GENETICS AND BEHAVIOR

Despite longstanding hostility to the biological explanation of human behavior, there are presently three general research programs aimed at the study of genetic influences on behavior: sociobiology and evolutionary psychology, behavioral genetics, and developmental psychobiology. Evolutionary psychology and its forebear, sociobiology, aim to discover species-typical traits that are adaptations (that is, traits that are in most cases the result of natural selection): Why do humans behave aggressively? What is the evolutionary source of altruism? Behavioral genetics aims primarily to uncover and disentangle genetic contributions (as distinct from environmental contributions) to individual differences in behavior: What are the predictors of aggressive versus nonaggressive behavior? Why does one person perform well on an IQ test, and another not? Developmental psychobiology aims to elucidate developmental pathways to particular behaviors: What is the mechanism by which organisms come to behave aggressively? What are the determinants of central nervous system development?

Such sample questions are by no means exhaustive; they are meant simply to illustrate the focal differences between these three approaches to genetics and behavior (see Table 1), the latter two of which will be the focus here. That is, rather than focus on how biological evolution as a whole has affected species-specific behaviors, the emphasis will be on how genetics can account for individual differences within species and on the more detailed pathways by which DNA causally influences human behavior.

Born of Controversy

Both behavior geneticists and developmental psychobiologists aim to move beyond the nature/nurture dichotomy, according to which traits are either genetically influenced (nature) or environmentally influenced (nurture), in favor of some collaborative interaction. What nature-and-nurture or nature-via-nurture actually means in practice is not always clear, however, because most scientists continue to partition correlational and causal influence in traditional terms (Schaffner 2001, Robert 2003).

The modern roots of the nature–nurture controversy are to be found in the writings of Francis Galton (1822–1911), a cousin of Charles Darwin, in the latter half of the nineteenth century. In 1869 Galton published Hereditary Genius, in which he attempted to discern what makes some humans geniuses and others exceptionally stupid. Based in part on anecdotal observations of twins, along with a questionnaire he administered to a small group of twins who were believed to be more similar in their youth than at the time of testing, Galton eventually concluded that "nature prevails enormously over nurture" in explaining variance in cognitive

outcome (Galton 1875, p. 576). Galton later coined the term eugenics as part of a program to increase the number of so-called desirables in a population and to decrease the number of so-called undesirables (Kevles 1985). The "eugenics movement" has, of course, had its own very controversial history—including the rationalization of human rights violations in the United States, Nazi Germany, and other countries.

Since its modern incarnation, then, and however well-intended, behavior genetics has been associated with the justification of class-based and racial prejudice, exemplified more recently with the argument of Arthur R. Jensen (1969) that genetic differences between "races" influence the lower intelligence (or the poorer performance on IQ tests) of blacks as compared with whites. While most behavior geneticists have disowned this and related work, in 1995 the outgoing president of the Behavior Genetics Association (BGA), Glayde Whitney, celebrated Jensen's putatively brilliant and bold 1969 work in his presidential address. Whitney's speech was widely disparaged, and the editor of the BGA journal, Behavior Genetics, refused to publish it.

Classical Behavior Genetics

Three key concepts in classical genetics that referred originally not so much to behavioral but to anatomical characteristics that need to be clarified are genotype, phenotypes, and allele. The genotype is simply the genetic make-up of the organism, its complement of DNA. Genes, now known to be sections of chromosomes, manifest themselves as the organism's phenotype, its outward appearance. Any one gene may also come in different or alternative forms called alleles. For example, the founder of genetics, Gregor Mendel (1822–1884), in his research with pea plants, identified that one gene controls seed color, and the two forms of this gene give either green or yellow peas. That is, one allele (for yellow pea color) will be expressed as one phenotype (yellow peas), whereas another allele (for green pea color) will be expressed as another phenotype (green peas). One question for behavior genetics is whether and to what extent there are genotypes with different alleles that control for phenotypical behavior as well as physical characteristics.

The attempt to answer this question through the practice of observing twins continues to this day—though now with considerably more sophistication and computational power. Modern behavior geneticists establish correlations between genes and behavioral outcomes on the basis of two general types of study, involving classical or quantitative genetics (family, twin, and adoption studies) and molecular genetics and genomics (linkage, association, allele sharing, quantitative trait locus mapping, and DNA microarray studies). Although it is not necessary to know the complete meaning of the technical terms here, linkage and association refer to kinds of connections between genes, alleles (as already explained) are different forms of the same gene, trait locus mapping seeks to locate genes at specific points on a chromosome, and DNA microarray studies aim to show which genes are expressed at any given time. Classical studies are used to reveal the relationship between genetic variation and variation in phenotypic outcome.

Twin studies, for instance, are premised on the notion that, on average, identical (monozygotic, or MZ) twins share almost 100 percent of their genes in common, while fraternal (dizygotic, or DZ) twins share approximately 50 percent of their genes in common. A fundamental assumption is that both kinds of twins are affected by their rearing environments in a similar way, and that their "equal environments" cannot make MZ twins any more alike than DZ twins. On the basis of this assumption, behavior geneticists argue that what makes MZ twins more alike than DZ twins is that they are more genetically similar.

In any given population, heritability refers to the proportion of phenotypic (or apparent, expressed) variance that can be explained by genotypic (or hidden, genetic) variation, and is quantified as between zero (no variation explained by genetic inheritance) and one (all variation explained by inheritance). In humans, the heritability of having two legs is just about zero: Because almost all humans are born with two legs, there is very little phenotypic variance to be explained. By contrast, the heritability of eye color in a random human population approaches one, inasmuch as the variation in eye color can be explained almost exclusively by genetic variance. T the heritability of height is somewhere in between. Like physical characteristics, behaviors of interest to behavior geneticists have nonzero heritability (often in the range of 0.4 to 0.6), though it is often unclear what inferences are justified on the basis of a heritability estimate (Turkheimer 1998).

Behavior geneticists distinguish between traits that are either present or absent, and those that are continuously distributed. Where presence/absence is appropriate, scientists calculate concordance rates. Where the trait is continuous, scientists calculate correlation coefficients. So if MZ twins both exhibit some noncontinuous phenotypic outcome (say, depression), they are said to be "concordant" for that trait; and where the concordance rate for MZ twins is greater than that for DZ twins, the greater concordance is attributed to genes. Where MZ and DZ concordance rates are similar, this is attributed to shared environmental influences. And where MZ twins are discordant for a trait, this is attributed to nonshared environmental influences. In many cases, genes, shared environment, and nonshared environment are invoked to partially explain phenotypic differences (Baker 2004, Parens 2004), although nonshared environmental effects remain very difficult to discern (Turkheimer 2000).

Molecular Behavior Genetics

Classical studies can reveal associations between genetic variance and phenotypic variance, but do not identify the particular genes that may generate a trait. In the 1980s, behavior geneticists began to take advantage of emerging molecular techniques to attempt to identify specific genes. Linkage studies are employed to detect genes of major effect shared by a disproportionately large number of family members manifesting a condition or trait of interest. Successful linkage studies require three conditions to have been met: that a gene of major effect is implicated; that there is only one such gene segregating in a given family; and that the mode of inheritance is known (Robert 2003). For most complex behaviors, at least one of these conditions is violated; for many complex behaviors, all three are violated.

Allelic association studies are employed to discern whether alleles or different forms of particular genes are transmitted preferentially to family members, or whether there are differences in the frequency of alleles between individuals and control populations. These studies avoid the requirement for a single gene of major effect; moreover, in the company of now-possible genome-wide scans, there is no need even to identify candidate genes or regions in order to turn up possible correlations. Further, success with these studies does not depend on knowing a specific mode of inheritance. But correlations are not causes, and allelic association studies risk turning up correlations that are causally spurious. For instance, where an allele is in linkage disequilibrium with another allele, allelic association studies will positively correlate both alleles with the phenotype, even if only one is actually involved in generating the phenotype.

Behavior geneticists are now using still more sophisticated techniques to reveal associations between genes and phenotypic outcomes. These include quantitative trait locus mapping and DNA microarray technology. Most phenotypes, especially of behaviors, are complex combinations of traits and thus governed by more than one gene. Quantitative trait locus mapping attempts to determine in quantitative terms what set of traits define a complex phenotype. DNA microarray technology, using what is variously called a biochip, DNA chip, or gene chip, allows for large-scale gene expression studies in order to identify interacting genes. Progress has nevertheless been slower than initially anticipated (e.g., Hamer 2002), and very few specific genes have been identified.

According to behavior geneticist Michael Rutter (2002), "knowing that a trait is genetically influenced … is of zero use on its own in understanding causal mechanisms" (p. 4). Some developmental psychobiologists take this as evidence of the sterility of behavior genetics (e.g., Gottlieb 1995). If the focus of behavior genetics is on the establishment of correlations and other associations between inherited genes and particular behavioral outcomes, the focus of developmental psychobiology is on the identification of the developmental pathways that lead to those outcomes. Often, these pathways involve heritable elements, including genes; sometimes, other levels of analysis are more apt to yield developmental insights.

Developmental Psychobiology

Behavior geneticists do not study behavior as such, but rather differences in behavior. Moreover, behavior geneticists study associations between genetic variance, environmental variance, and interactions between the two, not causal relationships between developmental factors. By contrast, developmental psychobiologists seek to unpack genetic and other influences on complex behavioral phenotypes by elucidating causal mechanisms and pathways within the developing organism.

There is a long history of research in animal behavior (ethology) and comparative psychology, including experimental studies of animal behavior. Many historians begin with the work of Konrad Lorenz (1903–1989) on innateness. Lorenz's research was not entirely well-received among "English-speaking ethologists," as he called them, particularly Daniel S. Lehrman (1919–1972). Lehrman's criticisms of Lorenz (1953) continue to inspire developmental psychobiologists (e.g., Johnston 1987, Lickliter 2000, Oyama 2000), while classical ethology has generally been dislodged by sociobiology and evolutionary psychology. (Developmental criticisms of the concept of innateness preceded the work of Lorenz; see, for instance, Kuo 1921.)

Experimental analyses of animal behavioral development have revealed aspects of development from conception through senescence, including factors, mechanisms, and causal interactions involved in central nervous system development. Coupled with results from brain science, developmental psychobiologists are shedding light on the pathways of neural, cognitive, and motor development in a wide range of animals, including those chosen as models for understanding human development. Nonetheless, developmental psychobiology has yet to yield a fully integrative account of behavioral development, in large part because of the complexity of the task.

Yet a framework for the integrative project is now in place. Timothy D. Johnston and Laura Edwards's series of increasingly specific (or "unpacked") representations of a model of the development of behavior are not intended to specify every molecular or cellular aspect of the complexity of development, but rather to provide "a useful intermediate level of detail that captures that complexity while at the same time rendering it reasonably comprehensible" and open to empirical investigation (Johnston and Edwards 2002, p. 31). Genes, neurons, and experience have indirect and reciprocal effects on the development of behavior, though their activity is mediated through multiple levels of biological, ecological, and social organization. The model is meant to focus investigative attention on developmental interactions and specific mechanisms, as depicted in Figure 1.

Any particular concrete use of this model would represent only a snapshot of a specific developmental moment. The model could also be transformed from two dimensions to three with the addition of information regarding the timing of individual influences on development, though this would obviously make it considerably less easy to represent graphically. This model of behavioral development can be used to organize existing knowledge and to make predictions about behavioral development that can be empirically investigated, yielding support for or requiring alteration of the underlying model.

In using this model of behavioral development in a research context, it is evident that scientists cannot do the kinds of studies with humans that would yield results of interest. There are limits on what is acceptable with human subjects. Accordingly, developmental psychobiologists (like all developmental researchers) must infer from animal models, a process that is both conceptually and ethically fraught (Gottlieb and Lickliter 2004). Are the behaviors observed (or created) in animal models in fact homologous (or even analogous) to human behaviors? How does a passive–aggressive rat or an alcoholic monkey behave? What can be learned about human neural development from a fruit fly? These challenges beset any attempt to understand human behavioral development on the basis of studies with nonhuman animals.

Ethical and Social Considerations

While both behavior genetics and developmental psychobiology continue to provide important insights into the development of behavior, ethical concerns persist. These range from eugenic fears about the discovery of so-called gay genes and genes predisposing to antisocial behavior, to worries about the possible genetic enhancement of human cognitive function.

Following the mapping and sequencing of the human genome, a project that was sometimes viewed in exaggerated terms, there has been a shift to functional genomics, that is, attempts to determine what genes do and how they interact. Some hope that functional genomics will tell us not just how genes produce certain proteins but also how genes produce phenotypes, including behavior. But according to one policy commentary in Science magazine:

The genetics of behavior offers more opportunity for media sensationalism than any other branch of current science. Frequent news reports claim that researchers have discovered the "gene for" such traits as aggression, intelligence, criminality, homosexuality, feminine intuition, and even bad luck. Rarely is it mentioned that traits involving behavior are likely to have a more complex genetic basis. This is probably because most journalists—in common with most educated laypeople (and some biologists)—tend to have a straightforward, single-gene view of genetics. (McGuffin et al. 2001, p. 1232)

Thus there is clearly a place for the lowering of expectations with regard to behavioral genetics.

More broadly, though, simply to study genetics and behavior by any means is to study what makes humans behaviorally different from one another. For many, any advances in this domain threaten to impinge, at least conceptually, on precisely what it is that distinguishes human from nonhuman nature. While these concerns may be ill-founded, behavioral scientists must take seriously the imperative to assuage these fears by promoting socially responsible public engagement with the science.

JASON SCOTT ROBERT

SEE ALSO Bioethics;Genethics;Genetic Research and Technology.

BIBLIOGRAPHY

Baker, Catherine. (2004). Behavioral Genetics: An Introduction to How Genes and Environments Interact through Development to Shape Differences in Mood, Personality, and Intelligence. Washington, DC: American Association for the Advancement of Science.

Galton, Francis. (1869). Hereditary Genius: An Inquiry into Its Laws and Consequences. London: Macmillan. A classic text that initiated behavioral genetics.

Galton, Francis. (1875). "The History of Twins, as a Criterion of the Relative Powers of Nature and Nurture." Fraser's Magazine 12(92): 566–576. Revised version published in Journal of the Royal Anthropological Institute, vol. 5, pp. 391–406.

Gottlieb, Gilbert. (1995). "Some Conceptual Deficiencies in 'Developmental' Behavior Genetics." Human Development 38(3): 131–141.

Gottlieb, Gilbert, and Robert Lickliter. (2004). "The Various Roles of Animal Models in Understanding Human Development." Social Development 13(2): 311–325.

Hamer, Dean. (2002). "Rethinking Behavior Genetics." Science 298(5591): 71–72.

Jensen, Arthur R. (1969). "How Much Can We Boost IQ and Scholastic Achievement?" Harvard Educational Review 39(1): 1–123.

Johnston, Timothy D. (1987). "The Persistence of Dichotomies in the Study of Behavioral Development." Developmental Review 7(2): 149–182.

Johnston, Timothy D., and Laura Edwards. (2002). "Genes, Interactions, and the Development of Behavior." Psychological Review 109(1): 26–34.

Kevles, Daniel J. (1985). In the Name of Eugenics. New York: Knopf.

Kuo, Zing-Yang. (1921). "Giving Up Instincts in Psychology." Journal of Philosophy 18(24): 645–664.

Lehrman, Daniel S. (1953). "A Critique of Konrad Lorenz's Theory of Instinctive Behavior." Quarterly Review of Biology 28(4): 337–363.

Lickliter, Robert. (2000). "An Ecological Approach to Behavioral Development: Insights from Comparative Psychology." Ecological Psychology 12(4): 319–334.

Lorenz, Konrad. (1957). "The Nature of Instinct." In Instinctive Behavior: The Development of a Modern Concept, trans. and ed. Claire H. Schiller. New York: International Universities Press. Originally published, 1937.

McGuffin, Peter; Brien Riley; and Robert Plomin. (2001). "Genomics and Behavior: Toward Behavioral Genomics." Science 291(5507): 1232–1233.

Oyama, Susan. (2000). The Ontogeny of Information: Developmental Systems and Evolution, 2nd edition. Durham, NC: Duke University Press.

Parens, Erik. (2004). "Genetic Differences and Human Identities: On Why Talking about Behavioral Genetics Is Important and Difficult." Hastings Center Report, special supplement, 34(1): S1–S36.

Robert, Jason Scott. (2003). "Developmental Systems and Animal Behaviour." Biology and Philosophy 18(3): 477–489.

Rutter, Michael. (2002). "Nature, Nurture, and Development: From Evangelism through Science toward Policy and Practice." Child Development 73(1): 1–21.

Schaffner, Kenneth F. (2001). "Nature and Nurture." Current Opinion in Psychiatry 14(5): 485–490.

Turkheimer, Eric. (1998). "Heritability and Biological Explanation." Psychological Review 105(4): 782–791.

Turkheimer, Eric. (2000). "Three Laws of Behavior Genetics and What They Mean." Current Directions in Psychological Science 9(5): 160–164.

Whitney, Glayde. (1995). "Ideology and Censorship in Behavior Genetics." Mankind Quarterly 35(4): 327–342.