You probably have a good sense of your level of intelligence. Most people have taken standardized intelligence tests and received scores ranking them in comparison to others. However, you may have less understanding of what, exactly, intelligence is. Is it something in your head—maybe a part of the brain? Or it is just a label that summarizes people’s performance when answering the questions on intelligence tests?

If you’ve ever asked yourself what intelligence is, you’re not alone. Psychologists puzzle over the question. Although researchers have developed measures of intelligence and identified factors that influence people’s intelligence test scores, basic questions about the nature of intelligence are debated to the present day. Let’s begin our study of intelligence, then, by examining definitions of the term.

342

We call people “intelligent” when we see them do something difficult that requires thinking. Whether it’s learning a foreign language, solving a math problem, inventing the electric light bulb, developing a novel form of poetry, or just correctly answering lots of questions on an intelligence test, the accomplishment—the learning, the solving of problems, or the creation of something new—indicates the person’s intelligence (Gardner, 1993, Sternberg, 1985). Intelligence, therefore, is the ability to acquire knowledge, to solve problems, and to use acquired knowledge to create new, valued products. To many psychologists, intelligence refers, more specifically, to a general ability to do these things—that is, an ability to comprehend information and respond effectively to problems that arise on different types of activities (Deary, Penke, & Johnson, 2010).

This definition indicates not only what intelligence is, but also what it is not. “Intelligence” does not refer to the following:

Our definition of intelligence leaves some important questions unanswered. The most significant is whether there is one type of intelligence or multiple types, that is, distinct mental abilities, each of which is a type of intelligence. Let’s look at the origins of intelligence testing, which focused on measuring one type of intelligence, known as general intelligence.

343

ORIGINS OF INTELLIGENCE TESTING AND IQ. More than a century ago, psychologists began to measure differences in people’s mental abilities (Brody, 1992; Fancher, 1979). A pioneer in the field was the French psychologist Alfred Binet. Binet reasoned that some children might have mental disabilities and thus would need extra help from teachers. To identify these children, Binet devised an intelligence test.

Before Binet, psychologists interested in intelligence had measured people’s perceptual ability and memory capacity. Binet’s test, however, was different. He measured logical reasoning, vocabulary use, and factual knowledge; children taking his test had to identify objects, explain the logical relation between ideas, and construct sentences using challenging vocabulary words. These test items tapped abilities that are important to school achievement. Binet’s test thus was a success; test scores predicted classroom achievement. Today’s intelligence tests, which employ items similar to those Binet developed, strongly predict school success (Greven et al., 2009).

In 1912 a German psychologist, William Stern, provided a valuable scoring system for intelligence tests (Wasserman & Tulsky, 2005). Stern capitalized on an insight from Binet, who recognized that, when assessing a child’s intelligence, one should consider the child’s age. Different children might develop the same ability at different ages. The child who develops the ability earlier is said to be more intelligent.

Stern included age in a formula for computing intelligence quotient, or IQ, which is a measure of overall intelligence that takes into account a person’s age. It does so by relating any given child’s test performance to the average age of children who perform similarly on that test. Here’s an example.

344

The mental age of the 7-year-old who answered 43 questions correctly, then, is 8. A child’s mental age is the average age of children who perform as well as he or she did on an intelligence test.

Stern included mental age in his formula for IQ:

IQ = (mental age/chronological age) × 100

An average child—one whose mental age equals his or her chronological age—would have an IQ of 100. The 7-year-old with a mental age of 8 would have an IQ of 114 (8/7 × 100). When intelligence tests are given to adults, with age no longer being a relevant factor, the tests are scored so that the mean score in a population is 100 and the standard deviation (an index of the degree to which numbers spread, or vary, from the mean; see Chapter 2) is 15; about two-thirds of people, then, receive IQ scores between 85 and 115.

THE NATURE OF GENERAL INTELLIGENCE. Computing a person’s IQ score is easy. But it leaves a difficult question unanswered: “What, exactly, does the IQ score measure?”

The first step in answering that question involves the following fact: Whenever people take tests measuring different mental abilities, those who get high scores on any one test (e.g., verbal skills) tend also to get high scores on others (e.g., math skills). Individual differences on the separate tests correlate positively (Deary, 2001). What explains the positive correlations? More than a century ago, the British psychologist Charles Spearman (1904) explained them by proposing the variable general intelligence, or g. People with high (low) g tend to perform well (poorly) on tests of mental ability.

IQ, then, measures g. IQ scores measure individual differences in general intelligence.

Let’s see exactly how this works by examining one popular test of intelligence, the Wechsler Adult Intelligence Scale, or the WAIS (Wechsler, 2008). The WAIS includes measures of four aspects of mental ability:

345

Although these four abilities are distinct—comprehending words differs conceptually from solving problems involving shapes—the four WAIS abilities are correlated positively (Deary, 2001). People with relatively high Verbal Comprehension scores tend also to get relatively higher Perceptual Reasoning, Working Memory, and Processing Speed scores. The positive correlations among test scores can be summarized by g. In a hierarchical model, the four WAIS factors are linked to the high-level factor of general intelligence (Figure 8.12). To assess g, the tester combines a person’s scores on the four measures.

Is g a single thing—for example, a single structure in your brain? No. If you look inside your brain, you won’t find “a g.” General intelligence is merely is a summary of some numbers: the positive correlations among test scores when a group of people take a number of different tests (Borsboom & Dolan, 2006; Deary, 2001). A wide variety of mechanisms in the brain could contribute to the correlations among test scores. But no single brain mechanism would correspond directly to g (van der Maas et al., 2006).

It might at first sound odd that the single variable g does not correspond to a single thing in the brain. An analogy makes it clear. Suppose you’re running a theatre group, putting on a musical, and need people for the show. You develop tests for the needed talents: a singing test, a dancing test, a test for stage design ability, and a test for technical skills such as lighting the stage. People’s scores on these tests might correlate positively. Even if they do, that doesn’t mean people’s brains contain a “General Theatre Ability.” There are other explanations for the positive correlations. People who sing well in childhood might be more likely to appear in stage shows. Once in the shows, they receive dance lessons. Eventually, they hang out with people who do stage design and lighting. As a result of these experiences, they pick up diverse theatre-related knowledge and skills, and the four tests correlate positively. Analogously, people who have diverse classroom educational experiences may pick up the diverse knowledge and skills that go into g.

These skills are significant. People with higher scores on tests of general intelligence often achieve higher levels of academic and professional success. A study of tens of thousands of British schoolchildren showed that general intelligence scores at age 11 substantially predicted school achievement in math and English at age 16 (Deary et al., 2007). Research relating g to professional success indicates that higher levels of general intelligence predict higher levels of success in job-training programs and higher levels of on-the-job work performance (Ree & Earles, 1992).

GENERAL INTELLIGENCE AND WORKING MEMORY. The psychologist Raymond Cattell (1987) explained that, to identify cognitive processes that contribute to intelligence, you first have to distinguish between two forms of general intelligence: fluid and crystallized. Fluid intelligence refers to mental abilities that can benefit performance on almost any challenging task; these abilities have “the ‘fluid’ quality of being directable to almost any problem” (Cattell, 1987, p. 97). Thanks to fluid abilities, people can perform intelligently even on tasks they have never seen before (e.g., puzzles, logic problems, or detecting patterns in a series of geometric shapes). Crystallized intelligence refers to socially acquired knowledge. Someone who knows a lot of facts, or an unusually large number of vocabulary words, would be said to have high crystallized intelligence. In contrast to fluid intelligence, the knowledge that comprises crystallized intelligence is generally more helpful on some tasks than others; knowing words, for example, helps in solving verbal problems but not arithmetic problems.

346

It is difficult to identify specific cognitive processes that contribute to crystallized intelligence. This aspect of knowledge is diverse—people acquire a wide range of knowledge through experiences—so the cognitive processes involved may be equally diverse. Fluid intelligence, however, is different. Many researchers suggest that a specific component of the mind is key to fluid intelligence: working memory.

Working memory (see Chapter 7) holds information for brief periods and functions as an “executive” that focuses a person’s attention on tasks and helps people to avoid distraction. Individual differences in working memory capacity predict individual difference in fluid intelligence (Engle, 2002). For instance, when researchers measure people’s ability to avoid distraction, those who are less distractible are found to have higher fluid intelligence (Engle et al., 1999) and higher scores on intelligence tests that combine fluid and crystallized intelligence abilities (Colom et al., 2004).

Individual differences in working memory capacity thus contribute to individual differences in fluid intelligence. This fact, however, leaves unanswered the following question: Why do people differ in working memory capacity and fluid intelligence? In other words, why are some people smarter than others?

One possible cause of individual differences in intelligence is biological inheritance. Just as people inherit genes that determine their eye color or height (see Chapter 4), they may inherit genes that determine their level of intelligence. Intelligence, then, may be a fixed, genetically determined quality that you couldn’t alter even if you wanted to.

347

Over the years, some psychologists have made this claim. If intelligence is biologically determined, then society perhaps should not spend much money educating people with low IQ scores, because education won’t boost their intellectual ability (Herrnstein & Murray, 1994; Jensen, 1969).

Do the facts support the controversial claim that intelligence is determined genetically? Let’s look closely at this question by examining a biologically determined trait and then comparing it to intelligence.

Here are four facts about height, a biologically determined trait:

These four facts are true, then, for biologically determined traits, such as height. Are they also true for intelligence? Let’s look at the evidence.

CHANGES ACROSS GENERATIONS. Intelligence tests have been around for a long time. Test publishers use many of the same test items from one year to the next. Even when they add new items, they do so carefully, conducting statistical analyses to ensure that the tests don’t vary in difficulty from one year to the next. This makes it possible to compare IQ scores across historical periods; researchers can thus see whether IQ changes from one generation to another.

Would you expect substantial changes? If biology completely determines intelligence, you shouldn’t expect changes; the human genome doesn’t vary much from one generation to the next. But research by the New Zealand psychologist James Flynn (1987, 1999) shows that IQ changes are substantial. Flynn analyzed IQ scores across decades of the twentieth century for different countries, one at a time. His results were startling. In each country, people became more intelligent across the decades of the second half of the twentieth century. IQ scores went up a lot (Figure 8.13). In England, for example, scores shifted by nearly 30 points (statistically, nearly 2 standard deviations) in only a half century, 1940 to 1990.

This increase in intelligence across generations has become known as the Flynn effect. The Flynn effect is the rise in a population’s average IQ over time.

348

The Flynn effect is so large that it couldn’t possibly result from changes in human biology. By comparison, there are no proportional changes in human height. The standard deviation of height is about 3 inches; a comparable change, then, would be 5–6 inches in average height in a half-century. No such changes occurred during the twentieth century or any other comparable period in human history (Steckel, 1995). The Flynn effect therefore shows that factors other than biology must be influencing IQ scores.

It is difficult to know precisely what those factors are (Neisser, 1997). Rather than truly being more intelligent, people today might simply be better at taking tests; people might acquire test-taking skills by working on puzzles or video games, or by taking classes in which test-taking tips are taught. Alternatively, improvements in education might provide people today with more knowledge and skills, that is, more crystallized intelligence. So let’s look at the effect of education on intelligence test scores.

THE EFFECTS OF A SOCIAL EXPERIENCE: EDUCATION. Does attending school increase your intelligence? Research evidence says yes. People who spend more time in school become more intelligent; they achieve higher scores on intelligence tests (Ceci, 1991).

Two sources of evidence support this conclusion. One is the correlation between years spent in school and IQ scores. Rather than being .00 (like the correlation between number of days in basketball camp and height), the correlation is large, about .80. The positive correlation is found even after accounting for other factors, such as the possibility that more intelligent kids start school early (Ceci, 1991). Education, then, is a social experience that increases intelligence.

The second source of evidence involves summer vacation. Summer vacation is great—the best part of the school year! But when it comes to intelligence, it’s costly. IQ declines during the summer break; students’ IQ scores are higher at the end of a school year than at the start of the next school year. This is particularly true among low-income children, who are less likely to engage in educationally enriching activities during the summer (Ceci, 1991; also see This Just In).

THIS JUST IN

STABILITY OF INDIVIDUAL DIFFERENCES. Are individual differences in intelligence as stable as individual differences in height, with some people always being taller than others? To find out, researchers studied adolescents at two time periods: age 14 (on average) and three years later (Ramsden et al., 2011). They expected intelligence to be stable, like height. But “we were very surprised,” the project leader reported. Many adolescents changed a lot. “We had individuals that changed from being in the 50th percentile, with an IQ of 100, [all] the way up to being in the (top) 3rd percentile, with an IQ of 127” (Aubrey, 2011). Scores also went down. One individual’s verbal IQ score dropped by 20 points—a change that, by comparison with height, would be like becoming more than 3 inches shorter!

Changes in IQ scores turned out to be correlated with changes in the brain (Ramsden et al., 2011). Through brain-imaging evidence, the researchers found that changes in the verbal component of IQ scores correlated with changes in the density of neurons in a part of the brain needed for producing speech (Figure 8.15). This result means that the variations in IQ weren’t due to “random” factors, such as test takers not trying as hard one of the two times they took the test. The changes were meaningful. People varied, across time, at the level of the mind (the ability to answer the test items) and the brain (neural density).

SOCIOECONOMIC STATUS. If intelligence were like height, then, within any of a variety of socioeconomic groups, rich or poor, genetic factors would be the major determinant of individual differences.

But it turns out that the impact of genes on intelligence is not like this. When it comes to intelligence, sometimes genetic effects are big and sometimes they’re small. Critical evidence comes from a study of hundreds of pairs of identical and fraternal twins from families of diverse economic backgrounds (Turkeimer et al., 2003; reviewed in detail in Chapter 4). In wealthy populations, genetics explained the vast majority of individual differences in intelligence. But in poor populations, genetics explained only a very small percentage of the individual differences.

Why would genetic factors have only a small effect within populations of people who are poor? Children living in poverty may experience poor nutrition, inconsistent parental care, and substandard school facilities. These environmental factors, not found in wealthy populations, can strongly affect children’s intellectual development (Nisbett, 2009).

350

We began this section of the chapter by asking whether intelligence is like height: determined almost entirely by inherited biology, and little affected by the environment. You’ve now seen that it is not. Thanks to today’s research evidence, we can, as one scholar put it, “shake off the yoke of hereditarianism in … our thinking about intelligence” (Nisbett, 2009, p. 199) and recognize that both genetic and environmental factors shape a person’s mental ability.

351

Consider the following three people:

These students don’t sound like a very promising lot. But the three boys are (1) Igor Stravinsky, commonly considered the twentieth century’s greatest composer of symphonic music; (2) Mohandas Gandhi, the revered Indian spiritual and political leader who led his nation to independence; and (3) Pablo Picasso, universally acknowledged as one of the greatest geniuses in the history of art (Gardner, 1993). They each displayed the highest possible level of brilliance in their fields.

Note two facts about Stravinsky, Gandhi, and Picasso:

The fact that Stravinsky, Gandhi, and Picasso each displayed exceptional intellectual ability in one domain of life but not others raises an important question. Is the concept of general intelligence the right way to think about human mental abilities?

Howard Gardner (1993, 1999) thinks not. According to his multiple intelligences theory, people possess a number of different mental abilities. Each is a distinct form of intelligence; that is, they are not merely aspects of general intelligence. Gardner emphasizes that one can have a high degree of one intelligence and a low degree of another.

352

Gardner has proposed eight different types of mental ability—that is, eight different intelligences. Table 8.2 lists them, along with professions in which you might expect to find people with particularly high levels of the given intelligence.

Howard Gardner’s multiple intelligences theory proposes that human intellect contains eight different forms of intelligence.

How can we know these are distinct intelligences, rather than different aspects of one overall, general intelligence? Gardner (1993) cites a variety of evidence. One is child prodigies, children who exhibit exceptional mastery of a skill at a very early age. Prodigies usually are not “generalized geniuses” who excel at everything. Rather, they often excel in one form of intelligence. For example, a twentieth-century musical prodigy, the jazz pianist Art Tatum, taught himself piano at age 3. He soon could play complete songs after hearing them only once. By age 6, he was playing, by himself, pieces he heard that were written as duets, apparently “unaware that there were supposed to be two players” (Art Tatum, n.d.). There is no evidence that Tatum was exceptionally skilled at the verbal, logical, or mathematical tasks included on the WAIS. But in musical intelligence, he was an off-the-charts genius.

353

Another form of evidence is savant syndrome, in which people who are mentally impaired in most areas of life display exceptional performance—an “island of genius” (Treffert, 2009, p. 1)—in one domain. Savant syndrome is sometimes seen among people with autism, a brain disorder whose symptoms include an impaired ability to interact socially with others. People with savant syndrome display exceptionally high levels of skill early in life in specific tasks, such as music, math, or art. The differences in their performance in different domains—impaired in some, exceptionally skilled in others—are consistent with Gardner’s claim that the mind contains multiple intelligences that are distinct from one another.

In savant syndrome, then, exceptional skill accompanies mental impairment early in life. There are also cases where people develop exceptional intellectual abilities late in life, in the midst of declining overall mental capacity. Researchers (Miller et al., 1998) report five cases in which patients in their 50s and 60s developed artistic abilities while suffering from dementia, a progressive decline in mental functioning, due to an abnormal biological deterioration of the front of the brain. These patients’ overall ability to think and act effectively grew worse and worse. But their artistic abilities improved!

The image shown in Figure 8.16 was painted by one of these patients, a woman in her mid-60s whose “speech became repetitive and rambling” (Miller et al., 1998, p. 979) due to the dementia, but whose artwork became remarkably skilled. The researchers suggest that the patients’ dementia made them less emotionally inhibited than they previously had been, which in turn allowed them to express themselves more freely in their art.

This evidence, too, supports Gardner’s multiple intelligences approach. If some mental abilities improve while others deteriorate, the pattern of changes cannot be explained in terms of general intelligence.

Gardner’s theory has been criticized, however. Some argue that the evidence in support of it is still sparse (Waterhouse, 2006). Yet others see great virtue in Gardner’s work. The people around you—even if they aren’t Stravinskys, Gandhis, and Picassos—often possess distinctive skills and abilities that are not captured by traditional intelligence tests. Multiple intelligences theory helps us appreciate the rich range of mental abilities possessed by the world’s diverse individuals.

354

Gardner’s approach is not the only theory that moves beyond the traditional “general intelligence” approach. For example, Robert Sternberg’s triarchic theory of intelligence proposes that intelligent behavior requires three distinct mental components:

In Sternberg’s model, the components of a person’s intelligence might be more highly developed in one aspect of life than another; for instance, you might have a lot of intelligence when it comes to fixing a car (i.e., a lot of knowledge, executive-planning skills, and ability to carry out the plans), but less intelligence when it comes to planning a meal (or vice versa). As in Gardner’s approach, then, intelligence is not a “general” ability. It is a collection of mental skills that enable you to solve problems in specific areas of life.

Let’s conclude by focusing on a biological level of analysis and asking whether observed individual differences in overall intelligence, or g, can be explained in terms of underlying individual differences in nervous system structure or functioning (Deary et al., 2010).

BRAIN SIZE. If you see one cartoon character with a small head, and one with a large head, it’s not hard to guess which is the smart one. But is there any truth to the stereotype? Are people with larger brains more intelligent?

It turns out that the answer is yes (Deary et al., 2010). The most convincing evidence comes from a meta-analysis, that is, a statistical summary of results of large numbers of studies. One meta-analysis summarized more than three dozen prior studies, with more than 1500 participants (McDaniel, 2005). Results revealed that IQ correlated positively with brain volume (r = +.33). Bigger-brained individuals got higher IQ scores.

The correlation, .33, is significantly greater than zero. Yet it is nowhere near a perfect correlation of 1.0. Many people with small brain volume have high IQ scores (and vice versa). For example, the brain of Walt Whitman, a genius of American poetry, was below average in size (Gould, 1981). Brain size thus does not determine your level of intelligence; it merely predicts individual differences in intelligence to a moderate degree.

355

BRAIN CONNECTIONS. Intelligence does not reside in any one part of the brain. Intelligent behavior requires coordination among large numbers of brain regions (Bullmore & Sporns, 2009). Individual differences in intelligence thus might be grounded in the way these regions are interconnected.

People differ in the strength of interconnections in the brain. One person might have well-developed connections between any two brain regions, whereas, in another person’s brain, these regions might be connected weakly. Evidence suggests that people with highly developed interconnections process information more efficiently and, as a result, get higher IQ scores (Deary et al., 2010). Just as a highway system is efficient if there are numerous roads connecting different places, a brain is efficient if it contains numerous connections among its different places.

In one study (Li et al., 2009), researchers measured brain interconnections among two groups of participants: people with (1) average to high IQ scores and (2) very high IQ scores (> 120). Using images of the strength of these interconnections (Figure 8.17), they computed a measure of brain efficiency, the overall strength of interconnections in the person’s brain. The average-to-high and very high IQ participants differed significantly in brain efficiency (Li et al., 2009). People who had stronger axon interconnections among the regions of their brains, and who thus were able to process information more efficiently, were more intelligent (Figure 8.18).

This research, conducted at a brain level of analysis, directly connects to research you learned about earlier that was conducted at a mind level of analysis. You saw that people’s mental powers are significantly shaped by their environmental experiences. Schooling, for example, boosts IQ. Environmental experiences affect brain connections, too; different experiences create different interconnections in the brain (Edelman & Tononi, 2000; see Chapter 3). Research on neural interconnections and intellectual ability thus can give us a brain-level understanding of how people’s intelligence is shaped by their experiences in the world.

356