Intelligence exerts a major influence on anyone’s thinking ability. It is easy to identify in people and even easy to observe in other animals. Yet intelligence is not at all easy to define. Despite a century of study, researchers have not yet reached agreement on what intelligence entails. We therefore begin this section by reviewing some hypotheses of intelligence.

553

In the 1920s, Charles Spearman proposed that although different kinds of intelligence may exist, there is also an underlying general intelligence, which he called the g factor. Consider for a moment what a general intelligence factor might mean for the brain. Presumably, brains with high or low g would have some general difference in brain architecture—perhaps in gyral patterns, cytoarchitectonics, vascular patterns, or neurochemistry, for example.

This difference could not be something as simple as size, because human brain size (which varies from about 1000 to 2000 grams) correlates poorly with intelligence. Another possibility is that g is related to special cerebral connectivity or even to the ratio of neurons to glia. Still another possibility is that g is related to the activation of specific brain regions, possibly in the frontal lobe (Duncan et al., 2000; Gray & Thompson, 2004).

The results of preliminary studies of Albert Einstein’s brain implied that cerebral connectivity and ratio of glia to neuron may be important. Sandra Witelson and her colleagues (Witelson et al., 1999) found that although Einstein’s brain is the same size and weight as the average male brain, its lateral fissure is short, and both the left and the right lateral fissures take a particularly striking upward deflection (Figure 15-21; compare Figure 15-11). This arrangement essentially fuses the inferior parietal area with the posterior temporal area.

The inferior parietal cortex has a role in mathematical reasoning, so it is tempting to speculate that Einstein’s mathematical abilities were related to neural rearrangements in this area. But another important difference may distinguish Einstein’s brain. Marion Diamond and her colleagues (1985) looked at its glia-to-neuron ratio versus the mean for a control population. They found that Einstein’s ratio in the inferior parietal cortex was higher than average: each neuron in this region had an unusually high number of glial cells supporting it.

The glia-to-neuron ratio was not unusually high in any other cortical areas of Einstein’s brain measured by these researchers. Possibly, then, certain types of intelligence could be related to differences in cell structure in localized brain regions. But even if this hypothesis proves correct, it offers little neural evidence in favor of a general intelligence factor.

One possibility is that g is related to the brain’s language processes, because language ability qualitatively changes the nature of cognitive processing in humans. So perhaps people with very good language skills also have an advantage in general thinking ability.

Many other hypotheses on intelligence have been set forth since Spearman’s, but few have considered the brain directly. One exception, proposed by Howard Gardner (1983), a neuropsychologist at Harvard, considers the effects of neurological injury on people’s behavior. Gardner concludes that seven distinct forms of intelligence exist and that brain injury can selectively damage any form. The idea of multiple human intelligences should not be surprising given the varied cognitive operations the human brain can perform.

Gardner’s seven categories of intelligence are linguistic, musical, logical-mathematical, spatial, bodily-kinesthetic, intrapersonal, and interpersonal. Linguistic and musical intelligence are straightforward concepts, as is logical-mathematical intelligence. Spatial intelligence refers to abilities discussed in this chapter, especially navigating in space, and to the ability to draw and paint. Bodily-kinesthetic intelligence refers to superior motor abilities, such as those exemplified by skilled athletes and dancers.

The two types of “personal” intelligence are less obvious. They refer to the frontal and temporal lobe operations required for success in a highly social environment. The intrapersonal aspect is awareness of one’s own feelings, whereas the interpersonal aspect entails recognizing others’ feelings and responding appropriately. Gardner’s definition of intelligence has the advantage not only of being inclusive but also of acknowledging forms of intelligence not typically recognized by standard intelligence tests, abilities such as theory of mind, described in Section 15-3.

554

One prediction stemming from Gardner’s analysis of intelligence is that brains ought to differ in some way when people have more of one form of intelligence and less of another. Logically, we could imagine that if a person were higher in musical intelligence and lower in interpersonal intelligence, then the brain regions for music (especially the temporal lobe) would differ in some fundamental way from the “less efficient” regions for interpersonal intelligence. One way to examine such differences is to use fcMRI or DTI to identify differences in pathways, as in the example of absolute pitch (see Figure 15-10).

One clear difference between lesions in the parietal and temporal lobes and lesions in the frontal lobes is in the way they affect performance on standardized intelligence tests. Posterior lesions produce reliable and often large decreases in intelligence test scores, whereas frontal lesions do not. This is puzzling. If frontal lobe damage does not diminish a person’s intelligence test score, why do people with this kind of damage often do stupid things? The answer lies in the difference between two kinds of intelligence.

According to J. P. Guilford (1967), traditional intelligence tests measure convergent thinking—applying knowledge and reasoning skills to narrow the range of possible solutions to a problem, then zeroing in on one correct answer. Typical intelligence test items using vocabulary words, arithmetic problems, puzzles, block designs, and so forth all require convergent thinking. They demand a single correct answer that can be easily scored.

In contrast, divergent thinking reaches outward from conventional knowledge and reasoning skills to explore new, more unconventional solutions to problems. Divergent thinking assumes a variety of possible approaches and answers to a question rather than a single “correct” solution. A task that requires divergent thinking is to list all the possible uses you can imagine for a coat hanger.

Clearly, a person who is very good at divergent thinking might not necessarily be good at convergent thinking and vice versa. The distinction is useful because it helps us to understand the effects of brain injury on thought. Frontal lobe injury is believed to interfere with divergent thinking. The convergent thinking measured by standardized IQ tests is often impaired in people with damage to the temporal and parietal lobes.

Injury to the left parietal lobe in particular causes devastating impairment in the ability to perform cognitive processes related to academic work. People with this kind of injury may be aphasic, alexic, and apraxic. Many have severe deficits in arithmetic ability. All such impairments would interfere with school performance and performance at most jobs.

Patient M. M., discussed in Section 15-4, had left parietal lobe injury and was unable to return to school. In contrast with people like M. M., those with frontal lobe injuries seldom have deficits in reading, writing, or arithmetic. And they show no decrement in standardized IQ tests. C. C.’s case provides a good example.

C. C. had a meningioma along the midline between the frontal lobes. Extracting it required removing brain tissue from both hemispheres. Before his surgery C. C. was a prominent lawyer. Afterward, although he still had a superior IQ and superior memory, he was unable to work, in part because he no longer had any imagination. He could not generate the novel solutions to legal problems that had characterized his career before the surgery. Thus, both M. M. and C. C. had problems that prevented them from working, but their problems differed because their injuries affected different kinds of thinking.

555

Donald Hebb proposed another way to categorize human intelligence. Like Guilford, Hebb thought of people as having two forms, which he called intelligence A and intelligence B. Unlike Guilford’s convergent–divergent dichotomy, Hebb’s intelligence A refers to innate intellectual potential, which is highly heritable: it has a strong genetic component. Intelligence B is observed intelligence, which is influenced by experience as well as other factors, such as disease, injury, or exposure to environmental toxins, especially during development.

Hebb (1980) understood that experience can influence brain cell structure significantly. In his view, experiences influence brain development, and thus observed intelligence, because they alter the brain’s synaptic organization. It follows that appropriate postnatal experiences can enhance development of intelligence B in people with lower than average intelligence A, whereas a poor or underresourced environment can hinder the development of intelligence B in people with higher than average intelligence A. The task is to identify a good and a bad environment so as to stimulate people to reach their highest potential intelligence.

One implication of Hebb’s view of intelligence: the brain’s synaptic organization is key. Synaptic organization is partly directed by a person’s genes but is also affected by epigenetic factors. Variations in the experiences to which people are exposed, coupled with variations in genetic patterns, undoubtedly contribute to the individual differences in intelligence that we observe—both quantitative differences (as measured by IQ tests) and qualitative differences (as in Gardner’s view).

Hikaru Takeuchi and colleagues (2012) used fMRI to characterize brain activation while participants performed working memory tasks of varied complexity. Performance on IQ tests and memory are highly correlated, so the investigators reasoned that brain activity during memory tests might reflect brain differences related to IQ score.

Performance speed correlated with increased activation in the right dorsolateral prefrontal cortex as well as an increase in the interaction between the prefrontal cortex and right posterior parietal cortex. Gray matter volume in the right dorsolateral prefrontal region correlated with the participant’s accuracy in working memory tasks, which in turn correlated with psychometric intelligence measures.

Others have obtained parallel results using event-related potentials (e.g., Langer et al., 2009). The general conclusion from ERP studies is that general intelligence is related to the efficiency of cortical networks linking prefrontal and parietal regions. We can speculate that the efficiency of different neural networks, as might be seen in fcMRI studies, will underlie the variation in each of Gardner’s seven forms of intelligence.

Finally, neuropsychological studies using tests of executive (frontal lobe) function show an advantage for bilingual speakers relative to monolinguals. The difference is hypothesized to reflect bilinguals’ consistent need to select language-appropriate words and to inhibit language-inappropriate words. Further, learning two languages early in life appears to confer an even greater advantage than acquiring a second language later in life.

Olulade and colleagues (2015) found that, versus monolinguals, increased gray matter volume in the frontal lobe of adults who learned two languages before age 6 reflects this behavioral advantage. No such difference appeared when the investigators compared individuals who acquired an oral language and American Sign Language. The finding is consistent with the idea that the challenge of selecting language-appropriate words is critical. How the increased gray matter volume correlates with other measures of intelligence is unknown, but the results are consistent with evidence showing long-term cognitive advantages in elderly bilinguals. This suggests a broader advantage for bilinguals than in executive functions alone.

556

Intelligence

Before you continue, check your understanding.

Question 1

Question 2

Question 3

Question 4

Question 5

Answers appear in the Self Test section of the book.