160 VOLUME 9, NUMBER 5, OCTOBER 2000 Three Laws of Behavior Genetics and What They Mean Eric Turkheimer' Department of Psychology, University of Virginia, Charlottesville, Virginia Abstract Behavior genetics has dem- onstrated that genetic variance is an important component of variation for all behavioral out- comes, but variation among families is not. These results have led some critics of behav- ior genetics t conclude that heritability is so ubiquitous as to have few consequences for scientific understanding of de- velopment, while some be- havior genetic partisans have concluded that family environ- ment is not an important cause of developmental outcomes. Both views are incorrect. Geno- type is in fact a more system- atic source of variability than environment, but for reasons that are methodological rather than substantive. Development is fundamentally nonlinear, interactive, and difficult to con- trol experimentally. Twin stud- ies offer a useful methodologi- cal shortcut, but do not show that genes are more fundamen- tal than environments. Keywords genes; environment; develop- ment; behavior genetics The nature-nurture debate is over. The bottom line is that every- thing is heritable, an outcome that has taken all sides of the nature- nurture debate by surprise. Irving Gottesman and I have suggested that the universal influence of the behavior be e enshrined as genes on behavior the first law of behavior genetics (Turkheimer & Gottesman, 1991), and at the risk of naming laws that I can take no credit for discovering, it is worth stating the nearly unani- mous results of behavior genetics in a more formal manner. First Law. All human behavioral traits are heritable. Second Law. The effect of being raised in the same family is smaller than the effect of genes. Third Law. A substantial portion of the variation in complex hu- man behavioral traits is not ac- counted for by the effects of genes or families. It is not my purpose in this brief article to defend these three laws against the many exceptions that might be claimed. The point is that now that the empirical facts are in and no longer a matter of serious controversy, it is time to tum atten- tion to what the three laws mean, to the implications of the genetics of behavior for an understanding of complex human behavior and its development. VARIANCE AND CAUSATION IN BEHAVIORAL DEVELOPMENT If the first two laws are taken lit- erally, they seem to herald a great victory for the nature side of the old debate: Genes matter, families do not. To understand why such views are at best an oversimplifica- tion of a complex reality, it is nec essary to consider the newest wave of opposition that behavior genet Somated. These new crit articulate spokes- ics has ics, whose man is Gilbert Gottlieb (1991, 1992, 1995), claim that the goal oal of devel- opmental psychology is to specify the actual developmental processes that lead to complex outcomes. In lower animals, whose breeding and environment can be brought under the control of the scientist, it is possible to document such devel- opmental processes in exquisite de tail. The critics draw an unfavor able comparison between these detailed animal studies and twin studies of behavior genetics, which produce only statistical conclu- sions about the relative importance of genes and environment in devel- opment. The greatest virtue of the new challenge is that it abandons the Published by Blackwell Publishers Inc CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE 161 implausible environmentalist con- tention that important aspects of behavior will be without genetic influence. Gottlieb (1992) stated, "The present viewpoint holds that genes are an inextricable com- ponent of any developmental sys- tem, and thus genes are involved in all traits" (p. 147). Unlike earlier critics who deplored the reduction- ism they attributed to behavior genetic theories of behavior, the developmental biologists take be- havior genetics to task for not be- ing mechanistic enough. Once vili- fied as the paragon of determinist accounts of human behavior, be- havior genetics is now chastised for offering vague and inconclusive models of development (Gottlieb, 1995; Turkheimer, Goldsmith, & Gottesman, 1995), and judged by the standards of developmental psychobiology in lower animals, it is true enough that behavior ge- netic theories of complex human behavior seem woefully poorly specified. But ultimately the charge is unfair, because there is no equivalent in developmental psy- chobiology to the behavior genetic study of marital status or school performance. great preponder- ance of the exquisite experimental science that goes into animal psy- chobiology is quite simply i impos- sible to conduct in humans. Human developmental social science is difficult equally so for the genetically and environmen- tally inclined because of the (methodologically vexing, human- istically pleasing) confluence of two conditions: (a) Behavior emerges out of complex, nonlinear developmental processes, and (b) ethical considerations prevent us from bringing most human de- velopmental processes under effec- tive experimental control. Figure 1 is a schematic illustration of the problem. Individual genes (Genes 1, 2, and 3) and their environments (which include other genes) inter- act to initiate a complex develop- mental process that determines adult personality. Most characteris- tic of this process is its interactivity: Subsequent environments to which the organism is exposed depend on its earlier states, and each new en- vironment changes the develop- mental trajectory, which affects fu- ture expression of genes, and so forth. Everything is interactive, in the sense that no arrows proceed uninterrupted from cause to effect, any individual gene or environ- mental event produces an effect only by interacting with other genes and environments. For the behavior geneticist, Variation in personality accounted for by genes Gene 1 Gene 2 Genes Gene 3 Environment Reinforcement Parenting Human Development Birth Order Personality "Variation in personality accounted for by environment Fig. 1. Schematic diagram of contrasting roles of genes and environment in development of personality. One-headed arrows link causes to effects; two-headed arrows indicate correlations. Genes and environments are both causal inputs into an interactive developmental system (represented by the network of arrows in the center of the figure), but because people select and shape their own environments (as represented by lighter one-headed arrows from personality to environments), correlations across the developmental system (dotted two-headed arrows) are easier to detect for genes than for environments. Copyright 2000 American Psychological Society 162 however, the quasi-experimental gift of genetically identical and nonidentical twins offers a remark- able, if deceptively simple, method to span this daunting interactive complexity. Thanks to the fact that identical twins are on average ex- actly twice as similar genetically as nonidentical twins, one can use straightforward statistical proce- dures to estimate the proportion of variability in complex outcomes that is associated with causally dis- genes, all the while maintain- ing a state of near-perfect igno rance about the actual causal processes that connect genes to be- havior. This methodological short- cut is not available to rivals of be- havior genetics who seek to measure the effects of families on behavior. How similar was my rearing environment to that of my siblings? And how similar was it to the environment of my adopted sibling, if I have one, o or to the en- vironment of my biological sibling who was raised by someone else? The apparent victory of nature over nurture suggested by the first two laws is thus seen to be more methodological than substantive In a world in which there were oc- casional occurrences of "identical environmental twins," whose ex- periences were exactly the same, moment by moment, and another variety who shared exactly (but randomly) 50% of their experi- ences, environmentalists could re- produce the precision of their ri vals, and like the behavior geneticists could measure with great precision the total contribu- tion of the environment while knowing almost nothing about the developmental processes that un- derlie it. The old-fashioned nature-nur- ture debate was about whether or not genes influence complex be- havioral outcomes, and that ques tion has been decisively answered in the affirmative. The new ques- VOLUME 9, NUMBER 5, OCTOBER 2000 partitioning sources of variance to specifying concrete developmental processes (Turkheimer, 1998), and although critics like Gottlieb are correct that heritability per se has few implications for a scientific un derstanding of development, they have failed to emphasize two cru- cial points. First, heritability does have one certain consequence: It is no longer possible to interpret cor- relations among biologically re- lated family members as prima fa- cie evidence of sociocultural causal mechanisms. If the children of de pressed mothers grow up to be de- pressed themselves, it does not necessarily demonstrate that being raised by a depressed mother is it self depressing. The children might have grown up equally depressed. if they had been adopted and raised by different mothers, under the influence of their biological mother's genes. For every behavior who gene heritabilities as though continues to report moderate they were news, there is an envi- ronmentalist who reports causally ambiguous correlations between genetically related parents and children. Second, the problem the critics have uncovered extends well beyond behavior genetics: It is a rare environmentalist who has never used statistical methods to predict behavioral outcomes from earlier events, in the hope that the specific developmental mecha- nisms can be filled in later. The dis connect between the analysis of variance and the analysis of causes, to use Lewontin's (1974) phrase, is not a proprietary flaw in behavior genetic methodology; in fact, it is the bedrock methodological prob- lem of contemporary social science. NONSHARED ENVIRONMENT AND THE GLOOMY PROSPECT Even after the effects of genes tion is how we can proceed from and the shared effects of families have been accounted for, around 50% of the differences among sib- lings is left unexplained. In recent years, scientists interested in the genetics of behavior have come to call this unexplained portion the "nonshared environment." Al- though according to the second law shared environment accounts for a small proportion of the vari- ability in behavioral outcomes, ac cording to the third law, nonshared environment usually accounts for a substantial portion. So perhaps the appropriate conclusion is not so much that the family environment does not matter for development, but rather that the part of the fam- ily environment that is shared by siblings does not matter. What does matter is the individual envi ronments of children, their peers, and the aspects of their parenting that they do not share. Plomin and Daniels (1987) reviewed evidence of the predominance of nonshared environmental variance and posed a seminal question: Why are chil dren in the same family so differ siblings ent? They proposed are different because nonshared environmental events are more potent causes of developmental outcomes than the shared environ mental variables, like socioeco- nomic status, that have formed the traditional basis of sociocultural developmental psychology. Plomin and Daniels's explana- tion involves a subtle conceptual shift, best described in terms of a distinction between the objective and effective environment (Gold- smith, 1993; Turkheimer & Wal- dron, 2000). What qualifies an en- vironmental event as nonshared? There are two possibilities. The first is objective: An event is non- shared if it is experienced by only one sibling in a family, regardless of the consequences it produces. The other possibility is effective: An environmental event is non- shared if it makes siblings different Published by Blackwell Publishers Inc. CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE rather than similar, regardless of whether it was experienced by one or both of them. Plomin and Daniels's proposal, then, is that the nonshared environment as an ef fectively defined variance compo- nent can be explained by objec- tively nonshared environmental events. The question, "Why are children in e same family so dif- ferent?" is answered, "Because measurable differences in their en- vironments make them that way." This proposal has been enor- mously influential, spawning an entire area of empirical inquiry into the consequences of measured en- vironmental differences among siblings. Ironically, that same lit- erature has quite decisively dem- onstrated that the conjecture is false. A review of 43 studies that measured differences in the envi- ronments siblings and related them to differences in the siblings' developmental outcomes (Turk- heimer & Waldron, 2000) has shown that although upwards of 50% of the variance in behavioral accounted for by the effectively defined variance com- ponent called nonshared environ- ment, the median percentage ac- counted for by objectively defined nonshared events is less than 2% outcomes What could be going on? Plomin and Daniels (1987) al- most identified the answer to this question, but dismissed it as too pessimistic: One gloomy be unsystematic, spect is that the salient environment might idiosyncratic, or serendipitous events such as accidents, illnesses, or other traumas. Such events, however, are likely to dead end for research. More interesting heuristi- cally are possible systematic sources of differences between families. (p. 8) The gloomy prospect is true. Non- shared environmental variability predominates not because of the systematic effects of environmental events that are not shared among siblings, but rather because of the unsystematic effects of all environ- mental events, compounded by the equally unsystematic processes that expose us to environmental events in the first place (Turk- heimer & Gottesman, 1996). A model of nonshared variabil- ity based on the gloomy prospect is radically different from the Plomin model based on systematic conse- quences of environmental differ- ences among siblings. Most impor tant, the two models suggest very different prospects for a genetically informed developmental psychol- ogy. Again and again Plomin and his colleagues have emphasized. that the importance of nonshared environment implies that it is time to abandon shared environmental variables as possible explanations of developmental outcomes. And although modern environmental- ists might not miss coarse mea- sures like socioeconomic status, it is quite another thing to give up on the causal efficaciousness of nor- mal families, as Scarr (1992), Rowe (1994), and Harris (1998) have urged. If, however, nonshared en vironmental variability in outcome is the result of the unsystematic consequences of both shared and nonshared environmental events, the field faces formidable method- ological problems-Plomin and Daniels's gloomy prospect-but need not conclude that aspects of families children share with sib- lings are of no causal importance. CONCLUSION: ANTICIPATING THE GENOME PROJECT It is now possible for behavior genetics to move beyond statistical analyses of differences between identical and nonidentical twins and identify individual genes that are related to behavioral outcomes. 163 What should we e expect from this endeavor? Behavior geneticists an- ticipate vindication: At long last, statistical variance components. will be rooted in the actual causal consequences of actual genes. Crit- ics of behavior genetics expect the opposite, pointing to the repeated failures to replicate associations be- tween genes and behavior as evi- dence of the shaky theoretical un which they have so derpins long There is an interesting parallel between the search for individual genes that influence behavior and the failed attempt to specify the nonshared environment in terms of measured environmental variables. In each case, investigators began with statistically reliable but caus- ally vague sources of variance, and set out to discover the actual causal processes that produced them. The quest for the nonshared environ- ment, as we have seen, got stuck in the gloomy prospect. Although in- dividual environmental events in- fluence outcomes in the most gen eral sense, they do not do so in a systematic way. One can detect their effects only by accumulating them statistically, using twins or adoptees. If the underlying causal struc- ture of human development is highly complex, as illustrated in Figure 1, the relatively simple sta- tistical procedures employed by developmental psychologists, ge- neticists, and environmentalists alike are being badly misapplied. But misapplied statistical proce dures still produce what appear to be results. Small relations would still be found between predictors and outcomes, but the underlying complex causal processes would cause the apparent results to be small, and to change unpredictably from one experiment to the next. So individual investigators would ob- tain "results," which would then fail to replicate and accumulate into a coherent theory because the Copyright 2000 American Psychological Society 164 simple statistical model did not fit the complex developmental pro- cess to which it was being applied. Much social science conducted in the shadow of the gloomy prospect has exactly this flavor (e.g., Meehl, 1978). The gloomy prospect looms larger for the genome project than is generally acknowledged. The question is not whether there are correlations to be found between individual genes and complex be- havior of course there are-but e domains instead whether there of genetic causation in which the gloomy prospect does not prevail, VOLUME 9, NUMBER 5, OCTOBER 2000 tics over chaotic psychological re- ality, it is a devil's choice: In the long run, the gloomy prospect al- ways wins, and no one would want to live in a world where it did not. Psychology is at least one good paradigm shift away from an em- onse pirical answer to the gloomy pros- pect, but the philosophical re- s becoming clear: The additive effect of genes may con- human development, but what is predictable about human develop- ment is also what is least interest- stitute what is predictable about ing about it. The gloomy prospect isn't. allowing the little bits f correla tional evidence to cohere into rep Recommended Reading licable and models of dulative genetic Gottlieb, G. (1992). (See References) development. My own Lewontin, R.C. (1974). (See Refer- prediction is that such domains will prove rare indeed, and that the Mechl, PE (1978). (See References) likelihood of discovering them will Plomin, R., & Daniels, D. (1987). (See be inversely related to the com- plexity of the behavior under study. Finally, it must be remembered that the gloomy prospect is gloomy only from the point of view of the working social scientist. Although frustrated developmental psy- chologists may be tempted to favor methodologically tractable heuris- ence References) Note 1. Address correspondence to Eric Turkheimer, Department of Psychol ogy, 102 Gilmer Hall, P.O. Box 400400, University of Virginia, Charlottesville, VA 22904-4400; e-mail: turkheimer virginia.edu. References Goldsmith, H. (1993) Nature-nurture isues in the behavioral genetic context: Overcoming barri ers to communication. In R. Plomin & G. Mc Clean (Eds), Nature, marfare and paychology Fenchelington, DC: American Gob, G. (1991). Experiential canalization of be havioral development Theory. Developmental Psychology, 27, 4-13 Go, G. (1992). Individual development and evo lution. New York Odord University Press Gob, G. (1995). Some conceptual deficiencies in "developmental behavior genetics Hama Dept, 38, 131-141. Harris, JR. (1998). The nurture amption: Why Hum the way they do. New York Genetic, 26, 400-411. Journal of Lewontin, R.C. (1974). The analysis of variance and the analysis of cases. American Mechl, P.E. (1978). Thecoetical risks and tabular asterisk Sir Karl, Sir Ronald, and the slow progress of soft psychology Journal of Consult Plomin, R., & Daniels, D. (1987). Why are children Rowe, DC. 0994). The lo Gene exprimer, and behavior. New York: Guil Scar, S. (1992). Developmental theories for the 1990 Development and individual eno Cald Delopment, 63, 1-19. differ Turkheimer, E. (1998) Heritability and biological explanation Prychological Review, 105, 782-291 Turker, E, Goldsmith, H.H, & Goeman 38, 142-153ary. Human Development, Turkheimer, E., & Gottesman, LL (1991). H-0 a null hypothesis anymore? Behavioral and Brain Sci, 14, 410-411. Turkheimer, E., & Gottesman, LL. (1996). Simulat ing the dynamics of games and environment in development. Development and Psychopathol- Turkheimer, E., & Waldron, MC. (2000). Non shared environment: A theoretical method 8667-677 Published by Blackwell Publishers Inc This document is a scanned copy of a printed document. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material.