Genetics Of Intelligence and Cognitive Abilities Research Paper

1. Genetic Perspectives On Intelligence

Nearly all genetic research on human intelligence focuses on individual differences within our species. In contrast, cognitive psychology generally considers species-typical cognitive functioning, asking questions such as the involvement of working memory and a central executive in processing information. Imaging research asks what parts of the brain are engaged during particular tasks. Such research is more concerned about universal themes of human cognitive functioning rather than individual variations on these themes, which is the provenance of genetic research. These represent different perspectives and entail different levels of analyses. Perspectives are not right or wrong, merely more or less useful for asking particular questions. Genetic research cannot address the etiology of species-typical behavior but it is well suited to investigate the etiology of individual differences within a species.

Genetic research to date is also largely limited to a particular model of cognitive functioning called the psychometric model that considers cognitive abilities to be organized hierarchically (Carroll 1993) from specific tests to broad factors to general cognitive ability (often called g). There are hundreds of tests of cognitive abilities. These tests measure several broad factors (specific cognitive abilities) such as verbal ability, spatial ability, memory, and speed of processing. These broad factors are not independent— they intercorrelate modestly, about 0.40. That is, in general, people who do well on tests of verbal ability tend to do well on tests of spatial ability. General cognitive ability, that which is in common among these broad factors, was discovered by Charles Spearman nearly a century ago. It has been suggested that g is preferable to the word intelligence because the latter has so many different meanings in psychology and in the general language (Jensen 1998).

Most people are familiar with intelligence tests, often called IQ (intelligence quotient) tests. These tests typically assess several cognitive abilities and yield total scores that are reasonable indices of g. For example, the Wechsler tests of intelligence, widely used clinically, include ten subtests such as vocabulary, picture completion (indicating what is missing in a picture), analogies, and block design (using colored blocks to produce a design that matches a picture). In research contexts, g is usually derived by using a technique called factor analysis that weights tests differently according to how much they contribute to g. This weight can be thought of as the average of a test’s correlations with every other test. This is not a statistical abstraction—one can simply look at a matrix of correlations among such measures and see that all the tests intercorrelate positively and that some measures (such as spatial and verbal ability) intercorrelate more highly on average than do other measures (such as nonverbal memory tests). A test’s contribution to g is also related to the complexity of the cognitive operations it assesses. More complex cognitive processes such as abstract reasoning are better indices of g than less complex cognitive processes such as simple sensory discriminations (Jensen 1998).

Although g explains about 40 percent of the variance among such tests, most of the variance of specific tests is independent of g. Clearly there is more to cognition than g. Moreover, there may be more to g than can be seen from the types of psychometric tests used to assess it at the end of the twentieth century. There are other ways to study cognitive processes than psychometric tests such as information-processing approaches that typically rely on reaction time. Especially exciting are neuroscience measures that directly assess brain function such as evoked potentials, positron emission tomography (PET) scans, and functional magnetic resonance imaging (f MRI). As new tests of cognitive abilities are developed and their reliability and validity established, their relationship to g and their genetic and environmental origins will be investigated. However, few genetic studies have as yet used these other cognitive measures.

Just as there is more to cognition than g, there is clearly much more to achievement than cognition. Personality, motivation, and creativity all play a part in how well someone does in life. However, it makes little sense to stretch a word like intelligence to include all aspects of achievement such as emotional sensitivity and musical and dance ability that do not correlate with tests of cognitive ability.

Despite the massive data pointing to g, considerable controversy continues to surround g and IQ tests, especially in the media. There is a wide gap between what lay people (including scientists in other fields) believe and what experts believe. Most notably, lay people often hear in the popular press that the assessment of intelligence is circular—intelligence is what intelligence tests assess. To the contrary, g is one of the most reliable and valid measures in the behavioral domain, its long-term stability after childhood is greater than for any other behavioral trait, and it predicts important social outcomes such as educational and occupational levels better than any other trait (Gottfredson 1997). Although a few critics remain, g is widely accepted by experts. It is less clear what g is, whether g is due to a single general process such as executive function or speed of information processing, or whether it represents a concatenation of more specific cognitive processes (Jensen 1998).

2. Historical Context

The relative influence of nature and nurture on g has been studied since the beginning of psychology. Indeed in 1865, a year before the publication of Gregor Mendel’s seminal paper on the laws of heredity, Francis Galton published a two-article series on high intelligence and other abilities, which he later expanded into the first book on heredity and cognitive ability, Hereditary Genius: An Inquiry into its Laws and Consequences (1992; originally published in 1869). The first twin and adoption studies in the 1920s also focused on g. Highlights in the history of genetic research on g include Leahy’s (1935) adoption study, in which she compared IQ resemblance for nonadoptive and adoptive families. This study confirmed an earlier adoption study that showed genetic influence, in that IQ correlations were greater in nonadoptive than in adoptive families. The first adoption study that included IQ data for biological parents of adopted-away offspring also showed significant parent–offspring correlation, suggesting genetic influence (Skodak and Skeels 1949). Begun in the early 1960s, the Louisville Twin Study was the first major longitudinal twin study of IQ that charted the developmental course of genetic and environmental influences (Wilson 1983).

In 1963, a review of genetic research on g was influential in showing the convergence of evidence pointing to genetic influence (Erlenmeyer-Kimling and Jarvik 1963). During the 1960s, environmentalism, which had been rampant until then in American psychology, was beginning to wane, and the stage was set for increased acceptance of genetic influence on g. Then, in 1969, a monograph on the genetics of intelligence almost brought the field to a halt, because the monograph suggested that ethnic differences in IQ might involve genetic differences (Jensen 1969). Twenty-five years later, this issue was resurrected in The Bell Cur e (Herrnstein and Murray 1994) and caused a similar uproar. The causes of average differences between groups need not be related to the causes of individual differences within groups. The former question is much more difficult to investigate than the latter, which is the focus of the vast majority of genetic research on IQ. The question of the origins of ethnic differences in performance on IQ tests remains unresolved.

The storm raised by Jensen’s monograph led to intense criticism of all behavioral genetic research, especially in the area of cognitive abilities. These criticisms of older studies had the positive effect of generating a dozen bigger and better behavioral genetic studies that used family, adoption, and twin designs. These new projects produced much more data on the genetics of g than had been obtained in the previous 50 years. The new data contributed in part to a dramatic shift that occurred in the 1980s in psychology toward acceptance of the conclusion that g is significantly associated with genetic differences between individuals (Neisser et al. 1996).

3. Genetic Influence On G

Dozens of studies including more than 8,000 parentoffspring pairs, 25,000 pairs of siblings, 10,000 twin pairs, and hundreds of adoptive families all converge on the conclusion that genetic factors contribute substantially to g (Bouchard and McGue 1981, Plomin et al. 2001). Correlations for first-degree relatives living together average 0.43 for more than 8,000 parentoffspring pairs and 0.47 for more than 25,000 pairs of siblings. However, g might run in families for reasons of nurture or of nature. In studies involving more than 10,000 pairs of twins, the average g correlations are .85 for identical twins and 0.60 for same-sex fraternal twins. These twin data suggest that genetic factors play an important role in the origins of familial resemblance in g scores. Adoption studies also indicate substantial heritability. For example, in two recent studies, identical twins reared apart are almost as similar for g as are identical twins reared together, with an average correlation of 0.78 for 93 such pairs (Bouchard et al. 1990). Adoption studies of other first-degree relatives also indicate substantial heritability, as illustrated by recent results from the longitudinal 25-year Colorado Adoption Project (Plomin et al. 1997).

Estimates of the effect size of genetic factors, called heritability, vary from 40 to 80 percent but estimates based on the entire body of data are about 50 percent, indicating that genes account for about half of the variance in g. Even a meta-analysis that attempted to ascribe as much variance as possible to prenatal effects estimated heritability as 48 percent (Devlin et al. 1997). It is not of much importance whether heritability is 30, 50, or 70 percent because few implications would follow from one or the other estimate being correct. The point is that genetic influence on g is not only statistically significant; it is also substantial. Although heritability could differ in different cultures, moderate heritability of g has been found, not only in twin studies in North American and western European countries, but also in Moscow, former East Germany, rural India, urban India, and Japan (Plomin et al. 2001).

Most of the genetic variance for g is additive; that is, genetic effects add up rather than interact across loci—offspring resemble their parents genetically for additive genetic effects, not for interactive effects among genes. The additivity of most genetic effects on g may be due to the fact that there is greater assortative mating (nonrandom mating) for g than for other behavioral traits. Spouses correlate about 0.10 for personality, about 0.20 for height and weight, but for g spouses correlate about 0.45. Because bright women are likely to mate with bright men, this double-barreled effect makes their offspring likely to be brighter on average than would be expected if mating were at random. In this way, assortative mating spreads out the distribution of additive genetic effects on g in the population (Plomin et al. 2001).

These same genetic data provide the best available evidence for the importance of environmental factors independent of genetics. If genetic factors account for about half of the variance then non-genetic factors account for the rest of the variance. Environment clearly is important. This is suggested as well by the steady rise in IQ scores during the past several generations, which would seem too short a time to allow genetic explanations (Flynn 1999), and by studies in which children from abusive families show gains in IQ when adopted (Duyme et al. 1999). However, little is known about specific environmental differences as they relate to g independent of genetic factors.

4. Beyond Heritability

The answers to the rudimentary questions of whether and how much genetic factors contribute to g now seem sufficiently clear—the answers are, respectively, ‘yes’ and ‘a lot.’ Genetic research on g has moved beyond heritability, for example, to investigate developmental changes, multivariate relations among cognitive abilities, and specific genes responsible for the heritability of g.

4.1 Genetic Influence On G Increases During Development

When Francis Galton first studied twins in 1876, he investigated the extent to which the similarity of twins changes during development. Other early twin studies of g were also developmental, but this developmental perspective faded from genetic research until the late twentieth century. One of the most interesting findings about g is that heritability increases steadily from infancy (20 percent) to childhood (40 percent) to adulthood (60 percent). For example, a recent study of twins aged 80 years and older reported a heritability of about 60 percent (McClearn et al. 1997).

The 20-year longitudinal Colorado Adoption Project (CAP) confirms this finding using the adoption design (Plomin et al. 1997b). CAP is a 25-year study of 245 children adopted away from their biological parents at birth and adopted into adoptive homes in the first month of life. CAP includes ‘genetic’ (biological) parents and their adopted away children, ‘environmental’ (adoptive) parents and their adopted children, and ‘genetic-plus-environmental’ (nonadoptive or control) parents and their children who have been matched to the adoptive families. Correlations between non-adoptive parents and children increase from less than 0.20 in infancy to about 0.20 in middle childhood and to about 0.30 in adolescence. The correlations between biological mothers and their adopted-away children follow a similar pattern, indicating that parent-offspring resemblance for g is due to genetic factors. In contrast, parent–offspring correlations for adoptive parents and their adopted children hover around zero, which suggests that family environment shared by parents and offspring does not contribute importantly to parent–offspring resemblance for g. Because CAP is a parent–offspring design with adult parents and young offspring, these results can be interpreted more precisely as showing that genetic effects on adult g do not come on line just in adulthood, but also to a considerable extent in adolescence, to a lesser extent in childhood, and even to some slight extent in infancy.

Why does heritability of g increase during the life span? Perhaps completely new genes come to affect g as more sophisticated cognitive processes come on line. A more likely possibility is that relatively small genetic effects early in life snowball during development, creating larger and larger phenotypic effects, perhaps as individuals select or create environments that foster their genetic propensities.

4.2 Genetic Effects Are Broad (Molar) Rather Than Modular

In the widely accepted hierarchical model of cognitive abilities mentioned earlier, g consists of what is in common among specific cognitive abilities such as spatial, verbal, speed of processing, and memory abilities. Less is known about the genetic and environmental origins of individual differences in specific cognitive abilities, but they also appear to show substantial genetic influence, although less than g (Plomin and DeFries 1998).

To what extent do different sets of genes affect these phenotypically different abilities? A technique called multivariate genetic analysis examines covariance among specific cognitive abilities and yields a statistic called the genetic correlation, which is the extent to which genetic effects on one trait correlate with genetic effects on another trait independent of the heritability of the two traits. That is, although cognitive abilities are moderately heritable, the genetic correlations between them could be anywhere from 0.0, indicating complete independence, to 1.0, indicating that the same genes influence a variety of cognitive abilities. Multivariate genetic analyses have consistently found that genetic correlations among specific cognitive abilities are very high, close to 1.0 (Petrill 1997). In other words, the same genetic factors largely influence different abilities. What this finding means concretely is that if a specific gene were found that is associated with verbal ability, the gene would also be expected to be associated with spatial ability and other specific cognitive abilities.

These genetic results have major implications for current theories of cognitive neuroscience. According to one theory, the brain works in a modular fashion, that is, cognitive processes are specific and independent. Implicit in this perspective is a bottom-up reductionistic view of genetics in which modules are the targets of gene action. In contrast, the findings from multivariate genetic analyses suggest a top-down view in which genetic effects operate primarily on g, rather than a bottom-up view in which genetic effects are specific to modules. Given that the brain has evolved to learn from a variety of experiences and to solve a variety of problems, perhaps it makes sense that the brain functions holistically. However, finding genetic correlations near 1.0 does not prove that genetic effects are limited to a single general cognitive process that works in a top-down way. Another alternative is that specific cognitive abilities as they are currently assessed might involve many of the same modular processes that are each affected by different sets of genes. This alternative hypothesis could be tested by multivariate genetic research on measures of more modular processes such as neuroimaging measures of brain function.

4.3 Specific Genes Are Beginning To Be Identified

Heritability of complex dimensions such as g seems likely to be due to multiple genes of varying but small effect size rather than a single gene that has a major effect. Genes in such multiple-gene systems are called quantitative trait loci (QTLs) (Plomin et al. 1994). Unlike single-gene effects like PKU that are necessary and sufficient for the development of a disorder, QTLs contribute interchangeably and additively like probabilistic risk factors. Traditional methods for identifying single-gene effects are unlikely to succeed in identifying QTLs.

It is interesting that the two best-replicated QTLs are both in the cognitive domain. Apolipoprotein-E was first reported in 1993 to be related to late-onset dementia using allelic association (Corder et al. 1993) and has since been replicated in scores of studies (Rubinsztein 1995). QTL sib-pair linkage designs were used to identify a linkage between chromosome 6p21 and reading disability (Cardon et al. 1994), a linkage that has been replicated in several subsequent studies (e.g., S. E. Fisher et al. 1999). The QTL perspective suggests that both dementia and reading disability are likely to be the quantitative extremes of continuous distributions.

If additive genetic factors are important for complex traits, then it should be possible to identify the genes responsible if we have sufficient power to detect QTLs of small effect size. The critical question is the distribution of effects sizes of these QTLs: What is the average effect size of QTLs and how are QTL effect sizes distributed? If the average effect size is 1 percent we will eventually detect many QTLs given a heritability of at least 50 percent. However, if the average QTL effect size is 0.1 percent, we will detect very few QTLs and they will be difficult to replicate.

A QTL study applying new genetic approaches to g yielded a replicated association in a study comparing groups of children of high g and children of average g (Chorney et al. 1998). The gene is insulin-like growth factor-2 receptor (IGF2R) on chromosome 6, which has been shown to be especially active in brain regions most involved in learning and memory. The frequency of one of the alleles was twice as high in two groups of children with high g as compared to two groups of children with average g (about 30 percent vs. 15 percent). Combini ng these results yielded a highly significant result (χ²= 12.41, p <0.0004). An ongoing systematic scan of the genome using these techniques has begun to identify several other QTLs associated with g (P. J. Fisher et al. 1999).

5. Implications Of Identifying Genes Associated With G

Identifying replicable QTLs associated with g will make it possible to use measured genotypes rather than indirect inferences about heritable influence based on familial resemblance in order to address issues such as developmental change and continuity and the multivariate relationships among cognitive abilities. Replicated QTL associations with g will also provide discrete windows through which to view pathways between genes and g. Functional genomics, understanding how genes affect traits, is generally viewed in terms of bottom-up molecular biological analyses of cellular function in which the gene product is identified and its effects studied at a molecular and cellular level. However, other levels of analysis are also likely to be useful in understanding how genes affect g such as anatomical neuroimaging, functional neuroimaging, electrophysiology, psychophysiology, cognitive processing, and psychometrics. As an antidote to the tendency to define functional genomics at the cellular level of analysis, the phrase ‘behavioral genomics’ has been proposed (Plomin and Crabbe 2000). The grandest implication for science is that the functional (behavioral) genomics of g will serve as an integrating force across diverse disciplines with DNA as the common denominator, opening up new scientific horizons for understanding learning and memory.

As is the case with most important advances, identifying genes for cognitive abilities and disabilities will also raise new ethical issues. Finding QTLs for g will have important implications for society as well as science (Plomin 1999). In terms of implications for society, it should be emphasized that no policies necessarily follow from finding genes associated with g because policy involves values. For example, finding genes for g does not mean that we ought to put all of our resources into educating the brightest children. Depending on our values, we might worry more about children falling off the low end of the bell curve in an increasingly technological society and decide to devote more public resources to those who are in danger of being left behind.

Many ethical issues related to DNA are being broached at the level of single-gene disorders that are hard-wired in the sense that a single gene is necessary and sufficient for the development of the disorder. This will benefit ethical deliberations about the genetics of g, which seems less pressing because genetic effects on g are probabilistic rather than deterministic for two reasons. First, heritability is closer to 50 percent than to 100 percent, which means that non-genetic factors make a major contribution. Second, because many genes contribute to the heritability of g, the system is inherently probabilistic. Potential problems related to finding genes associated with g have been discussed such as prenatal and postnatal screening, discrimination in education and employment, and group differences (Newson and Williamson 1999).

The fear lurks in the shadows of such discussions that finding genes for g will limit our freedom and our free will. In large part such fears involve misunderstandings about how genes affect complex traits like g. Finding genes for g will not automatically open a door to a genetic version of Huxley’s brave new world where babies are sorted out at birth (or before birth) into alphas, betas, and gammas. Although the balance of risks and benefits to society of finding genes for g is not clear, basic science has much to gain from functional genomic studies of brain functions related to learning and memory. We need to be cautious and to consider carefully societal implications and ethical issues. There is also much to celebrate here in terms of the increased potential for understanding our species’ nonpareil ability to think and learn.

Bibliography:

1. Bouchard T J Jr, Lykken D T, McGue M, Segal N L, Tellegen A 1990 Sources of human psychological differences: The Minnesota study of twins reared apart. Science 250: 223–8
2. Bouchard T J Jr, McGue M 1981 Familial studies of intelligence: A review. Science 212: 1055–9
3. Cardon L R, Smith S D, Fulker D W, Kimberling W J, Pennington B F, DeFries J C 1994 Quantitative trait locus for reading disability on chromosome 6. Science 266: 276–9
4. Carroll J B 1993 Human Cognitive Abilities. Cambridge University Press, New York
5. Chorney M J, Chorney K, Seese N, Owen M J, Daniels J, McGuffin P, Thompson L A, Detterman D K, Benbow C P, Lubinski D, Eley T C, Plomin R 1998 A quantitative trait locus (QTL) associated with cognitive ability in children. Psychological Science 9: 1–8
6. Corder E H, Saunders A M, Strittmatter W J, Schmechel D E, Gaskell P C, Small G W, Roses A D, Haines J L, Pericak Vance M A 1993 Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science 261: 921–3
7. Devlin B, Daniels M, Roeder K 1997 The heritability of IQ. Nature 388: 468–71
8. Duyme M, Dumaret A-C, Tomkiewicz S 1999 How can we boost IQs of ‘dull children’?: A late adoption study. Proceedings of the National Academy of Sciences USA 96: 8790–4
9. Erlenmeyer-Kimling L, Jarvik L F 1963 Genetics and intelligence: A review. Science 142: 1477–9
10. Fisher P J, Turic D, McGuffin P, Asherson P, Ball D M, Craig I, Eley T C, Hill L, Chorney K, Chorney M J, Benbow C P, Lubinski D, Plomin R, Owen M J 1999 DNA pooling identifies QTLs for general cognitive ability in children on chromosome 4. Human Molecular Genetics 8: 915–22
11. Fisher S E, Marlow A J, Lamb J, Maestrini E, Williams D F, Richardson A J, Weeks D E, Stein J F, Monaco A P 1999 A quantitative-trait locus on chromosome 6p influences different aspects of developmental dyslexia. American Journal of Human Genetics 64: 146–56
12. Flynn J 1999 Searching for Justice: The discovery of IQ gains over time. American Psychologist 54: 5–20
13. Galton F 1992 Hereditary Genius: An Enquiry into its Laws and Consequences. World, Cleveland, OH
14. Gottfredson L S 1997 Why g matters: The complexity of everyday life. Intelligence 24: 79–132
15. Herrnstein R J, Murray C 1994 The Bell Cur e: Intelligence and Class Structure in American Life. Free Press, New York
16. Jensen A R 1969 How much can we boost IQ and scholastic achievement? Harvard Educational Review 39: 1–123
17. Jensen A R 1998 The g Factor: The Science of Mental Ability. Praeger, Westport, CT
18. Leahy A M 1935 Nature–nurture and intelligence. Genetic Psychology Monographs 17: 236–308
19. McClearn G E, Johansson B, Berg S, Pedersen N L, Ahern F, Petrill S A, Plomin R 1997 Substantial genetic influence on cognitive abilities in twins 80 years old. Science 276: 1560–3
20. Neisser U, Boodoo G, Bouchard T J, Boykin A W, Brody N, Ceci S J, Halpern D F, Loehlin J C, Perloff R, Sternberg R J, Urbina S 1996 Intelligence: Knowns and unknowns. American Psychologist 51: 77–101
21. Newson A, Williamson R 1999 Should we undertake genetic research on intelligence? Bioethics 13: 327–42
22. Petrill S A 1997 Molarity versus modularity of cognitive functioning? A behavioral genetic perspective. Current Directions in Psychological Science 6: 96–9
23. Plomin R 1999 Genetics and general cognitive ability. Nature 402: C25–9
24. Plomin R, Crabbe J C 2000 DNA. Psychological Bulletin 126(6): 806–28
25. Plomin R, DeFries J C 1998 Genetics of cognitive abilities and disabilities. Scientific American 278(5): 62–9
26. Plomin R, DeFries J C, McClearn G E, McGuffin P 2001 Behavioral Genetics. Worth, New York
27. Plomin R, Fulker D W, Corley R, DeFries J C 1997 Nature, nurture and cognitive development from 1 to 16 years: A parent-offspring adoption study. Psychological Science 8: 442–7
28. Plomin R, Owen M J, McGuffin P 1994 The genetic basis of complex human behaviors. Science 264: 1733–9
29. Rubinsztein D C 1995 Apolipoprotein-E—A review of its roles in lipoprotein metabolism, neuronal growth and repair and as a risk factor for Alzheimers-disease. Psychological Medicine 25: 223–9
30. Skodak M, Skeels H M 1949 A final follow-up on one hundred adopted children. Journal of Genetic Psychology 75: 84–125
31. Wilson R S 1983 The Louisville Twin Study: Developmental synchronies in behavior. Child Development 54: 298–316