Advances in the Psychology of Human Intelligence

Volume 4

Robert J. Sternberg, Robert J. Sternberg

• Publisher: Psychology Press
• Year: 2014
• ISBN:
  9781317785958
• Edition: 1
• Topic: Psychologie
• Subtopic: Histoire et théorie en psychologie

First published in 1987. Since the 1960s, we have witnessed a remarkable resurgence of interest in the psychology of human intelligence. In the late 1960s, research in the field of intelligence seemed to have gone into at least partial remission. But today, a large number of investigators are pursuing active research programs concerning human intelligence. Advances in the Psychology of Human Intelligence, of which this is the fourth volume, contains chapters by leaders in the field that document the progress being made toward understanding human intelligence.

1	The Nature and Nurture of Cognitive Abilities

Robert Plomin

The Pennsylvania State University

Nearly a century ago, Francis Galton (1883) coined the alliterative phrase, nature-nurture, a phrase that has come to evoke a strong association with the word “controversy,” especially when the implicit conjunction between nature and nuture is versus. Fortunately, the conjunction and appears to be in ascendance; nature and nurture is the theme of this chapter.

During the past decade, recognition and acceptance of a role for genetics in complex human behavior, including intelligence, has increased dramatically. Indeed, we may be nearing the time when it will become important to prevent the pendulum of fashion from swinging too far in the direction of genetic influence. The purpose of this chapter is to review new findings from the field of behavioral genetics that demonstrate a critical point: Behavioral genetic research provides strong evidence for the importance of environmental influences as well as genetic influence on individual differences in cognitive abilities, and suggests new ways to think about these environmental influences.

This chapter emphasizes research rather than methodology; behavioral genetic methodologies relevant to the study of intelligence have recently been reviewed (Plomin, 1985). The chapter begins with a general discussion of the relationship between behavioral genetic research and intelligence, reviews the results of behavioral genetic research on general and specific cognitive abilities, and then presents a provocative hypothesis about the environment that emerges from this research.

INDIVIDUAL DIFFERENCES

It is critical at the outset to recognize that behavioral genetic research addresses the origins of individual differences in a population. Its goal is to partition observed variance for a particular measure into genetic and nongenetic components of variance. It has little to say about the causes of average group differences (e.g., why females excel on average for tests of verbal ability) and it has nothing to say about normative or universal features of development (e.g., why the human species is a natural language user). This distinction is important because the description and explanation of group differences bear no necessary relationship to the description and explanation of individual differences.

For example, individual differences in verbal ability could be largely due to genetic differences among individuals, yet the average difference between males and females could be entirely environmental in origin. The best known example of the confusion that results when the two approaches are not distinguished is the argument that genetic influence on intelligence within ethnic groups implies that average differences between ethnic groups are also due to genetic factors (Mackenzie, 1985). Another example is the frequently read statement that the effects of nature and nurture cannot be disentangled because “there can be no behavior without an organism and there can be no organism without genes” (Gottlieb, 1983, p. 5). Behavioral genetic research does not study an organism or the organism; its focus is on variance, differences among individual organisms in a population. We can assess the extent to which observed variance for a characteristic is due to environmental factors—as legions of environmental studies have attempted to do—and we can estimate the contribution of genetic differences among individuals.

This distinction between individual differences and group differences is important in relation to research on intelligence because much cognitive research, especially information-processing approaches and developmental research such as that of Piaget, is normative. Behavioral genetic research is not relevant to these legitimate normative concerns. Furthermore, it has been my experience that researchers who are not interested in individual differences will not be interested in behavioral genetics.

There are good reasons to be interested in individual differences, however. First of all, individual differences are real and must be part of a comprehensive theory. Ernst Mayr (1982) refers to the distinction between group differences and individual differences as essentialism versus population thinking and suggests that it represents a fundamental schism in biology. His view is that “the differences between biological individuals are real, while the mean values which we may calculate in the comparison of groups of individuals (species, for example) are man-made inferences” (p. 47).

A second, and more controversial, reason for studying individual differences is that behavioral issues of greatest relevance to society are issues of individual differences. Although, as scientists, we might like to know why children in the human species are natural language users, society is undoubtedly more concerned about individual differences: For example, why are some children significantly delayed in their use of language? Societal interest in psychology comes from the study of things that make a difference—the description, prediction, and explanation of individual differences.

Third, differences between groups are usually trivial in magnitude compared to individual differences within groups. For example, one of the best documented gender differences in cognition is that girls exceed boys on tests of verbal ability (Maccoby & Jacklin, 1974). However, this average difference accounts for less than 1% of the variance of verbal ability (e.g., Plomin & Foch, 1981). Finally, I believe that questions concerning the origins of individual differences are more easily answered than questions about the etiology of group differences. Although average group differences can easily be described, going beyond descriptions to explanations is difficult. Elaborate theories may be concocted to explain average group differences or normative phenomena, but the core explanatory constructs are seldom tested or even testable. Using Piagetian theory as one of a multitude of examples, assimilation and accommodation are posited as the essential explanatory processes of development, but how can we test whether these processes indeed explain the dramatic leaps in mental development that occur on average during the transition from infancy to early childhood?

Once an interest in individual differences is evoked—especially when that interest goes beyond description to explanation—the armamentarium of behavioral genetics can be appreciated for its power to take a reasonable first step in understanding the etiology of individual differences by asking the extent to which observed differences are due to individuals’ genetic differences or to experiential differences.

QUANTITATIVE GENETIC THEORY

Although the methods of behavioral genetics—family, twin, and adoption designs—have been frequently discussed, it is less often recognized that a powerful theory underlies these methods. It may be worthwhile describing this theory briefly because one reason for the relative disregard of individual differences is that such research often appears atheoretical at the descriptive and predictive levels of analysis and, especially, at the level of explanation. In my view, quantitative genetic theory, which underlies behavioral genetic research, offers a theory of unparalleled power in the behavioral sciences for understanding the origins of individual differences. It is a general theory that is not limited to a specific substantive domain such as intelligence.

Focusing on the pragmatic view of a psychological researcher rather than philosophy of science issues, theories should clarify our thinking by describing, predicting, and explaining behavior. At a minimum, a theory should be descriptive—organizing and condensing already existing facts in a reasonable, internally consistent manner. It should also make predictions concerning phenomena not yet investigated and permit clear tests of these predictions. At its best, a theory explains phenomena in addition to describing and predicting them. It is in terms of these criteria that I suggest that quantitative genetic theory is so powerful.

Most fundamentally, quantitative genetics organizes a welter of data on individual differences so that they are no longer viewed as imperfections in the species type or as nuisance errors in analyses of variance, but rather as the quintessence of evolution. Genetic variance is the raw material without which evolution cannot occur.

In terms of genetics, the essence of quantitative genetic theory is that Mendel’s mechanism of discrete inheritance also applies to normally distributed complex characteristics if we assume that the effects of many genes, each with slight influence, add up to produce observable differences among individuals in a population. This component of the theory was worked out in the early 1900s as a resolution to a dispute between Mendelians, who looked for single-gene segregation ratios, and those who felt that Mendel’s laws of inheritance, derived from experiments with qualitative characteristics in pea plants, were not applicable to complex characteristics in higher organisms, which are nearly always distributed continuously as a normal, bell-shaped curve (Fisher, 1918). In other words, if more than three or four genes affect a trait, the observed distribution cannot be distinguished from a normal curve. For example, a trait influenced by two alleles at each of three loci yields 27 different genotypes. Even if the alleles at the different loci equally affect the trait and there is no environmental variation, seven different phenotypes can occur and their distribution will be indistinguishable from a normal curve. For this reason, behavioral phenomena demand a polygenic (multi-gene) approach—the complexity of these phenotypes makes it unlikely that any single gene can account for much variance in the population. Although single-gene mutations can be devastating for affected individuals—such as those who are severely retarded as a result of untreated phenylketonuria—no single gene has been found to account for a detectable amount of variance in the normal range of individual differences in IQ scores.

The theme of this chapter is that quantitative genetic theory is as important for understanding environmental sources of variance as for understanding genetic sources. Indeed, the label quantitative genetics is misleading because the theory and its methods are as informative about environmental components of variance as they are about genetic factors. As discussed later, the most important contemporary contributions of behavioral genetic research to the study of intelligence involve nurture rather than nature.

Quantitative genetic theory is unlike most contemporary theories in the behavioral sciences in that it is not limited to a particular substantive domain such as intelligence. It is analogous in this respect to learning theory, which does not predict what is learned but rather the processes by which learning occurs. Quantitative genetics is also formally similar to learning theory in that neither specifies molecular mechanisms. That is, learning theory does not predict which neurotransmitters are involved in the learning process, nor does it even predict that neurotransmitters are involved at all. Although quantitative genetics is based on the proposition that variation in DNA leads to phenotypic variation, it does not specify which genes or environmental factors are responsible for phenotypic variance. By the end of this century, it will be possible to identify DNA segments responsible for genetic sources of variance even for behavioral traits for which hundreds of genes may each contribute miniscule portions of variance. Although specific environmental factors are not predicted by the theory, quantitative genetics provides some powerful tools for the exploration of environmental influences, as discussed later.

In order to emphasize the descriptive, predictive, and explanatory powe...