accomplishment for students identi- fied as gifted.
Standards for Evaluating Intelligence Tests By what standards can we evaluate whether an intelligence test is actually useful? The ques- tion arises when a child is given a cognitive-related diagnosis. We want to know whether the test informing the diagnosis is trustworthy. Professional guidelines and legal requirements for fairness help ensure that intelligence tests are extensively vetted for their soundness (Gottfredson & Saklofske, 2009). Two fundamental criteria for evaluating an assessment are its reliability and validity. These constructs are key aspects of the vetting process. Each is discussed in turn in this section.
As we discuss reliability and validity, keep in mind that these concepts apply to all psychologi- cal tests. In recent decades, testing has become increasingly common in schools. Education policy in the United States has emphasized guidelines that define what students should know and the assessment of whether those guidelines have been met (Hamilton, Stecher, & Yuan, 2012). The criteria we use to evaluate these educational assessments include reliability and validity.
Recall our case study. If Michael was retested a few days later and his IQ score was vastly different from his first test, we would not know which result, if either, to believe. We would question the reliability of the test. Test reliability refers to the extent a measurement pro- duces consistent results (Cook & Beckman, 2006). Consistency can be measured by compar- ing whether an individual’s test and retest scores are similar. The reliability of a test is dif- ferent from the stability of the construct of intelligence (Miller, 2012a). Reliability is typically measured across a period of days or months. Stability is measured across a period of years. Commonly administered intelligence tests generally have very good reliability (Flanagan & Harrison, 2012).
Second, an assessment is only useful to the extent it is accurate. Test validity is established when we are confident a test is measuring what it claims to measure (Cook & Beckman, 2006). Historically, the purpose of intelligence tests was to inform educators’ decisions about a child’s academic capabilities and potential. In the context of their historical purpose, school grades are a typical criterion for establishing whether an intelligence test is measuring what it purports to measure. A valid IQ test should bear some relationship to academic achievement.
Generally speaking, the correlation between IQ scores and school grades is approximately .5, which means the relationship is moderate (Duckworth, Quinn, Lynam, Loeber, & Stouthamer- Loeber, 2011; Neisser et al., 1996). Additionally, those with higher IQs tend, on average, to have higher incomes, educational attainment, and occupations associated with attaining a
Section 10.1Intelligence Tests
college education and holding a professional position (Gottfredson, 2004a; Gottfredson & Deary, 2004; McCall, 1977). Taken as a whole, these various findings are evidence for the validity of commonly administered intelligence tests.
An ongoing question about intelligence test validity asks if the tests are fair to children from all cultural backgrounds. If individuals have a set of experiences that bias them to consistently misinterpret and miss test questions, then the test is not a valid measure of their ability. An invalid test would be a poor predictor of outcomes for members of that particular culture.
There are cases in which intelligence test content does not meaningfully reflect a child’s back- ground and experiences. For instance, questions asking about U.S. presidents and history would be unfair for a child who recently immigrated to the United States and has not studied U.S. history. Similarly, children learning English as a second language are at an obvious dis- advantage on intelligence tests that assess verbal knowledge like vocabulary (Gottfredson & Saklofske, 2009). Consequently, the validity of IQ tests in the United States is in doubt when a child is not English-speaking and/or is new to the culture and unfamiliar with the knowledge and skills IQ tests assess (Kranzler & Floyd, 2013).
In the News: Intelligence Tests and the Death Penalty The Supreme Court recently examined arguments about whether a death row inmate’s IQ score should make him ineligible for capital punishment. As you read the article, consider the arguments for and against the use of strict IQ cutoff points to determine whether a death sentence can be carried out.
ht tp://w w w.pbs.org/newshour/rundown/supreme-court-skeptical-iq-scores-deciding- execution
Critical-Thinking Question
What criteria should an intelligence test have to meet, in your opinion, to be a factor in legal questions such as death penalty cases?
Aside from these exceptions, many researchers point out that the weight of evidence indi- cates intelligence tests possess acceptable validity across cultural and ethnic groups in the United States (Kranzler & Floyd, 2013). This is because intelligence tests predict outcomes in a comparable, reasonably effective manner for different groups. In short, validity across a wide spectrum of groups is evidence of fairness.
Of course, each child brings a unique set of circumstances to the assessment process. Clini- cians and educators must always strive for cultural fairness in testing on a case-by-case basis. Guidelines to this end include:
• use assessments that are psychometrically sound, • use multiple methods of assessment, and • be aware of each individual’s background and circumstance when making evalua-
tions (Reynolds & Ramsay, 2003).
Section 10.1Intelligence Tests
Psychometrically sound tests (see the first point in the above list) are those that are reli- able, valid, and have a normative sample that is culturally and ethnically diverse. Commonly administered intelligence tests like those summarized in Table 10.1 generally possess ample evidence of psychometric soundness (Bain & Jaspers, 2010; Flanagan & Harrison, 2012).
Nonintellective Influences on IQ Scores Psychological assessment relies on the objective administration of tests. This means that each test should be administered in a consistent manner for all children through strict adherence to instructions and other testing procedures.
If testers vary in how they present material—for example, if some offer hints and praise, while others simplify instructions—their well-intentioned modifications actually undercut the test’s usefulness. This is because the child’s score cannot be accurately compared to the normative sample if there are differences in the way the test was administered to the child and the comparison group. It would be impossible to know, for instance, if a child’s above average test score reflected above average intelligence or the advantages of receiving hints and suggestions that were unavailable to the normative sample.
In our case study we asked if Dr. Williams should have provided prompts to help Michael dur- ing testing. Williams may have been concerned that the child’s performance was not reflect- ing his true competence. However, if Williams’s goal was to identify the child’s intellectual abilities by comparing his score to the normative sample (which is typically the goal of such assessments), then he should not have offered hints and prompts.
Though psychologists and other professionals who administer intelligence tests are highly trained to minimize nonintellective influences on IQ scores, it is not possible in all cases. These influences refer to factors impacting a child’s IQ score that have nothing to do with the child’s intelligence per se. One factor we just noted is called assessor bias. This occurs when a test administrator strays from the testing procedures by, for instance, prompting children with suggestions, not enforcing time limits, or erroneously presenting items (McDermott, Watkins, & Rhoad, 2014).
The term assessor bias can be misleading because it suggests deliberate fraud. In reality, it simply means that errors can occur in the administration of intelligence tests. Ensuring tests are appropriately and consistently administered is crucial for making the results useful.
The concern over assessor bias is a longstanding one that dates to the creation of the modern intelligence test:
An inexperienced examiner has no idea of the influence of words; he talks too much, he aids his subject, he puts him on the track, unconscious of the help he is thus giving. He plays the part of pedagogue, when he should remain psychologist. Thus his examination is vitiated. It is a difficult art to be able to encourage a subject, to hold his attention, to make him do his best with- out giving aid in any form by an unskillful suggestion. (Binet & Simon, 1916, pp. 44–45)
Section 10.2Theories of Intelligence
Another potentially important source of nonintellective influence is the child’s level of motivation during testing. A child’s disinterest during testing can obscure his or her true cognitive abilities. A common experimental method for researching the effects of motivation is to offer chil- dren rewards that are tied to test performance. If rewards improve performance compared to typical test settings, then we can infer that outcomes are influenced, to some extent, by the child’s level of motivation.
A number of studies conclude that providing incentives (for example, tokens that can be cashed in for rewards) improves children’s test performance, particularly for those identi- fied with below-average IQ scores (Duckworth et al., 2011). Such findings raise the possibility that correlations between IQ scores and academic outcomes are influenced to a degree by individuals’ motivation (Duckworth et al., 2011). That is, an individual with low motivation during IQ testing may also experience relatively low levels of motivation in aca- demic assessments at school. Low motivation would impact performance in both settings.
We caution that although motivation may influence test performance for some children, IQ tests are, as we have already noted, generally valid measurements that use- fully predict school performance and other related out- comes (Duckworth et al., 2011). It would be misleading to immediately assume a relatively low IQ score simply reflects a lack of motivation. However, the influence of motivation is worth our consideration because it high- lights the performance–competence theme. Consider- ing the role of motivation usefully draws our attention to the possibility that a child’s intelligence score reflects factors that extend beyond the child’s actual intellectual competence.
10.2 Theories of Intelligence A score on an intelligence test is not identical to intelligence itself. Intelligence is a construct that is measured, or manifested in, intelligence test performance. Theories of intelligence influence how we define and measure the construct of intelligence and how we interpret the results of intelligence tests (Goldstein, 2013).
Theories clarify our understanding of the nature of intelligence. Consequently, theories guide interventions that target specific cognitive processes and help answer why some interven- tions are more effective than others (Kaufman, Kaufman, & Plucker, 2013). By identifying the underlying cognitive processes of intelligence, theories also inform educational goals. For
Fuse/Thinkstock Providing incentives can help improve children’s test performance.
Questions to Consider
1. What are some reasons why a child might be unmotivated to take an intel- ligence assessment?
2. If a child is generally unmotivated and disengaged when tested, does this mean his low IQ score is not valid? Why or why not?
Section 10.2Theories of Intelligence
instance, if widely accepted theories target abstract reasoning as an important aspect of intel- ligence, then educational efforts to develop intelligence will focus curricular goals on foster- ing reasoning.
Throughout the history of modern intelligence testing, theories and tests have had a mutual influence. If a theory defines quantitative reasoning as a critical component of intelligence, then a test will contain quantitative problems. It is also the case, as we will see, that the results of intelligence tests refine and clarify our understanding and our definitions of intelligence. For instance, the results of intelligence tests are analyzed to examine if intelligence is a singu- lar ability that impacts performance across a range of problems.
At present there is no single agreed-on and overarching theory of intelligence (Gardner, 2011). In this section, we will describe the dominant theoretical approach—the psychometric approach—and often-cited alternatives to this dominant approach.
The Psychometric Approach Sometimes an individual will refer to herself as a “math person” who particularly enjoys, and excels at, working with numbers. A “history buff ” may be someone whose cognitive strengths lie in quickly mastering and remembering lots of verbal information. Theories of intelligence ask how different cognitive tasks—like mastering math or history— may or may not be linked together by underlying cognitive processes (Gardner, 2011). How closely related are verbal and math abilities, and are they each influenced by a single, broad ability? The psychometric approach attempts to understand the nature and structure of intelligence by looking for pat- terns in the results of large data sets of intelligence test scores (Flanagan, Alfonso, Ortiz, & Dynda, 2013).
In this approach we can understand the qualities of intelligence using statistical procedures like factor analysis. Factor analysis is a statistical technique for detecting how variables are structured and related to one another. For instance, factor analysis informs us whether per- formance on language tasks is related to performance on memory tasks. It also addresses whether abilities like language and memory are separate or related by an underlying cogni- tive ability.
Intuitively, cognitive performance seems to vary from one task to another. One child may find math coursework easier than coursework that emphasizes reading comprehension. The reverse may be true for another child. However, decades of intelligence testing reveal this conventional wisdom is somewhat misleading. Scores on tests that assess seemingly diverse areas like spatial reasoning, vocabulary, memory, math, and others all tend to positively cor- relate with one another (Gottfredson, 2005). The phenomenon of positive intercorrelations among measurements of cognitive ability is known as the positive manifold. It is a consis- tent, frequently made observation that dates to the dawn of intelligence testing in the 20th century (Spearman, 1904; Van der Maas, Kan, & Borsboom, 2014).
The positive manifold indicates that cognitive tasks have something in common with one another. That is, diverse tasks such as defining vocabulary items and solving math problems draw on the same general intellectual ability. This common factor underlying scores on a wide variety of intelligence tests is often denoted by psychologists as g (Spearman, 1904), which stands for a global, or general, intelligence that is responsible for widely impacting cognitive performance (Gottfredson, 2002).
Section 10.2Theories of Intelligence
Findings suggest that g is not simply located in a single area of the brain. Instead, intelligence appears related to activity in a network of neural connections distributed throughout the brain (Barbey et al., 2012; Deary, 2012; Jung & Haier, 2007). The efficient flow of informa- tion across these regions appears to be associated with higher intelligence test scores (Deary, Penke, & Johnson, 2010). This indicates processing speed is an influential component of intelligence.
Information-Processing Theory and Intelligence As just noted, the psychometric approach identifies processing speed as an underlying com- ponent of intelligence. In IP theory, processing speed refers to the efficiency of cognitive func- tioning (Sweet, 2011). It is measured by presenting timed cognitive tasks and calculating, often in milliseconds, the speed with which someone responds.
To illustrate the measurement of processing speed, in one condition children would be instructed to push a button on a computer keyboard upon seeing a letter appear on a screen (Carlozzi, Tulsky, Kail, & Beaumont, 2013). In a second condition the particular button chil- dren push would depend on whether a lowercase or uppercase letter appeared. Notice that the second condition requires making a decision before responding as quickly as possible.
From an IP theory perspective, the second condition requires the child to attend to the type of letter presented on the screen and decide how to respond based on the task instructions held in memory. The average difference in time it takes to complete the more difficult task is an index of processing speed. This is because the time difference reflects how long it takes the child to decide how to respond compared to how long it takes to respond when no decision is required (as in the first condition) (Carlozzi et al., 2013).
When given such tasks, children and adults with faster processing speed times tend to have higher scores on measurements of g than those with lower times (Carlozzi et al., 2013; Deary, Der, & Ford, 2001; Sheppard & Vernon, 2008). Thus, there is a link between processing speed and individual differences (that is, idiographic differences) in intelligence. Theoretically, speed in encoding information, scanning and organizing it in working memory, and retriev- ing information from long-term memory contributes to the link between processing speed and intelligence (Vernon & Jensen, 1984).
Neural maturation during childhood and adolescence underlies increases in processing speed for typically developing children (Ferrer et al., 2013). Consequently, faster process- ing impacts age-related changes in intelligence. In other words, developmental changes in processing speed are linked to normative changes in intelligence (Coyle, Pillow, Snyder, & Kochunov, 2011). By normative changes we mean that as children mature they answer more questions correctly—their raw scores improve on measures of intelligence. Process- ing speed, therefore, impacts both individual differences and age-related developmental differences.
Crystallized and Fluid Intelligence Psychologists working from the psychometric approach examine performance on a variety of tasks to examine how intelligence is organized. Crystallized intelligence is the accumulated
Section 10.2Theories of Intelligence
knowledge and ability generally acquired from one’s culture and from formal instruction (Cattell, 1987). Tasks that assess vocabulary, general knowledge, and comprehension tend to be interrelated (grouped) and comprise the crystallized intelligence factor. Crystalized intel- ligence is typically assessed with verbal items that measure one’s store of knowledge (for example, “What is the state capital of Idaho?,” “What does the word democracy mean?,” and so forth).
Fluid intelligence, on the other hand, involves reasoning and problem solv- ing, particularly on unfamiliar prob- lems that require efficient and in-the- moment inferences and solutions (Cattell, 1987). Completing patterns of matrices is a task that draws on fluid intelligence (see the right-hand col- umn in Figure 10.1). Drawing an infer- ence from a series of relationships— for instance, reasoning that the next number in the series 1, 2, 4, 8, 16 ___ is “32”— also illustrates fluid intelli- gence. The answer is derived by rea- soning that the next number in the sequence is the result of multiplying the previous number by 2 (1 × 2 = 2; 2 × 2 = 4; 4 × 2 = 8, etc.). Because mea- surements of fluid intelligence focus on novel problems, they are assumed to rely less on prior learning than do measures of crystallized intelligence (Nisbett et al., 2012).
Consider an example that will illustrate both positive manifold and the different intellectual abilities we have just discussed. Imagine that William is a fourth grader who plays with puz- zles and video games that require reasoning skills. He does reasonably well at these games. However, William enjoys reading about dinosaurs most of all. In fact, he is reading books on the topic that are well above his grade level. Based on the concept of positive manifold, we can predict that if William scores above average on tasks that assess reasoning (fluid intelli- gence), he will probably also score above average on other tasks that assess reading compre- hension (crystallized intelligence). This is positive manifold.
We would not be surprised, however, to also learn that given William’s enthusiasm for read- ing about dinosaurs, he scored very high on verbal tests that measure reading and vocabulary (crystallized intelligence) while scoring lower (but still above average) on the measurements of nonverbal reasoning (fluid intelligence). In other words, crystallized and fluid intelligence are related (positive manifold), but also distinct and separable.
The psychometric approach has been refined in recent years by uncovering additional broad intellectual abilities that emerge from analyses of large data sets of intelligence test scores (Carroll, 1993; Horn & Blankson, 2005). Like crystallized and fluid intelligence, each broad category is made up of cognitive tasks that share more in common with each other than with tasks in the other categories.
age fotostock/SuperStock Fluid intelligence involves problem solving and detecting relationships, skills children use to solve puzzles.
Section 10.2Theories of Intelligence
These domains of intellectual abilities are organized hierarchically in what is known as the three-stratum model of intelligence (Carroll, 1993). The highest level of the hierarchy, g, is the most general stratum (layer) of intelligence (see Figure 10.3). The next broadest stratum is characterized by eight intellectual abilities (see Table 10.2). The third and most narrow layer of intelligence includes a number of specific abilities; for instance, memory for sound patterns, foreign language proficiency, and reading comprehension (see Flanagan et al., 2013 for a review).
Figure 10.3: The three-stratum model of intelligence
In the three-stratum model of intelligence, general intelligence, g, is at the top of the hierarchy. Broad abilities related to g are at the second tier. The broad abilities are reflected in specific abilities at the third tier.
General intelligence (g)
General memory and
learning
Broad visual
perception
Broad auditory
perception
Broad retrieval capacity
Broad cognitive
speediness
Processing speed
(Decision speed)
Crystallized intelligence
Fluid intelligence
Reasoning Vocabulary knowledge
Memory span
Detecting spatial
relations
Speech- sound
discrimination Creativity
Rate of test-taking
Reaction time
Source: From “The Three-Stratum Theory of Cognitive Abilities” by J. B. Carroll. In D. B. Flanagan, J. L. Genshaft, and P. L. Harrison (Eds.), Contemporary Intellectual Assessment: Theories, Tests, and Issues. Copyright © 1996 by Guiford Publications, Inc. Adapted with permission of the publisher.
Broad abilities
Specific abilities: Examples
Source: Adapted from Carroll, J. B. (1996). The three-stratum theory of cognitive abilities. In D. B. Flanagan, J. L. Genshaft, & P. L. Harrison (Eds.), Contemporary intellectual assessment: Theories, tests, and issues. New York: Guilford Press. Copyright © 1996 by Guilford Publications, Inc.
The three-stratum model conceptualizes intelligence as unitary: The abilities classified beneath g are interrelated and, therefore, not considered independent intelligences. The broad ability at the top of the hierarchy, g, influences second-level abilities that are more specific (such as crystallized intelligence). At the third and most specific level, an even more specialized skill is impacted by the second-level ability (which is impacted by g).
For the most widely used intelligence tests, items and subtests generally correspond to the abilities outlined in the three-stratum model (Flanagan et al., 2013; Gottfredson, 2004a). For instance, the Stanford-Binet (fifth edition) measures five broad abilities that correspond to abilities in the three-stratum model (Roid, 2003): fluid reasoning, crystallized knowledge, quantitative reasoning, visual-spatial reasoning, and working memory.
Section 10.2Theories of Intelligence
Table 10.2: Eight broad intellectual abilities in the three-stratum model of intelligence
Ability Description
Fluid intelligence Using reasoning and inference to identify relationships and solve novel problems; includes quantitative reasoning
Crystallized intelligence Cultural knowledge pertaining to concepts, vocabulary, and other infor- mation typically derived from experience and education
General memory and learning The limited-capacity system for maintaining immediately relevant infor- mation during a short period
Broad visual perception Transforming and spatially orienting shapes, figures, and other visual information
Broad auditory perception Discriminating, analyzing, and manipulating patterns of sound (for example, music, speech).
Broad retrieval ability Consolidating new information and retrieving it from long-term memory when cued; for instance, fluently naming examples of a category or remembering ideas, names, and other facts
Broad cognitive speediness Efficiency in performing familiar cognitive tasks that require attention (for example, reading)
Processing speed Reaction time when presented with simple tasks such as matching stimuli
Source: Adapted from McGrew, K. S. (2009). CHC theory and the human cognitive abilities project: Standing on the shoulders of the giants of psychometric intelligence research. Intelligence, 37(1), 1–10.
Intelligence Outside of the Classroom As we indicated at the outset of this chapter, definitions of intelligence typically center on cognitive abilities associated with success in school, such as reasoning, comprehending com- plex ideas, and learning from experience. These cognitive abilities can also apply to situa- tions and problems outside of formal schooling. For instance, as we discussed in the theory of mind chapter (Chapter 6), complex reasoning is employed when we infer the thoughts and intentions of others in social situations. Everyday phrases like “street smart” capture this broader notion of intelligence, extending it beyond school performance. In fact, adults often view social competence as an important aspect of intelligence (Silvera, Martinussen, & Dahl, 2001; Sternberg, Conway, Ketron, & Bernstein, 1981).
Following from these observations, broader definitions more explicitly associate intelligence with abilities that are useful in contexts outside of the classroom. One prominent researcher on the subject defines successful intelligence as the “ability to choose and successfully work toward the attainment of one’s goals in life, within one’s cultural context or contexts” (Stern- berg, 2014, p. 209). In this definition, intelligence is identified by behaviors and outcomes that are particularly adaptive and valued in one’s culture. Although basic cognitive processes like memory and attention may be important components of intelligence across different cul- tures, the type of problems that draw on those resources varies according to culture and con- text (Kranzler & Floyd, 2013; Sternberg, 2014).
Section 10.2Theories of Intelligence
For instance, Luo children in a village in Kenya who were identified as particularly competent were actually removed from formal schooling in order to serve in an apprenticeship (Stern- berg, 2014). Intelligence in that culture was not closely tied to the type of problem solving typically encountered in school. Practical knowledge valued in the community involved prob- lems such as distinguishing among different herbal medicines to treat parasitic illnesses. If the skills this culture considered valuable and adaptive were measured in a written intelli- gence test, the questions might look like the following hypothetical item:
Sample Item Measuring Successful Intelligence in a Rural Kenyan Village
“A small child in your family has homa. She has a sore throat, headache, and fever. She has been sick for 3 days. Which of the following five Yadh nyaluo (Luo herbal medicines) can treat homa?
i. Chamama. Take the leaf and fito (sniff medicine up the nose to sneeze out illness).∗
ii. Kaladali. Take the leaves, drink, and fito.∗
iii. Obuo. Take the leaves and fito.∗
iv. Ogaka. Take the roots, pound, and drink.
v. Ahundo. Take the leaves and fito.”
*Correct answers (Sternberg, 2014)
Situating intelligence within a social context is related to social constructivist views of cogni- tion. Recall that within a zone of proximal development, cognitive performance is measured as it unfolds in the context of social guidance (Vygotsky, 1978a). In contrast, intelligence test- ing in Western societies is typically decontextualized, meaning that the setting is neutral and strictly limits any influence from the tester.
However, Vygotsky (1978a) claimed that guidance from a tester could help draw out intel- lectual potential that would otherwise remain hidden in a testing situation. In reaction to this concern, he engaged in dynamic testing, a flexible assessment in which feedback, social prompts, and support are provided to the child during testing (Murphy, 2011).
To illustrate, Vygotsky wrote that testers “might run through an entire demonstration and ask the children to repeat it, others might initiate the solution and ask the child to finish it, or offer leading questions” (Vygotsky, 1978a, pp. 85–86).
The interplay between tester and child is the strength of dynamic testing because it pro- vides information about a child’s potential to benefit from experience. This interplay is also a weakness, however, from a practical standpoint. Drawbacks of dynamic testing include its time-intensive nature and its relatively high cost in terms of training the testers (Murphy, 2011). Consequently, it is used relatively infrequently (Fuchs, Compton, Fuchs, Bouton, & Caffrey, 2011).
Section 10.2Theories of Intelligence
Dynamic testing is of theoretical interest, however, because it can provide direct evidence of the child’s ability to learn when provided with assistance (Sternberg & Grigorenko, 2006). Recall that learning from experience is a feature of intelligence as per our definition found at the outset of the chapter. Practically speaking, dynamic testing seems most compatible, at present, with attempts to assess children who are underperforming and who do not appear to be exhibiting their intellectual potential within more conventional ways of testing and assess- ment (Murphy, 2011).
Also differing from the psychometric approach, Howard Gardner theorizes that intelligence is not a general ability captured by a single score. The theory of multiple intelligences pro- poses that separate core abilities and processes operate to “solve problems or create prod- ucts that are of value in a culture” (Gardner, 1999, p. 34).
In this theory there are many domains of intelligence extending beyond the analytical and problem-solving abilities typically associated with school performance. Instead, Gardner asks us to consider how intelligence operates in a wide variety of pursuits and contexts, rang- ing from activities like playing a musical instrument, competing in athletic competitions, or playing a game of chess (Gardner, 1983, 2006). By tying intelligence to culture and values, Gardner’s theory highlights the importance of opportunities in the environment to nurture intelligence (Gardner, 1999). If, for instance, virtuosity in music is not particularly valued in a culture, a child’s potential in that particular domain of intelligence will often go unrealized.
Altogether, Gardner has identified eight separate intelligences (see Table 10.3). The intelli- gences are, to some extent, distinct from one another. This means we might expect uneven- ness in ability; for instance, a performing artist might exhibit exceptional musical and bodily-kinesthetic intelligence while other intelligences are much less advanced (Gardner & Moran, 2006).
Table 10.3: Gardner’s theory of multiple intelligences
Intelligence Characteristic core components
Linguistic Sensitivity to language and effective use of it; the ability to learn a new language
Logical–mathematical Carrying out mathematical and logical operations; scientific thinking
Spatial Recognizing and manipulating visual patterns
Bodily–kinesthetic Using one’s body for creative expression or problem solving
Musical Producing and appreciating musical components like pitch and rhythm
Interpersonal Discerning mental states of others and responding by working effectively with others
Intrapersonal Apprehending one’s own thoughts and feelings to guide behavior; knowl- edge of one’s own strengths and weaknesses
Naturalistic The ability to recognize and classify animate and inanimate objects
Source: Gardner, 1999.
Section 10.2Theories of Intelligence
Gardner’s theory is derived from evidence gathered from a variety of sources beyond the IQ and experimental tasks that make up the psychometric approach. For instance, he examined patterns of evidence from brain-damaged individuals indicating a particular cognitive fac- ulty was localized in the brain (Gardner, 1999). The localization of function is evident when damage to a specific part of the brain impairs a specific cognitive function but spares other cognitive abilities. Damage to particular areas of the left hemisphere specifically impairs lan- guage, suggesting to Gardner that humans have a separate linguistic intelligence (Gardner, 1999). Essentially, evidence tying specific cognitive faculties to specific areas of the brain is counted as evidence for an intelligence.
Evidence from prodigies also informs the theory of multiple intelligences. A prodigy is some- one with exceptional, rare ability in a specific area of expertise, typically coupled with at least average general intellectual ability (Gardner, 1999). An example of a child prodigy’s accom- plishments might be committing 100 pages of music to memory and playing at Carnegie Hall before kindergarten (Ruthsatz, & Urbach, 2012). The uniqueness of the extraordinary ability suggests it is independent from the other, more typical, intellectual abilities of the prodigy. A prodigy in an area like math or music allows us to observe “a particular intelligence in sharp relief ” (Gardner, 1999, p. 39).
Gardner’s theory reminds us that emphasis in school curricula on cognitive abilities like math and language may neglect cognitive capacities such as interpersonal and musical intel- ligences that are also valuable in society (Gardner, 1995). In addition, Gardner’s theory draws attention to the possibility that education might become more personalized and less uni- form once we broaden our definition of intelligence (Gardner, 1999). In this view, accounting for differences in the ways children process information and learn—acknowledging there are multiple intelligences—calls for individualized approaches to teaching and assessment (Gardner, 2009).
Although the theory of multiple intelligences is thought-provoking in expanding our ideas of intelligence, it has been criticized on empirical grounds. For instance, in a study with adults, many of the intelligences with higher order cognitive requirements (for example, linguis- tic, logical, interpersonal, spatial) were intercorrelated and appeared to reflect a common, g factor of intelligence (Visser, Ashton, & Vernon, 2006). This evidence raises the question of whether there are really multiple intelligences or just one g that impacts performance on multiple tasks.
To date, theories that extend the concept of intelligence beyond traditional academic set- tings have not produced measurements that have the same practical utility as conventional intelligence tests that measure cognitive abilities closely linked to school performance (Gott- fredson, 2004a). Abilities like bodily-kinesthetic, musical, and intrapersonal intelligence do not readily translate to the conventional approach, in which children answer questions that are clearly quantified as right or wrong. For instance, measuring musical intelligence might necessitate assessing activities, such as singing a new melody, that do not immediately lend themselves to an objective right or wrong response (Gardner, 1995).
Section 10.3Heredity, Environment, and Intelligence
Questions to Consider
1. How do the three theoretical approaches that characterize intelligence—psycho- metric, multiple intelligences, and successful intelligence—compare and contrast to one another?
2. If you were given the opportunity to design an elementary school curriculum (including teaching materials, lesson plans, and content) that was inf luenced by Gardner’s theory of multiple intelligences, what are some possible differences that might emerge in comparison to traditional school curricula?
10.3 Heredity, Environment, and Intelligence The relationship of IQ to outcomes like school performance, job performance, and health highlight the relevance of asking why individual differences in intelligence emerge in the first place (Neisser et al., 1996). Ideally, answers to the question will provide information we can use to reduce disparities. A theme highlighted throughout this book is the role heredity (nature) and environment (nurture) jointly play in influencing cognitive development. This theme guides our discussion in this section. We will attempt to clarify how the contributions of genes to individual differences are quantified, how genes and environment interact to pro- duce differences in intelligence, and how the nature/nurture theme applies to possible group differences in IQ test performance.
Twin Studies Efforts to understand and quantify the different contributions of genes and environments to differences in intelligence are greatly advanced by studies of twins. In contemporary research, twin studies are sometimes very large scale and involve thousands of twin pairings (Haworth et al., 2010). In this section we will first address why researchers would go to such great lengths to study thousands of twins. What do twins tell us about the nature/nurture theme? Second, we will summarize the results of twin studies in the context of intelligence differ- ences. Hopefully, the more we understand the sources of intelligence differences, the better we can design methods to intervene and optimize the growth of individuals’ intelligence.
At the beginning of modern intelligence testing, Francis Galton, a half-cousin of Charles Dar- win, traced the lineage of famous individuals in Great Britain (Galton, 1908). He observed that prominent, accomplished individuals were often genetically related to other prominent indi- viduals. Many times we make similar observations in our own lives. Teachers, for instance, might observe that the classroom performance of younger siblings often bears some resem- blance to their older siblings.
Galton inferred that success “runs in families” because intelligence is inherited. An alternative explanation, of course, appeals to the similarities in cultural and economic background often shared by relatives. Psychological resemblances among relatives could, in short, be plausibly explained by genetic similarities and/or by social-environmental similarities.
Section 10.3Heredity, Environment, and Intelligence
Modern twin studies originated in the 1920s and help clarify the influences of nature and nurture on intelligence (Rende, Plomin, & Vandenberg, 1990). They compare the correspondence between identical (monozygotic) twins with the correspondence between fraternal (dizygotic) twins on the same measurement. Monozygotic (MZ) twins are genetically identical while dizygotic (DZ) twins share, on average, 50% of their genes. These dif- ferences occur because DZ (fraternal) twins come from two separate fertil- ized eggs, whereas MZ (identical) twins develop from a single fertilized egg that has split in two.
If identical twins are more closely related in intelligence than fraternal twins, a genetic influence is inferred.
The greater relationship between identical twins is attributed to their greater genetic simi- larity. The results of a number of seminal twin studies indicate that IQ scores of identical twins are more closely related than are IQ scores of fraternal twins, which is evidence that IQ is influenced by genetic factors (Bouchard & McGue, 1981). Because of such results, the exis- tence of a relationship between genetic and intellectual differences is generally considered well established (Nisbett et al., 2012).
A common concern about twin studies is the possibility that identical twins are treated more similarly by others than their fraternal twin counterparts (Miller, 2012a). Perhaps the physi- cal similarities of the identical twins leads to similar treatment, which then leads to stronger correlations compared to fraternal twins. One way to address this concern is to examine iden- tical twins who were adopted into different homes.
Many studies have shown that IQ scores are more strongly correlated for identical (MZ) twins reared in different homes compared to fraternal (DZ) twins reared together in the same home (Bouchard & McGue, 1981). This indicates that even when identical twins grow up in different homes, they are still related to one another in terms of their IQ scores. Because the magnitude of the correlation exceeds the correlation for fraternal twins raised in the same home, we again see evidence for the influence of nature on intellectual differences. Twins who share all of their genes but do not share a home are more closely related (in terms of IQ scores) than twins who share a home but do not share all of their genes.
The genetic influence on intelligence reminds us that children bring to each environment their own unique combination of skills and interests. Although some children may thrive in a particular environment or under a particular method of instruction, other children may not benefit to the same extent. As we see in the next section, the influence of the environment depends on the influence of the genes (and vice versa).
Creatas Images/Creatas/Thinkstock Identical twins share 100 percent of their genes, whereas fraternal twins share only about 50 percent. Twin studies help researchers determine the extent that genetic similarities are related to intellectual similarities.
Section 10.3Heredity, Environment, and Intelligence
Heritability A teacher looking out on a classroom of students will notice a variety of differences in hair color, height, and weight. The children also, of course, differ psychologically. Heritability refers to the extent the differences among the children are related to their genetic differences. More formally, heritability is the proportion of the variance within a population that is related to genetic differences (Nisbett et al., 2012). In the context of cognitive development, heritabil- ity estimates the extent to which psychological differences within the group relate to genetic differences.
Heritability is quantified through calculations yielding a coefficient ranging from 0 to 1. The higher the number, the more heritable is the trait being measured. Estimates of the heritabil- ity of IQ range between .4 and .8 across many studies (Nisbett et al., 2012). This means that some, but not all, of the observed differences in intelligence within a population are related to genetic differences.
The heritability of intelligence is often misinterpreted to mean that differences in IQ are unalterably fixed (Block, 1995). This misinterpretation stems from confusing heritability with inheritability (Block, 1995; Keller, 2010; Moore, 2013b). The concepts are very differ- ent! Inheritance refers to an individual’s genetic material. Heritability is a measurement of differences within a population, a statistic that does not tell us anything about the DNA one inherits. An individual’s genetic material—inheritance—does not change because of the envi- ronment. However, the reasons why people within a group differ from one another—heri- tability—can change. In fact, intelligence can become more or less heritable depending on culture and circumstance.
We can illustrate how heritability changes by discussing the association between heritabil- ity and a child’s age (Briley & Tucker-Drob, 2013). Specifically, the heritability of intelligence increases from infancy to young adulthood (Haworth et al., 2010). In other words, during development, individual differences in intelligence become more strongly related to genetic differences (based on the results of twin studies). A common explanation for this finding illus- trates how environment and genes work together to influence intelligence.
In particular, many theories propose that heritability increases with age because of the trans- actional nature of development (Haworth et al., 2010; Trzaskowski et al., 2013; Tucker-Drob, Briley, & Harden, 2013). This means that features related to one’s genes change one’s environ- ment, and the environment impacts features that are influenced by genes (Sameroff, 2010). More succinctly, it means that nature affects nurture and nurture affects nature. Differences in the environment can impact differences in intelligence, even though intelligence is a heri- table characteristic (Eisenberg, 2004).
For instance, two children may possess slight genetic differences that are reflected in differ- ent IQ scores measured during their early childhood. By middle childhood, the differences in their environments may become magnified as the children choose different levels of aca- demic challenges, different friends, different extracurricular activities, and so on (Briley & Tucker-Drob, 2013). From this perspective, children select their own environments follow- ing, at least in part, from their genetic propensities. As a consequence of these choices, the differences in intelligence test performance between the children become multiplied because of their different environmental experiences (Dickens & Flynn, 2001).
Section 10.3Heredity, Environment, and Intelligence
An example from outside of intelligence testing will help clarify the transactional nature of development. Imagine two children differ in height, and that one is very tall for her age. These differences relate to the children’s genetic differences. Parents of the tall child respond by encouraging her to develop skills in basketball. The other child, in contrast, may have an interest in basketball but does not receive the same encouragement and training. These envi- ronmental differences magnify the differences in basketball skills between the two children. Nature (height) influenced nurture (environment), and then nurture amplified the differ- ences in basketball skills.
This discussion of heritability has implications for people who work with children. First and foremost, it helps us see that even though intelligence is heritable, it is not a fixed trait in the sense that eye color or other characteristics are fixed by DNA. Differences in intelligence are the product of both environmental and genetic differences that work together. Intelligence is not simply a characteristic that impacts how much a child will learn and succeed in school. It is also the product, or the outcome, of educational experiences both inside and outside of school (Martinez, 2000).
Moreover, the transactional view of development reminds us that children are active par- ticipants in their learning as they choose and select specific opportunities and environments (Scarr & McCartney, 1983). It may be the case that individual differences, related to some extent to genetic differences, predispose children to pursue certain educational activities with greater interest than other activities (Krapohl et al., 2014). A uniform, one-size-fits-all approach to education is somewhat ill-suited to children’s individual differences. Perhaps with technological advances, such as educational software and individualized means of infor- mation delivery, education can become increasingly personalized to reflect children’s indi- vidual differences (Krapohl et al., 2014).
Group Differences Differences in cognitive performance occur between the sexes and between races. Before dis- cussing these differences, we point out that there is substantial overlap between groups on cognitive task performance. In other words, if group A has a higher average score than group B, it is still the case that many individuals in group B outperform many individuals in group A. Group membership is not a basis for inferring an individual’s capabilities. As we will see, stereotypes are not only unfair—they can also adversely impact cognitive performance.
Race and IQ When IQ differences arise between groups, a question immediately arises: What do these differences mean? Put another way, how should we interpret group differences? Address- ing these questions will help us understand, and critique, highly controversial proposals that public policy should be influenced by group differences in IQ test scores.
For instance, when someone claims that immigration policy in the United States should restrict the admittance of groups with purportedly below-average IQ scores, an effective, informed response would include an understanding of what the data on group differences mean (and what they do not mean) (Richwine, 2009). Aside from the ethical problems associ- ated with discriminating between groups, one would also need to be prepared to point out
Section 10.3Heredity, Environment, and Intelligence
flawed interpretations of data. To foreshadow our discussion, the data do not support making decisions about groups based on group differences in IQ scores.
Data indicate IQ scores of African Americans (AA) tend to be lower than the scores of Euro- pean Americans (EA). Summaries of the data have placed the difference at about 15 points (Neisser et al., 1996; Rushton & Jensen, 2006). Others have found the 15 point gap is decreas- ing, with AA gains of about 4 to 7 points since the 1970s (Dickens & Flynn, 2006). We will use these AA and EA differences to address the question central to this section—how do we interpret group differences in IQ?
Our first consideration when interpreting findings of racial differences is that for many scien- tists, the very concept of “race” is fuzzy. This is because most genetic variation among individ- uals exists within racial categories and not between them (Sapp, 2012; Sternberg, Girgonko, & Kidd, 2005). In essence, members of the same race are, genetically, very diverse on aver- age. Generally speaking, people within racial categories are vastly different from one another and the biological boundaries between races are difficult to define and are subject to debate (Sternberg et al., 2005).
However, the concept of race is a social reality in the United States, and related to differences in socioeconomic conditions (Fryer, 2011; Sapp, 2012). Given social realities, the topic of race differences in IQ continues to be studied, and the findings need to be properly interpreted.
Consistent with our nature/nurture theme, researchers tend to explain the group differences by emphasizing either biological or environmental influences (Rushton & Jensen, 2006; Nis- bett, 2009). Two extensive summaries of the research literature on intelligence have con- cluded that while the issue remains unsettled, the weight of evidence points to the impor- tance of environmental differences in understanding the reasons for the gap in IQ scores (Neisser et al., 1996; Nisbett et al., 2012). Thus, we see a preliminary answer to our question; in particular, when racial differences occur, the evidence suggests we look to cultural and environmental differences for explanations.
One clear strand of evidence showing the importance of the environment comes from the relationship between social change and IQ change for African Americans. The gains in IQ scores among African Americans since the 1970s indicate that some cultural/environmental factors (for example, progress in school funding that benefits minority groups) have posi- tively impacted their intelligence test performance (Dickens & Flynn, 2006). We will see in the next section how environmental factors may produce positive change in IQ for other groups as well.
Another strand of evidence occurs when we pinpoint counterproductive features in the envi- ronment that can artificially lower IQ scores. One adverse environmental factor is the influ- ence of negative racial stereotypes on test performance among minorities (Walton & Spen- cer, 2009). Stereotype threat occurs when awareness of a negative stereotype (for example, “Black people are not intelligent”) causes anxiety and stress when completing a test or mea- surement related to the stereotype (Steele, 1997). These negative feelings in turn interfere with standardized test performance (Good, Aronson, & Inzlicht, 2003).
Section 10.3Heredity, Environment, and Intelligence
Real-World Application: Reducing Stereotype Threat Scientific findings can helpfully close gaps in cognitive achievements between groups. The discovery of stereotype threat in the mid-1990s spurred interventions in the classroom. As the linked article indicates, results of these interventions are encouraging.
http://www.apa.org/monitor/2011/09/achievement.aspx
Critical-Thinking Question
What other classroom exercises can you think of that might protect children from stereotype threat?
Finally, and most directly, we can compare individuals of substantial, if not 100%, African ancestry, with individuals that possess a combination of African and European ancestry. In particular, we can examine whether IQ scores are associated with such racial differences. If race differences in IQ are genetic in origin, then IQ scores among African Americans should rise and fall depending on the degree of their European ancestry. Data indicate that the rela- tionship between IQ and the degree of European ancestry among African Americans is negli- gible (Nisbett, 2005). Again, the weight of evidence indicates that we should look to environ- mental factors rather than genetic ones when interpreting race differences in IQ.
Cognitive Sex Differences As with racial differences, the study of cognitive sex differences can help identify factors that potentially create inequalities, which in turn informs attempts to improve children’s cogni- tive performance (Halpern, 2013). As we investigate this topic, keep in mind that we are not investigating which sex is smarter or better—instead, research asks to what extent, and under what circumstances, differences might exist (Halpern, 2013).
When IQ tests are constructed, items that differentiate between the sexes are eliminated so that there are no sex differences in average IQ scores (Halpern, 2013; Halpern & LaMay, 2000). However, as we discussed earlier, general intelligence consists of more specific cogni- tive abilities (such as verbal and nonverbal spatial reasoning subscales). Decades of psycho- logical research suggests four cognitive sex differences at this more specific level of cognition (Miller & Halpern, 2014). These differences are listed in Table 10.4.
Precise explanations that investigate the interplay of nature and nurture on these differences continue to be sought by investigators (Halpern, 2013; Miller & Halpern, 2014). In this sec- tion, we will discuss some preliminary evidence in the areas of math and spatial reasoning to provide a sense of the types of explanations and evidence being sought.
Section 10.3Heredity, Environment, and Intelligence
Table 10.4: Cognitive sex differences suggested by research evidence
Area Findings
Math • Males are more likely than females to score exceptionally high (for example, top 1%) on measures of mathematical ability.
• Averaging across the entire range of scores on measures of mathematical ability, males have a small to negligible advantage.
Memory • Females outperform males in some areas of memory, such as face recognition and object location.
Spatial • Males score higher than females on some measures of spatial reasoning, such as mental rotation tasks.
Verbal • Females consistently outperform males on verbal measures (for example, reading achievement and writing).
Source: Adapted from Miller, D. I., & Halpern, D. F. (2014). The new science of cognitive sex differences. Trends in Cognitive Sciences, 18(1), 37–45.
The influence of the environment is evident because cognitive sex differences are changing in some societies. For instance, in the United States the ratio of males to females exhibiting exceptional mathematical abilities has decreased from 13:1 in the 1970s and 1980s to 4 (or less):1 in more recent years (Miller & Halpern, 2014). This means there is a greater represen- tation of females among top performers than in previous decades.
Cultural and educational changes that place a greater emphasis on gender equity are presum- ably responsible for females’ gains on assessments of mathematical ability (Kane & Mertz, 2012). For example, cross-cultural research indicates that providing equal access to edu- cation reduces sex differences in math performance. To illustrate, 8th grade boys and girls tended to perform similarly on a measurement of math achievement in nations where there was equal access to formal education, and less similarly in nations where females’ access was more limited (Else-Quest, Hyde, & Linn, 2010).
As with racial differences, negative stereotypes (“girls are not good at math”) can interfere with performance on math assessments (Shapiro & Williams, 2012). In general, anxiety about cognitive tests, including IQ tests, can impair performance (Hembree, 1988; Hopko, Critten- don, Grant, & Wilson, 2005). Stereotyped beliefs may cause anxieties that in turn negatively impact performance (Maloney, Schaeffer, & Beilock, 2013).
Math anxiety is anxiety that sets in when a person thinks about or performs math problems. A link between math anxiety and math performance is evident as early as first grade and can occur throughout schooling (Hembree, 1990; Ramirez, Gunderson, Levine, & Beilock, 2013). Both males and females experience math anxiety, although females tend to display higher levels in some studies (Beilock, Gunderson, Ramirez, & Levine, 2010; Else-Quest et al., 2010; Hembree, 1990).
Math anxiety and stereotype threat may share the adverse effect of impairing working memory performance during testing (Maloney et al., 2013). Negative thoughts about doing poorly can take up working memory capacity that might otherwise be devoted to problem solving. Interventions to reduce intrusive thoughts that accompany anxiety can be helpful in this regard.
Section 10.3Heredity, Environment, and Intelligence
For instance, in one study, the negative impact of test anxiety was diminished when 9th graders were given the opportunity to express their worries in writing about an upcoming high-stakes exam (Ramirez & Beilock, 2011). The writing intervention took place just minutes before the exam and afforded students an opportunity to express and process negative thoughts, pos- sibly freeing working memory capacity during the exam (Ramirez & Beilock, 2011).
The joint influence of nature and nurture is evident in research investigating sex differences in mental rotation. Explanations for sex differences include the effects of hormones, matura- tional rates of the brain, experiences, and task factors (such as time limits) that may be dis- advantageous for females (Hyde, 2014; Moore & Johnson, 2008). To illustrate the interplay of nature and nurture in understanding sex differences in mental rotation, we look at studies of the sex hormone androgen, which stimulates and maintains the development of masculine characteristics. Before birth, androgen may play a role in brain organization (Wallen, 2009).
Female adolescents and young adults (ages 16 to 30) who were exposed to higher than nor- mal levels of androgen before birth (because of a genetic disorder) outperformed typically developing females on measures of mental rotation like the one illustrated in Figure 10.4 (Berenbaum, Bryk, & Beltz, 2012). However, research does not always find a link between elevated levels of prenatal androgen and spatial abilities, and the relationship continues to be investigated (Malouf, Migeon, Carson, Petrucci, & Wisniewski, 2005).
Figure 10.4: A test of mental rotation skills
On average, boys outperform girls on tests of mental rotation like the one shown here.
Source: Adapted from Vandenberg, S. G., & Kuse, A. R. (1978). Mental rotations, a group test of three-dimensional spatial visualization. Perceptual and Motor Skills, 47(2), 599–604.
Theories highlighting the transactional nature of development draw on findings that early excess androgen exposure in females is related to play behaviors like climbing and playing with tool sets, behaviors more commonly associated with males (Berenbaum et al., 2012; Pas- terski et al., 2005). Theoretically, these behaviors positively impact spatial development by giving children enriched experiences with spatial information (Beltz, Swanson, & Berenbaum, 2011; Berenbaum et al., 2012; Golombok & Rust, 1993). We see in this example how nature and nurture can have reciprocal effects.
Section 10.4IQ: Stability and Modifiability
Questions to Consider
1. In what ways could findings regarding group differences be applied to the classroom?
2. How might the transactional nature of development help explain the improvement in IQ scores among African Americans over successive generations?
10.4 IQ: Stability and Modifiability Returning to our case study that opened this chapter, if a young child scores poorly on an intelligence test, does this mean the child will continue to have low intelligence scores into adulthood? Or will intelligence change? This section considers these questions. When using the terms stability and change, we are referring to the position of one’s IQ scores relative to same-age peers over time.
Our concerns in this section are (a) whether there is stability in IQ scores and (b) what envi- ronmental interventions might modify or change IQ over time. These concerns are, naturally, motivated by the desire to ensure that we optimize children’s development. To the extent change is possible, it is useful to know which interventions show particular promise. In addi- tion, the stability of IQ is an important consideration for educators when IQ scores are used as evidence for a child’s placement in a particular educational program. For instance, if a young child scores poorly on an IQ test and receives special education services, can we assume the poor IQ score is sufficiently stable so that the child does not need to be retested at a later date?
Development and Stability in IQ Recall that we distinguished earlier in the chapter between test reliability and the stability of intelligence. Test reliability is consistency in performance typically measured over a period of days or weeks. If intelligence is a stable characteristic of the individual, a person’s score on an intelligence test should remain about the same over a prolonged period. By “the same,” we mean that a person’s rank order in comparison to peers should be relatively stable. In other words, someone who is above or below average in early childhood would remain above or below average at later ages if IQ is stable. By “prolonged period” we are referring to a period measured in years.
At the outset of this section we point out two generalizations with regard to stability and IQ. First, the shorter the interval between test and retest, the greater the stability in IQ scores (Schneider, Niklas, & Schmiedeler, 2014). For example, there is typically less change in IQ between ages 7 and 9 than there is between a longer interval like between ages 7 and 11. Second, the older the child when first tested, the more stability his or her scores exhibit upon retest (Neisser et al., 1996; Schneider et al., 2014; Moffitt, Caspi, Harkness, & Silva, 1993).
We begin investigating stability in intelligence by reaching back to the very beginnings of development. In particular, we look at whether there are measureable differences
Section 10.4IQ: Stability and Modifiability
in intelligence in infancy and if those differences remain stable over time. The Bayley Scales of Infant and Tod- dler Development measure motor skills, language, and cognitive/percep- tual abilities in infants and toddlers ages 1 to 42 months (Bayley, 2006). In the most recent version of the Bayley Scales, items include object and sound recognition, direction following, and fine and gross motor skills such as sit- ting upright or grasping objects.
Because the latest version of the assessment is relatively new, findings about long-term stability in intelli- gence involve earlier versions of the measurement. For instance, an earlier
version of the Bayley Scales was administered to 3-month-olds and was positively related to later IQ at age 6 (Wilson, 1978). In general, however, evidence that individual differences on the Bayley Scales predict later IQ differences is mixed (Colombo, 1993; Hack et al., 2005; Lut- tikhuizen dos Santos, de Kieviet, Königs, Van Elburg, & Oosterlaan, 2013).
More conclusive are those studies that measure infant intelligence by drawing on IP theory. In IP theory the amount of time it takes an infant to habituate to a stimulus is used as a mea- sure of early intelligence (Bornstein, Hahn, & Wolke, 2013; Rose & Feldman, 1995). The rea- soning is that infants who quickly and efficiently process information will need fewer trials than other infants to recognize that a stimulus is familiar and habituate to it. Remember that IP theory emphasizes continuity in development. Consequently, the ability to efficiently and accurately encode information into memory should continue to influence cognitive perfor- mance after infancy (McCall & Carriger, 1993).
For example, in one study, the efficiency with which infants habituated to a stimulus was mod- erately correlated eleven years later with performance on many subtests of the Wechsler test (Rose & Feldman, 1995). Other studies find continuity between infant information processing and IQ extending into adulthood (Fagan, Holland, & Wheeler, 2007). Such findings indicate long-term continuity and indicate that infant cognition contributes to later intelligence test performance (Rose, Feldman, Jankowski, & Van Rossem, 2012).
It is important to note that these correlations are modest in magnitude, and it would be a mis- take to assume that individual differences in intelligence do not change from infancy onward. The correlations are noteworthy, however, given the amount of time they encompass and the differences between measuring intelligence with looking time and with paper-and-pencil IQ assessments.
Generally, beginning around age 6 or 7, IQ scores begin to stabilize and correlate more sub- stantially with later assessments taken in childhood and adolescence (Keage et al., 2015; Schneider et al., 2014). To illustrate, in one study intelligence was measured at ages 2, 4, 7, and 11 to 13. Age 7 was the youngest age at which IQ scores predicted the number of years
Monkeybusinessimages/iStock/Thinkstock IQ test scores tend to be more stable (a) the shorter the time between the first test and retest and (b) the older the child at the time of the first test.
Section 10.4IQ: Stability and Modifiability
of schooling eventually completed by the participants (Keage et al., 2015). Long-term studies show that childhood IQ remains somewhat stable even after a nearly 70-year gap between tests (Deary, Whiteman, Starr, Whalley, & Fox, 2004).
In our case study at the outset of this chapter, Michael’s age (6 years) is young enough that it should give us pause and lessen our confidence that his IQ score is an exceptionally strong indicator of future IQ scores. An additional reason for caution is evidence that IQ scores some- times substantially change over time. To illustrate, in one study children who were tested for possible placement in special education services with the WISC-IV were tested again approxi- mately 3 years later. Although the overall stability, encompassing the entire group of children, was relatively high, a closer look revealed occasional instability in individual scores. Approxi- mately 25% of the children’s scores differed by 10 or more points between test sessions (Wat- kins & Smith, 2013).
Comparable changes in IQ scores over time were also observed in a nonclinical sample of elementary school children ages 7 to 12 tested approximately 2 years apart (Kieng, Favez, Jérôme, Geistlich, & Lecerf, 2014). In another study, a subset of preschoolers were initially classified with low intellectual abilities but their scores were well above average by age 17 (Schneider et al., 2014).
Thus, on average, the long-term stability for intelligence is relatively high, but we can also expect some individual exceptions. For this reason, clinicians and educators must be cautious in assuming long-term stability. Some researchers question the exclusive reliance on early IQ scores to warrant children’s continued placement in special education, and recommend readministering IQ tests at a later date (Watkins & Smith, 2013).
Interventions The apparent relationship between IQ scores and outcomes like academic achievement, job performance, and socioeconomic status motivates significant societal and scientific interest in whether interventions can positively impact intelligence (Gottfredson, 2004b). Evidence for the effectiveness of interventions supports the view that nurture plays a role in individual differences in intelligence.
Recall from Chapter 1 that plasticity of the brain means neural pathways can be reorganized and new neural connections can form in response to experiences. Researchers are attempt- ing to identify the types of experiences that can produce significant and long-lasting changes in intelligence. One approach is to target domain-general mechanisms like working memory that correlate with IQ (Kane & Engle, 2002).
Working memory (discussed in Chapter 3) is the ability to hold information in mind while working on cognitive tasks. It is related to general intelligence in children (Cornoldi, Orsini, Cianci, Giofrè, & Pezzuti, 2013; Giofrè, Mammarella, & Cornoldi, 2013). There are a variety of commercial programs available that claim to boost intelligence through a combination of cognitive activities that include working memory practice (a quick Internet search with key- words “brain training”, “IQ”, and “working memory” will reveal some popular programs).
Researchers have investigated whether working memory training can actually improve chil- dren’s intelligence. As discussed in Chapter 3, training typically involves repeated practice on
Section 10.4IQ: Stability and Modifiability
working memory tasks over a period of several weeks. To date, evidence for the effectiveness of working memory training on intelligence is mixed. For instance, in one study, elementary and middle school children completed working memory sessions that involved remembering whether a stimulus had appeared on a computer screen in the same location as on previ- ous trials (Jaeggi, Buschkuehl, Jonides, & Shah, 2011). Training lasted for a month, and three months later a positive effect on fluid intelligence was still evident.
However, such findings are not consistently replicated (Shipstead, Hicks, & Engle, 2012). A number of variables could impact the relationship between working memory training and intelligence, such as length and type of training as well as the age, background, and ability of the children being trained. Research continues on the question, and at this point caution is warranted when considering whether working memory training can produce noticeable long-term gains in intelligence (Shipstead, Redick, & Engle, 2012). For now, we can conclude that working memory training has the potential to positively impact intelligence, but more research is needed to identify how and when that potential is realized and how enduring and impactful such improvements might be in children’s lives.
Spotlight on Research: Music Training and IQ in Children In addition to working memory, another well-researched intervention attempt is music train- ing (Schellenberg, 2011). In one study, children ages 4 to 6 years received either visual art or music training via computerized programs (Moreno et al., 2011). The visual art group served as the control in the experiment—these children learned about concepts like shape, color, and dimension. Children receiving music training heard lessons that emphasized musical con- cepts like rhythm, pitch, and melody. Each training session was administered twice a day for 4 weeks.
Only the children who received the music training showed significant improvement on a test of verbal IQ assessing vocabulary. Also, children in the music training group showed a dif- ference in neural activity when completing an EF task compared to children in the visual art group. The distinctive pattern of neural activity is one that is associated with higher-order cog- nitive processing (Moreno et al., 2011). Thus, one possibility is that music training impacted inhibitory control, attention, and/or memory skills, which in turn led to better performance on the verbal assessment (Moreno et al., 2011).
The study did not assess whether the gains in IQ were long-lasting. And while other studies have also demonstrated cognitive enhancement resulting from music training (Fujioka, Ross, Kakigi, Pantev, & Trainor, 2006; Ho, Cheung, & Chan, 2003; Kaviani, Mirbaha, Pournaseh, & Sagen, 2014) sometimes other studies find no benefits (Mehr, Schachner, Katz, & Spelke, 2013). Overall, the present finding suggests that intelligence is potentially modifiable by experience.
Critical-Thinking Question
Does the effect of musical training on intelligence better support the psychometric theories that posit a single intelligence (g) or Gardner’s theory of multiple intelligences? Why?
There are a variety of other intriguing findings that intelligence in young children is mallea- ble (Protzko, Aronson, & Blair, 2013). Supplementing a pregnant mother’s diet or an infant’s formula with long-chain polyunsaturated fatty acids (omega-3 fatty acids) may result in the
Section 10.4IQ: Stability and Modifiability
young child’s IQ score being about 3.5 points higher than it would be otherwise. Omega-3 fatty acids support brain development, which is the hypothesized basis for the effect (Innis, 2009; Protzko et al., 2013).
To illustrate, in one study pregnant mothers were randomly assigned to receive the omega-3 supplement or a placebo. The positive effect on intelligence was evident when the children were 4 years old (Helland, Smith, Saarem, Saugstad, & Drevon, 2003). However, in another study, there was no effect by age 6, and the long-term benefits of omega-3 supplements on intel- ligence continue to be investigated (Gould, Smithers, & Makrides, 2013; Willatts et al., 2013).
Another finding is that preschool enrollment of children from low-income homes can raise IQ scores by as many as 7 points, particularly if the preschool focuses on enhancing language skills (Protzko et al., 2013). More generally, school attendance positively impacts IQ through- out childhood (Becker, Lüdtke, Trautwein, Köller, & Baumert 2012; Ceci, 1991).
As evidence of this, IQ scores tend to decline during summer months when children are away from school, particularly when children have little or no involvement in enriching academic activities over the summer (Ceci & Williams, 1997; Nisbett et al., 2012). Moreover, dropping out of school is associated with a subsequent decline in IQ scores (Ceci & Williams, 1997). Explanations for the positive effects of schooling include improving children’s vocabulary and their store of factual content information as well as engaging them in complex problem solv- ing (Protzko et al., 2013).
Additional evidence that education has an impact on IQ scores comes from the Flynn effect. The Flynn effect refers to the finding that in many parts of the world, intelligence scores have risen by perhaps 15 or more points since the 1940s (Flynn, 1999; Trahan, Stuebing, Fletcher, & Hiscock, 2014). The effect is evident from the changes in performance that have occurred in normative samples. Recall that intelligence tests are administered to large samples in order to obtain norms (that is, average scores and standard deviations).
In successive generations normative samples have shown increased intelligence test perfor- mance compared to the preceding generation. This means their raw scores are higher than the previous generation’s scores. These changes necessitate adjustments of the scoring scale of IQ tests so that the average remains 100. Common explanations for the advancement in intelligence include better nutrition and increased access to formal schooling and college (Liu, Yang, Li, Chen, & Lynn 2012; Trahan et al., 2014).
Not only has access to schooling increased in many countries, but the type of material taught in schools has also evolved. An increasing emphasis in schools in the United States on the type of problem-solving skills underlying fluid intelligence may be one reason for gains in IQ (Blair, Gamson, Thorne, & Baker, 2005).
For example, at the beginning of the 20th century, math education in grade school empha- sized rote learning and memorization (Blair et al., 2005). By the 1950s and 1960s, math edu- cation began to emphasize problem solving such as detecting visual-spatial relationships like those depicted in Figure 10.1. Today an emphasis on recognizing patterns and relationships fundamentally related to geometry is found in first- and second-grade school curricula. Simi- lar concepts would not have been addressed until seventh or eighth grade in the 1950s (Blair et al., 2005).
Section 10.4IQ: Stability and Modifiability
IQ and Poverty Over the years, a number of studies have found a relationship between particular character- istics of children’s home life and intelligence (Totsika & Sylva, 2004). Researchers commonly assess these characteristics with the Home Observation for Measurement of the Environment (HOME) (Caldwell & Bradley, 1984). As Table 10.5 indicates, physical surroundings, stimula- tion, and positive emotional support are key aspects that support positive cognitive develop- ment. The more features like those in Table 10.5 are characteristic of the child’s home, the more conducive the environment for development.
Table 10.5: Sample items on the HOME inventory for infants and toddlers (ages 0 to 3)
Item Description
Parental responsiveness • Parent spontaneously vocalized to the child at least twice during visit. • Parent responded verbally to child’s speech. • Parent caressed, kissed, or hugged child at least once. • Parent kept child in view and/or looked at child often.
Learning environment • Child has three or more books of his or her own. • Parent reads stories to child three times a week or more. • Child has several types of age-appropriate toys available. • Child’s play environment is safe.
Sources: Bradley, Corwyn, McAdoo, & García Coll, 2001; Totsika & Sylva, 2004.
In general, decades of research demonstrate a strong correlation between socioeconomic sta- tus (SES) and a child’s IQ (Bradley & Corwyn, 2002). Factors associated with low SES—such as poor nutrition, poor parental education, difficult schooling environments, early high school dropout, and family stress—are all potential contributors to low IQ scores (Blair & Raver, 2012; Hanscombe et al., 2012; Noble, Houston, Kan, & Sowell, 2012; Walker et al., 2011). The relationship between SES and IQ is supported by studies of children born in low SES circum- stances who were placed into higher SES adoptive homes. The adopted children’s IQ scores
tended to be 12-18 points higher than those of siblings remaining with birth parents or children adopted into low SES homes (Nisbett et al., 2012).
Instruction programs for low-income parents help them learn how to reduce stress in the home by institut- ing predictable routines (for example, bedtime). Another feature involves teaching parents to monitor and increase the amount of positive lan- guage (for example, praise) directed toward children in order to improve the parent–child relationship and increase the child’s confidence (Nev- ille et al., 2013).
Monkeybusinessimages/iStock/Thinkstock Parents can reduce stress at home and improve children’s confidence by implementing routines and praising their children.
Section 10.4IQ: Stability and Modifiability
Large-Scale Intervention Although interventions discussed earlier demonstrate that IQ is modifiable, at least in the short term, evidence for the long-term benefits of educational interventions on IQ perfor- mance is inconsistent (Barnett, 2011). Large-scale educational interventions are most often directed at children living in poverty. For instance, Head Start was created in 1965 as a pro- grammatic attempt to enhance and stimulate cognitive development in young children prior to formal schooling (Gibbs, Ludwig, & Miller, 2011).
Generally speaking, IQ gains from large-scale interventions in early childhood are initially evident, but then the gains tend to decrease and fade over time (Wachs, Georgieff, Cusick, & McEwen, 2014). One explanation is that risk factors associated with poverty can persist beyond early childhood (Ramey & Ramey, 1998; Walker et al., 2011). The concern is that interventions need to extend due to the fact that “over time, the initial positive effects of early interventions will diminish to the extent that there are not adequate environmental supports to maintain children’s positive attitudes and behavior and to encourage continued learning related to school” (Ramey & Ramey, 1998, p. 117).
An early intervention program that produced lasting improvement in IQ scores is the Caro- lina Abecedarian Project (Ramey et al., 2000). Children, mostly African American, who were living in poverty received pediatric healthcare and early childhood education. Family sup- port services, nutritional supplements, and transportation to the preschool were also avail- able. The enriched preschool education emphasized language and conversation, favorable teacher–child ratios that permitted individualized attention, and positive, responsive interac- tions with teachers.
Supplemental education support from kindergarten through second grade occurred year- round. The support included assisting parents in addition to their children. Intervention ser- vices helped parents learn to provide a stimulating educational environment at home. Efforts also helped parents learn how to work with the school system and ensure their children were receiving necessary services to support learning.
For the children who participated in the intervention program from preschool through the 2nd grade, positive gains in IQ and academic achievement persisted during childhood. In par- ticular, children’s IQ scores were positively impacted, relative to a control group, even at age 15 (Ramey et al., 2000). Children in the Abecedarian program were 2½ times more likely than the control group to enter college (Duncan & Magnuson, 2013). This finding illustrates that IQ is modifiable, and it offers promise that continued research and educational efforts will enable children to overcome obstacles and reach their fullest potential.
Questions to Consider
1. Why do you think IQ scores tend to stabilize around ages 6 or 7?
2. What does your answer imply about the timing and focus of interventions intended to boost children’s IQ scores?
Summary and Resources
Summary and Resources
Chapter Summary
• Intelligence tests are primarily designed to measure cognitive abilities associated with school performance.
• An intelligence test is reliable if it yields consistent performance and valid when it measures the construct it purports to measure. Individual differences on intelligence tests are normally distributed, meaning the average score is most frequent and scores occur less frequently the farther removed they are from the average.
• IQ tests do a reasonably good job of predicting academic achievement, but they are far from perfect indicators of success in school.
• Potential nonintellective influences on intelligence test scores include motivation and assessor bias.
• Psychologists who adopt the psychometric approach theorize the existence of a gen- eral intellectual ability that impacts a broad range of cognitive abilities.
• The three-stratum model of intelligence views intelligence as hierarchically orga- nized, with the most general intellectual ability at the top and more specific intel- lectual abilities, such as crystallized and fluid intelligence, on lower tiers of the hierarchy.
• Some theories of intelligence focus on abilities that are applied to problems and situations occurring in contexts beyond conventional school material. In Gardner’s theory of multiple intelligences, some of the eight intelligences he proposes—includ- ing musical and bodily kinesthetic intelligences—are often manifested in contexts different from schooling.
• Intelligence scores of identical twins are more closely related than intelligence test scores of fraternal twins. Intelligence is heritable, with some, but not all, differences in intelligence within a population related to genetic differences.
• Intellectual differences occur between the sexes and between races. In both cases evidence indicates that issues related to the environment, such as stereotype threat, explain such differences.
• Transactional development occurs when the influence of the environment is multi- plied by characteristics related to one’s genes. Thus, environment and genes jointly influence the child’s intelligence.
• Overall, IQ scores begin to stabilize in individuals around age 6 or 7. However, in a subset of cases, children’s IQ scores substantially change over time.
• IQ scores are potentially modifiable through factors like schooling and nutrition. • Long-lasting gains in IQ scores can occur. This is evident from evaluations of a sub-
stantial intervention that targeted the child’s home life and provided early education from preschool through second grade.
Posttest Questions
1. Intelligence tests were originally designed to .
a. measure social skills b. identify academically at-risk children c. identify group differences in intelligence d. pinpoint the stages of cognitive development
Summary and Resources
2. All of the following are nonintellective influences on IQ scores EXCEPT .
a. processing speed b. low motivation c. assessor bias d. administrating the IQ test in a language unfamiliar to the child
3. Test reliability refers to .
a. how well a test predicts a child’s performance in school b. whether a test measures what it is supposed to measure c. whether repeated presentations of the test yield consistent results d. whether a test is free from bias
4. refer(s) to accumulated knowledge and information.
a. Crystallized intelligence b. Spatial intelligence c. Fluid intelligence d. Multiple intelligences
5. The consistent finding of correlations among different measures of different aspects of intelligence is known as .
a. successful intelligence b. localization of function c. the positive manifold d. the Flynn effect
6. Which of the following statements about the psychometric approach to intelligence is NOT true?
a. It emphasizes cognitive skills closely linked to school performance. b. It emphasizes individual differences within specific age groups. c. It emphasizes stages of cognitive development. d. It is associated with a general factor of intelligence (g).
7. The heritability of intelligence during childhood.
a. increases b. is not possible to measure c. decreases d. stays the same
8. The statement “nature impacts nurture and nurture impacts nature” is MOST descriptive of which of the following concepts?
a. nonintellective influences b. transactional development c. positive manifold d. localization of function
Summary and Resources
9. According to the IP view of intelligence, an infant’s intelligence can be predicted by measuring .
a. how long an infant takes to habituate to a stimulus b. an infant’s recall memory of words c. the sensitivity of an infant’s sensory register to bright and colorful objects d. an infant’s motor skills
10. IQ scores have been rising worldwide. This phenomenon is known as the .
a. positive manifold b. genetic effect c. Flynn effect d. unreliability effect
Critical-Thinking Questions
1. At various points in the chapter, we discussed not only how IQ predicts school performance but also how schooling itself enhances IQ. Explain how these two considerations argue for or against tracking (that is, placing a child in an educational setting modified to match his or her intellectual potential).
2. Why is it misleading to say “intelligence is inherited” even though genetic differ- ences are related to intellectual differences? How might the belief that intelligence is inherited impact children’s motivation and academic achievement for better and/or for worse?
3. After reading this chapter, how would you respond to a senator or congressional representative who wants to reduce support for early educational interventions because they do not apparently produce long-lasting gains in IQ?
Key Terms
androgen A sex hormone related to the development of masculine characteristics.
assessor bias A situation in which a test administrator does not follow the prescribed procedures and instructions.
crystallized intelligence Skills and knowl- edge acquired from learning and experience.
dynamic testing An assessment designed to measure a child’s intellectual potential by examining how the child responds to feed- back and prompts during testing.
fluid intelligence The intellectual ability to reason and solve novel problems.
Flynn effect The discovery that intelligence test performance rose in the 20th century over successive generations in many nations.
g A global factor of intelligence influencing performance across a very broad range of cognitive measurements.
gifted A designation of children’s intel- lectual potential that is typically associated with extraordinarily high performance on a measure of intelligence.
heritability A statistical estimate of the extent that individual differences on a char- acteristic are related to genetic differences within a population.
Summary and Resources
intellectual disability A diagnosis of an impairment in mental ability and in every- day living skills generally originating before age 18.
intelligence The ability to reason, solve problems, and analyze, along with the ability to profit from experience.
intelligence quotient (IQ) An index of intelligence derived by calculating the ratio of a person’s test performance, referred to as “mental age,” to the person’s chronological age.
localization of function The association of a cognitive function with a particular area or areas of the brain.
math anxiety An unpleasant emotional experience instigated by math-related thoughts and assessment.
multiple intelligences The theory that separate core abilities and processes oper- ate in specific contexts to produce outcomes and solutions in culturally valued areas like music, interpersonal relationships, and number.
nonintellective influences Factors other than the child’s intellectual competence that occur during testing and impact an IQ score.
normal distribution A distribution in which outcomes on a measurement occur most frequently around the overall average score and less frequently for scores that are farther from the average.
normative sample A group of individu- als whose scores on a measurement are intended to be representative of the larger population.
positive manifold The occurrence of posi- tive interrelations among measurements of a wide variety of cognitive abilities.
prodigy An individual with at least average levels of general intelligence who also dem- onstrates unusually high ability in one area.
psychometric approach The idea that intelligence is characterized through the analysis of patterns of outcomes of intelli- gence tests.
standard deviation A measurement of the extent to which scores vary from the average.
stereotype threat A situation in which awareness of a negative stereotype relevant to oneself impairs performance on a mea- surement related to the stereotype.
successful intelligence The ability to progress toward attaining goals within one’s cultural context.
test reliability A situation that occurs when a measurement produces consistent results.
test validity A situation that occurs when a test is actually measuring what it purports to assess.
three-stratum model The idea that intel- ligence is organized in a hierarchy from general to increasingly specific intellectual abilities.
transactional The idea that abilities and behaviors related to one’s genes both influence and are influenced by one’s environment.
twin studies Studies in which the simi- larities between identical twins on a mea- surement are compared to the similari- ties between fraternal twins on the same measurement.
Summary and Resources
Additional Resources Web Resources
Intelligence Overview http://www.apa.org/topics/intelligence/index.aspx The American Psychological Association offers a definition of intelligence and links to articles covering a wide range of current issues about intelligence, including factors that may increase it and the Flynn Effect (that is, “Smarter than Ever?”).
Motivation and IQ http://www.nih.gov/researchmatters/may2011/05022011motivation.htm An interview with a researcher and brief story about the influence of motivation on IQ test performance.
Multiple Intelligences http://multipleintelligencesoasis.org/about An overview of the theory of multiple intelligences and implications for education.
Further Reading
Deary, I. J. (2001). Intelligence: A very short introduction. Oxford, UK: Oxford University Press. An accessible overview of the field of intelligence.
Deary, I. J. (2012). Intelligence. Annual Review of Psychology, 63, 453–482. A summation of contemporary research on intelligence.
Flynn, J. R. (2012). Are we getting smarter?: Rising IQ in the twenty-first century. New York: Cambridge University Press. The Flynn effect discussed by the researcher who identified the rise in IQ scores that occurred over successive generations in the 20th century.
Gardner, H. (2006). Multiple intelligences: New horizons (Rev. ed.). New York: Basic Books. An overview of Gardner’s theory of multiple intelligences.
Kranzler, J. H., & Floyd, R. G. (2013). Assessing intelligence in children and adolescents: A prac- tical guide. New York: Guilford. A practical reference describing the concepts, theories, and findings underlying the contemporary assessment of intelligence.
Nisbett, R. E. (2009). Intelligence and how to get it: Why schools and cultures count. New York: Norton. A researcher discusses the role of the environment in the development of intelligence.
Answers and Rejoinders to Chapter Pretest
1. False. The intelligence test was developed in the early 20th century to identify chil- dren who were at risk academically and needed special attention.
2. True. Scores on measurements of a wide variety of cognitive abilities tend to be interrelated. This finding is often interpreted as reflecting a general intellectual ability.
3. False. Behaviors and abilities influenced by genes impact, and are impacted by, the environment. Genes and environment work together to produce individual differ- ences in intelligence.
Summary and Resources
4. True. Intelligence scores begin to stabilize around age 6 or 7. Prior to that age, the relationships between earlier and later IQ scores tend to be smaller, though still detectable.
5. False. Studies find that initial gains in IQ due to interventions often fade over time. One exception is an intervention that focused on parents as well as children and extended children’s educational support from preschool into the grade school years.
Answers and Rejoinders to Chapter Posttest
1. b. identify academically at-risk children Alfred Binet created the intelligence test to identify children who needed extra attention to reach their academic potential.
2. a. processing speed Nonintellective influences are factors other than the child’s true intellectual compe- tence that impact test performance.
3. c. whether repeated presentations of the test yield consistent results Test reliability is assessed by the degree to which performance on a test is consistent.
4. a. Crystallized intelligence Crystallized intelligence refers to knowledge and abilities such as vocabulary that are acquired from experience.
5. c. the positive manifold The positive manifold refers to the interrelation among a wide variety of cognitive tasks.
6. c. It emphasizes stages of cognitive development. The psychometric approach to intelligence is concerned with measuring and explaining individual differences among children rather than general discontinuities in development.
7. a. increases The heritability of intelligence is not fixed but changes during development as a result of the interaction between genes and the environment.
8. b. transactional development Transactional development describes the process by which features related to genes impact the environment and the environment, in turn, impacts those same features. Nature and nurture jointly influence development.
9. a. how long an infant takes to habituate to a stimulus Processing speed and efficiency are measured by the rate of habituation in infancy.
10. c. Flynn effect The Flynn effect is the discovery that successive generations of normative samples outperformed earlier generations.