This paper reflects the research and thoughts of a student at the time the paper was written for a course at Bryn Mawr College. Like other materials on Serendip, it is not intended to be "authoritative" but rather to help others further develop their own explorations. Web links were active as of the time the paper was posted but are not updated.

Contribute Thoughts | Search Serendip for Other Papers | Serendip Home Page

Biology 202, Spring 2005
First Web Papers
On Serendip

Understanding Intelligence

Xuan-Shi, Lim

What is intelligence? While it is common to relate intelligence to IQ, or intelligence quotient, one should understand that IQ is a social construct. It refers to the scores on psychometric intelligence tests, which are constructed to measure qualities that enable people to be successful within that culture (1). Although intelligent behavior has different manifestations across and within cultures, it is intuitive to think that there may be underlying similarities in the brains of intelligent people. Does higher intelligence equate to larger brains and/or more synaptic connections between neurons? Can intelligence be localized to specific brain regions? This paper examines the major theories of intelligence and attempts to understand the concept of intelligence in terms of neuroanatomy.

Presently, theories of intelligence are divided into two camps: the psychometric and multiple intelligences approaches. Intelligence tests, such as the Wechsler tests, are typical psychometric instruments used to measure general intelligence, or g, for assessment and research purposes. The g factor was first proposed by Charles Spearman who also developed factor analysis (2), a statistical tool that has uncovered correlations among people's performance on groups of test items; this suggests that g underlies groups of specific abilities, as outlined in Spearman's two-factor theory of intelligence (3). Test items or tasks that involve a high degree of complexity have also been found to tap more heavily on g (4); one example is the Raven's Advanced Progressive Matrices (5). Thus, there is reason to believe that g is related to cognitive abilities, although g is not a cognitive ability by itself (2). Other researchers have since expanded on the concept of g. Cattell and Horn proposed that there are many types of g, including fluid (Gf) and crystallized (Gc) intelligences; Gf is "essentially nonverbal, relatively culture-free mental efficiency," while Gc refers to the skills and information acquired through acculturation (6). Carroll later superimposed a g factor, akin to Spearman's g, above the different types of general mental abilities, which also include Gf and Gc; the general abilities are in turn composed of specific abilities, such as general reasoning and induction that comes under Gf (6).

Howard Gardner and Robert Sternberg, proponents of the multiple intelligences approach, believe that intelligence consists of several constructs. Briefly, Sternberg's successful intelligence theory states that intelligence is comprised of three components: practical, analytic, and creative abilities (1). Gardner's multiple intelligences theory consists of seven intelligences: linguistic, logical-mathematical, spatial, bodily kinesthetic, musical, interpersonal, and intrapersonal. He has since suggested other intelligences: naturalist, spiritual, and existential (7). Gardner believes that everyone possesses seven intelligences that are in constant interaction with one another; they are utilized in different combinations to complete a task (8). Gardner's theory is of special interest to this paper because he has attempted to map out the brain areas associated with each intelligence, presumably drawing most of his data from clinical studies of patients with brain injuries (8). The observations that selective damage to a brain area impairs only a specific ability or "intelligence" and leaves other abilities unaffected suggest to Gardner that there are different intelligences, each of which may be localized to different brain regions.

Sternberg and Kaufman observed that since 1997, there were no empirical studies to test whether the multiple intelligences theory is valid (1). Despite its unpopularity among researchers, Gardner's theory is well-received by many in the field of education, which also uses psychometric tests to assess students. The fact that intelligence tests measure Gc, in addition to Gf, may contribute to the association of g with performance in academics. According to Gardner, most intelligence tests focus mainly on "linguistic and logical faculties" (5); traditionally, schools have nurtured these abilities or "intelligences." Seen in this light, Gardner's theory expands the concept of intelligence beyond what is measured on IQ tests, acknowledging performance in other domains. Gardner's expansive concept of intelligence is complementary to the idea that learners are unique individuals, with different strengths and weaknesses. However, does this imply that psychometric g is a reductionistic model of intelligence?

Psychologist Arthur Jensen believes that the "g factor reflects individual differences in information processing (9)." According to Jensen, tests that yield an IQ score correlate to some extent (6). Because these cognitive tests differ in their content and requisite skills, Jensen reasons that the correlation among them is not contaminated by the tests or the statistical tool of factor analysis; instead, loadings on g reflect an attribute of the brain (9). It seems reasonable for Jensen to view g as closely related to information processing capacity. At the level of the organism, information processing encompasses a wide range of cognitive functions such as attention, memory, and problem-solving. At the cellular level, a neuron both receives and transmits signals. If g is related to information processing, and given that information processing is a fundamental aspect of our nervous system, it is possible that g has a biological basis.

How are the brains of individuals with high g and average g different? In terms of neuroanatomy, MRI studies has previously shown that there is about a +0.4 correlation between brain size and IQ (9). More recently, Richard Haier et al (2004) first discovered that more gray matter in specific brain regions is associated with higher IQ, as measured by the Wechsler Adult Intelligence Scale (WAIS) (10). These regions include several Brodmann areas within the frontal, temporal, and parietal lobes. Interestingly, Gardner has also associated these lobes with several intelligences, such as interpersonal and logical-mathematical (8), although he did not name specific areas. Haier's study appears to endorse the idea that differing strengths and weaknesses of people with the same IQ arise as a result of differences in cognitive abilities associated with information processing, rather than from interactions among various intelligences. Such cognitive differences may be explained by the variation of gray and white matter volumes in brain areas that are associated with IQ (10).

If it is assumed that an individual with a high IQ score has a high g, then it is questionable whether g may be associated with gray and white matter location and volume. Is g an abstract function or does it has a material existence? With respect to Gardner's theory, one wonders whether poets have more gray matter in the same frontal areas identified by Haier's study, when compared to scientists with the same IQ. Does more gray and white matter in specific IQ-associated brain areas also translate into enhanced performance in certain domains under Gardner's theory? Theoretically, psychometric g and multiple intelligences theories are starkly different in the way each has conceptualized intelligence. In reality, the two theories may be different interpretations of the same neural map, as the existence of g is contingent on the variety of abilities or "intelligences" which one possesses.

Haier's study has also reported that the location of increased gray matter associated with IQ varies across age groups, i.e. young and middle-aged adults (10). Specifically, more gray matter in the frontal and parietal lobes are associated with IQ in the middle-aged cohort, while in young adults, more gray matter in the temporal lobe and less in the frontal lobe are associated with IQ. Comparing gray matter within the frontal brain region, the researchers found that the location of gray matter associated with g varies across different age groups. Haier et al suggest that the shifting pattern may be a response to age-related neuronal loss that is more likely to affect the frontal region, particularly the anterior cingulated gyrus, an area reported by Wilke et al (2003) to be associated with IQ in children (10). Given that Gf decreases with age while Gc remains relatively stable, could these findings relate to decreases in Gf? Taken together, these findings provide a dynamic and developmentally-sensitive view of intelligence that is influenced by biological mechanisms. Although Gardner's theory has also taken development into account, it appears that experience is the primary factor that brings about changes in the brain; Gardner claims that the seven multiple intelligences evolve with age and follow developmental trajectories in terms of emerging expertise.

To conclude, both the psychometric and multiple intelligences approaches contribute to one's understanding of intelligence. Studies emerging from neuroscience and cognitive psychology are more promising because they help one to better understand intelligence, as the g factor, in terms of individual differences in neurological and information processing systems. The multiple intelligences approach as a whole seems to convey the important reminder that intelligence is also influenced by experience and that intelligence manifests itself in different aspects of behavior. Clearly, how a society defines intelligence is likely to affect the ways in which it allocates resources among individuals. Therefore, the presence of different theories of intelligence is necessary to emphasize the idea that intelligence is not a fixed and concrete entity (11), which may be measured by culturally biased intelligence tests.


1)Human Abilties, An article by Sternberg and Kaufman that explores the definition of intelligence and provides a good summary of theories from past to present

2)The G Factor: the Science of Mental Ability, An article by Arthur Jensen, provides comprehensive and critical overview of psychometric g

3)Key Players in the History and Development of Intelligence and Testing, A very brief but organized introduction to the major intelligence theories

4)The General Intelligence Factor, A reader-friendly article on g that also discusses how g is measured

5)The Importance of Spearman's g, Reviews major intelligence theories and contains an informative section on the importance of g as an "educational and social construct"

6) Sattler, J. M. Assessment of Children: Cognitive Applications. 4th Edition. San Diego: J. M. Sattler, 2001.

7)Howard Gardner and Multiple Intelligences, Evaluates Gardner's multiple intelligences theory

8) Armstrong, T. Multiple Intelligences in the Classroom. Virginia: Association for Supervision & Curriculum Development, 2000.

9) Jensen, A. R. "The g Factor and the Design of Education." Intelligence, Instruction, and Assessment. Ed. Robert J. Sternberg & Wendy M. Williams. New Jersey: Lawrence Erlbaum Associates, Inc, 1998.

10)Structural Brain Variation and General Intelligence, The first study to report that more gray matter in specific brain areas is associated with IQ

11)Current Views of Intelligence Testing, 1997 Lecture notes created by Elizabeth Johnson

| Course Home | Serendip Home |

Send us your comments at Serendip

© by Serendip 1994- - Last Modified: Thursday, 19-Jul-2012 20:33:46 EDT