The Fourth and Fifth Laws of Behavioral Genetics


In 2000, behavioral geneticist Eric Turkheimer published an article entitled “Three Laws of Behavior Genetics and What They Mean.”1 Based on what he saw as the “nearly unanimous results” of behavioral genetic studies of families, twins, and adoptees, Turkheimer’s “Three Laws” were:

  1. “All human behavioral traits are heritable.”
  2. “The effect of being raised in the same family is smaller than the effect of genes.”
  3. “A substantial portion of the variation in complex human behavioral traits is not accounted for by the effects of genes or families.”

Critics have argued for decades, however, that most behavioral genetic assumptions, models, concepts, “laws,” and “discoveries” do not hold up under critical examination.2 In particular, these critics have shown that the key assumptions underlying genetic interpretations of studies of twins reared together, as well as those that underlie studies of “reared-apart” twin pairs, range from questionable to false. This also holds true for twin studies of psychiatric disorders.

Turkheimer, in this 2000 “Three Laws” article, wrote that it “is now possible for behavior genetics to move beyond statistical analyses of differences between identical [MZ] and nonidentical [DZ] twins and identify individual genes that are related to behavioral outcomes.” He believed that the completion of the Human Genome Project and subsequent behavioral molecular genetic research would produce one of two main results: (1) that behavioral geneticists would be “vindicated” by the discovery of genes that underlie behavioral differences and psychiatric disorders, or (2) that as the critics expected, genes for behavior are not found, which would support their argument that behavioral genetic twin and adoption studies are based on “shaky theoretical underpinnings.”3

Due to the subsequent failure to discover genes for behavior, it appears that the critics—not the behavioral geneticists—have been vindicated.4 Turkheimer himself was forced to admit in 2011“Everyone assumed that once the human genome was sequenced the ‘genes for’ the phenomena that had been demonstrated to be heritable would be just around the corner, but it hasn’t happened.”5 In 2012, while continuing to support most behavioral genetic methods and theories, Turkheimer wrote that most people “would agree that [the] new era of human genomics has been a disappointment so far.”6 Or as some critics argue, the “genes for behavior” project turned out to be a colossal failure.

Nevertheless, in 2015 behavioral genetic researcher Christopher Chabris and his colleagues endorsed Turkheimer’s original laws, while ignoring his unfulfilled prediction of gene discovery. In support, they cited a 2015 meta-analysis of behavioral twin studies which, as I have shown, merely repeated and combined the acceptance of the false assumptions underlying all classical twin method studies, including the utterly false assumption that reared-together MZ and DZ twin pairs grow up experiencing equal environments.7 Ignoring these problems, Chabris and colleagues proposed an additional “Fourth Law of Behavior Genetics”:

4. A typical human behavioral trait is associated with very many genetic variants, each of which accounts for a very small percentage of the behavioral variability.8

“Behavioral traits” were defined as traits “commonly measured by psychometric measures” (for example, IQ and personality), “a serious psychiatric disease,” or “a social outcome, such as educational attainment.”

Chabris and his colleagues proposed their new “law,” interestingly enough, after decades of failure to find genes (genetic variants) of major effect that contribute to differences in human behavior. Like most researchers attempting, yet failing, to uncover genes for behavior, Chabris and colleagues interpreted these failures as evidence that behavioral variation must be caused by thousands of genes each with a very small effect, and that much larger samples are needed to flush them out. This has been the standard argument in behavioral genetics and psychiatric genetics for the past 15 years or so. In the process, the leaders of these fields reject the obvious conclusion that their results show that genes for most human behavioral differences and psychiatric disorders do not exist.

Although Chabris and colleagues recognized the general failure of candidate-gene studies and genome-wide association studies (GWAS) to identify causal genes, a failure that they attributed to these studies’ “insufficient statistical power” due to small samples, they highlighted a few supposed gene-behavior associations in support of their claim that genes will be discovered once larger samples are collected. They pointed to “a SNP designated ‘rs9320913’… identified in a recent genome-wide association study (GWAS) of educational attainment,” where the original researchers had established a “stringent” significance threshold. Nevertheless, this variant is “estimated to account for only 0.02% of the overall variability in educational attainment.” Clearly, the decision of where to set statistical significance thresholds is subjective, and can affect the researchers’ results and conclusions when conducting potential molecular genetic “fishing expeditions.” In addition, an association (correlation) between genes and behavior doesn’t necessarily mean that the genes cause the behavior.

The many important and interconnected social, political, psychological, and economic factors that contribute to a person’s level of educational attainment are obvious. In order to help society increase levels of educational attainment in the future, behavioral genetic researchers would do better to study the genetic variants of governmental leaders whose behavior leads to raising the cost of public higher education to astronomical levels (such as in the United States), versus the genetic variants of political leaders whose behavior leads to supporting low-cost or free higher education.

Chabris and colleagues also mentioned a widely reported 2014 schizophrenia molecular genetic study, whose authors claimed to have identified 108 schizophrenia-related genetic loci. In addition to the questionable relevance and importance of such studies, these results are likely to join those already residing in the vast “genetics graveyard,” currently inhabited by thousands of false-positive (non-replicated) claims going back to the 1960s.9

The second area of gene discovery claimed by Chabris and colleagues is a relatively new molecular genetic research method called “Genome-wide Complex Trait Analysis,” or GCTA (referred to as “GREML” by Chabris et al.).10 GCTA studies have shown mixed results, and are subject to producing false positive findings based on systematic error and a reliance on questionable assumptions, and on the validity of controversial concepts such as heritability. Another major problem in GCTA research is the failure to adequately account for genetic differences based on variation found among differing populations (population stratification), which introduces a potential environmental confound into GCTA studies.11

Molecular genetic researchers of behavior are in the business of producing genes, which they unquestionably believe exist and play an important role. They are not in the business of producing failure, and so they are compelled to create new explanations (such as “missing heritability”) and “laws” in order to justify receiving the large sums of money they need to continue their work. The scientific-looking “Fourth Law” uses language to transform non-discovery into discovery, thereby justifying careers, beliefs, and funding, while continuing to divert attention from the obvious and important social causes of human behavioral differences. As Joanna Moncrieff put it:

“We will likely never be able to fully account for why some people experience extreme mental states, but we know that poverty, unemployment, insecure attachments, familial disruption, low self-esteem, abuse etc. play a role for many. We would be better concentrating on how to eliminate these from our society if we really want to reduce the impact of mental disorder, rather that pouring more money into the bottomless pit of genetic research.”

Chabris and colleagues believe that their proposed law “explains several consistent patterns in the results of gene-discovery studies.” In reality, these patterns (misinterpreted in favor of genetics) have created only the illusion that such genes exist, enabling money to continue to pour into the “bottomless pit of genetic research.”12

Given the decades of gene finding failures, and the massive problems in behavioral genetic and psychiatric genetic research in general, I propose here a Fifth Law of behavior genetics—actually, the only “law” one needs to know in order to properly understand this field:

5. Behavior genetic Laws 1-4 should be ignored because they are based on many false assumptions, concepts, and models, on negative gene finding attempts, and on decades of unsubstantiated gene discovery claims.

* * * * *

I thank M.C. Jones for contributing ideas
incorporated into some areas of this posting/article.


  1. Turkheimer, E., (2000), Three Laws of Behavior Genetics and What They Mean, Current Directions in Psychological Science, 9, 160-164.
  2. Joseph, J., (2004), The Gene Illusion: Genetic Research in Psychiatry and Psychology under the Microscope, New York: Algora; Joseph, J., (2006), The Missing Gene: Psychiatry, Heredity, and the Fruitless Search for Genes, New York: Algora; Joseph, J., (2015), The Trouble with Twin  Studies: A Reassessment of Twin Research in the Social and Behavioral Sciences, New York: Routledge.
  3. Here is the relevant passage from page 163 of Turkheimer’s 2000 “Three Laws of Behavior Genetics” article: “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. What should we expect from this endeavor? Behavior geneticists anticipate vindication: At long last, statistical variance components will be rooted in the actual causal consequences of actual genes. Critics of behavior genetics expect the opposite, pointing to the repeated failures to replicate associations between genes and behavior as evidence of the shaky theoretical underpinnings of which they have so long complained.”
  4. See Joseph, 2015, Chapters 9 & 10.
  5. Turkheimer, E. (2011), Still Missing, Research in Human Development, 8, 227-241, p. 231.
  6. Turkheimer, E., (2012), “Genome Wide Association Studies of Behavior are Social Science,” in K. Plaisance & T. Reydon (Eds.), Philosophy of Behavioral Biology, (pp. 43–64), Dordrecht, The Netherlands: Springer, p. 44.
  7. Joseph, J., Chaufan, C., Richardson, K., Shultziner, D., Fosse, R., James, O., Latham, J., & Read, J., (2015), The Twin Research Debate in American Criminology, Logos, 14 (Nos. 2-3), Published online 8/9/2015.
  8. Chabris, C. F., Lee, J. J., Cesarini, D., Benjamin, D. J., & Laibson, D. I., (2015), The Fourth Law of Behavior Genetics, Psychological Science, 24, 304-312.
  9. See also Fosse R., Joseph J., & Richardson, K., (2015), A Critical Assessment of the Equal Environment Assumption of the Twin Method for SchizophreniaFrontiers in Psychiatry, 6:62, 1-10. doi:10.3389/fpsyt.2015.00062
  10. In their endnote 3, Chabris et al. (2015) wrote that GREML “is also sometimes called genome-wide complex trait analysis, or GCTA,” although these methods are not identical.
  11. Charney, E., (2013, September 19th), Still Chasing Ghosts: A New Genetic Methodology Will Not Find the “Missing Heritability,” Independent Science News,
  12. Joseph, 2004.


Mad in America hosts blogs by a diverse group of writers. These posts are designed to serve as a public forum for a discussion—broadly speaking—of psychiatry and its treatments. The opinions expressed are the writers’ own.


Mad in America has made some changes to the commenting process. You no longer need to login or create an account on our site to comment. The only information needed is your name, email and comment text. Comments made with an account prior to this change will remain visible on the site.


  1. I always appreciate your work Jay.

    If “scientists” like Turkheimer and Chabris were to admit that their entire career of research is founded on false assumptions which render almost everything they’ve written meaningless and mistaken, they would feel like idiots and lose considerable prestige and future financial opportunity.

    These threats are the main reason why they don’t admit that their brand of behavioral genetics is what it is: garbage.

    Report comment

  2. Hi Jay, Listen, thanks ongoing. Your taste in fight reminds me of Spinoza, by the way, although at my little slight language science level. I took some time placing your nice textbook that looks worth the time to sit around (here And there…hooray, man), that is situating it in the Universe of Discourse–intellectually well. The effort was simple and basic and took the appropriate level of attention to the facts of the details and was of the difficulty that suits the plain reality that most “aeffects” us here in America. And will for ever, or at least For Good for “so long”. As I say unerring in no moanism. Sorry. Just saw who had posted and have to Ave’-Say Save the first read through for later. I have about finished my looks at fundaments and sci and fi arrangements of rhetoric and obsolete SEP talk, etc. That is done. No more viability for them inside, at all. Pleased to let you know that, while still “almost” no thanks due to the fun fanny shape up (so meaty and guerry and stitchy) on the Left Way Out on the Web.

    I like the difference (“the” Relevance) you keep tagged for keeping the talc ahead of the rosin here, for sure. Very much until later, thank you.

    Report comment

  3. “Genes for behaviour” is an absurd idea to begin with. All there is are genes for potential for a behaviour. Sure, if you don’t have legs it’s hard to walk and if you don’t feel emotions it’s hard to be emphatic to your fellow humans but the genes underlying the development of legs or brain structures related to emotion processing do not determine when and how and how much we walk or empathize. We do that in response to the environment. And even when our genes provide us with certain capacities it’s totally possible to lose these abilities, either permanently or temporarily in response to trauma.

    You see this even when you study fruit flies. It’s relatively easy to impair animal’s ability to perform a behaviour but inducing a behaviour is a tricky thing and you almost never get it with a single gene. Usually you need to present an environmental stimulus anyway.

    It’s hard to argue with the fact that people do inherit personality traits but that’s a different thing from saying that they inherit behaviours.

    Report comment

  4. Thanks as always for your exceptionally solid scientific analysis, Jay.

    I have to ask, if we accept that each gene contributes only a tiny amount to this mythical “heritability,” and hundreds or even thousands of genes could be involved in a single personality trait, what the hell would be the point of discovering which set of 350 genes contributed to say, “ADHD”? Especially as not everyone with the postulated “disorder” would have the same subset of those 350 genes involved? And as none of these genes would be modifiable in any case once their presence was discovered?

    Wouldn’t it make a lot more sense to study things that are controllable variables that have a more direct and knowable impact on behavior, such as poverty, education, parenting styles, nutrition, and protection from toxic exposures pre- and post-natally?

    What is wrong with these people? It appears all they are worried about is being “right” even if their “rightness” is of no use to anyone. Oh, well, I guess if you’ve wasted your career on a fruitless line of research, it might be hard to admit it. Though of course, if they are real scientists, proving the null hypothesis isn’t fruitless – it’s just more information – you’ve proven that specific genes don’t create specific behavioral traits, and that’s a good thing to know. Too bad their egos and their mis-education in the sciences forbids them from observing the actual data!

    —– Steve

    Report comment

    • Dear Steve,

      Thanks. You make some really good points, and we must always keep in mind that the people/corporations/governments funding this research have an interest in diverting our attention from the environmental “controllable variables” you mention, in addition to many others.

      Report comment

    • I agree with you, and I have pointed out in the past that genetically oriented researchers usually overlook countless real word examples running counter to their claims. Behavioral geneticists frequently refer to twin studies as “natural experiments,” but they overlook many other natural experiments that point to the importance of environment, such as the one you mention in relation to Australia and the crime rate.

      Report comment

  5. An enjoyable post and comments, especially Steve above.

    It is at least a bit questionable how much we solved our genetic code. The complex process from DNA to RNA to protein to body/behavior is still very much an ongoing research project. There are many things that we can continue to flush out as regards our baseline genes and how that structures who we are. It is still an important area of continued exploration, even if certain parts of the project are ill-conceived.

    That of course is not vindication of most lines of thought of behavioral genetics today, especially as regards complex behaviors. And it does not mean that the far easier and more robust changes to our lives will not come from changes in our social and environmental world. The inability to ask the appropriate questions about social changes is a broader issue than just the gene question. Our failure to ask appropriate social questions falls inline with a more widespread social conservatism, one that is only partially bolstered by such genetic claims (as opposed to our post-60’s, post-Socialist present political atmosphere, for example).

    Report comment