Fascination with identical twins goes back a long way: think of the founders of Rome, Romulus and Remus, and the astrological symbol Gemini. Our modern understanding of genetics has given us a scientific perspective on identical twins, and yet it too is accompanied by a mythology.
By virtue of being genetically identical, identical twins would seem to provide an ideal source of evidence concerning the heritability of traits. Since identical twins share physical characteristics, is it not conceivable that they also share traits such as intelligence and personality? There is, however, a fly in the ointment: the influence of the environment in which twins are raised. While eye colour has a purely genetic basis, couldn’t intelligence depend on how twins are jointly brought up? There are two ways around this obstacle.
Twins: Take 1
The first depends on a rather unlikely occurrence: identical twins separated at birth. Which brings us, my dear Watson (and Crick) to the curious case of Sir Cyril Burt (1883–1971), an English educational psychologist. In “The Mismeasure of Man”, Stephen Jay Gould writes that Burt:
… published several papers that butressed the [claim that IQ is inherited] by citing very high correlation between IQ scores of identical twins raised apart. Burt’s study stood out among all others because he had found fifty-three pairs, more than twice the total of any previous attempt.
But perhaps this was too good to be true:
Princeton psychologist Leon Kamin first noted that, while Burt had increased his sample of twins from fewer than twenty to more than fifty in a series of publications, the average correlation between pairs for IQ remained unchanged to the third decimal place—a statistical situation so unlikely that it matches our vernacular definition of impossible.
Gould goes on to review further evidence that Burt faked many of his results. But by the time this came to light, the damage was already done: Burt’s studies had influenced British educational policy for decades.
Twins: Take 2
A second way around the impact of environment is to compare identical twins with same-sex fraternal twins. If a trait is inherited, we would expect identical twins (known as monozygotes, MZ) to be more similar than fraternal twins (dizygotes, DZ), irrespective of environmental influences. A particularly disturbing example of this argument is in a 2005 paper titled “Evidence for substantial genetic risk for psychopathy in 7-year-olds” (Viding et al. J. Child Psychol. Psychiatry 46:6, pp 592–597). Measures of psychopathy were reported by teachers for each member of nearly four thousand twin pairs. A child with a high score (the “proband”) was then compared with his or her co-twin. The figure below is a hypothetical example, but it qualitatively represents the study findings:
As depicted in the figure, identical co-twins (MZ) were closer to the proband than fraternal twins (DZ). In fact, the authors studied two traits: callous-unemotional (CU) and antisocial behaviour (AB). The reported “remarkably high heritability for CU, and for AB children with CU”.
Megan McArdle cited this study (via a summary article) in a recent blog post:
Some years ago, I remember reading Jonathan Kellerman’s Savage Spawn, a book on sociopathic children, and how nearly impossible it is to treat them. […] It’s truly heartbreaking: a child who doesn’t seem capable of loving its parents, or anyone else. It seems to be mostly genetic, and nearly completely immune to any current treaments.
While I can’t comment on the claim that it’s “nearly impossible to treat” (though I wonder what the substantive basis for such a claim would be), I do question the claim that it’s “mostly genetic”. Do identical twins really grow up in the same environment as same-sex fraternal twins? I doubt it. As Lea Winerman wrote in A Second Look at Twin Studies in the April 2004 issue of the American Psychological Association Monitor: “… some research suggests that parents, teachers, peers and others may treat identical twins more similarly than fraternal twins.” Winerman provides a balanced review of the continuing controversy about twin studies, raising a number of other problematic assumptions.
Time to move on …
Back in April I expressed my skepticism about genetic determinism in the context of sex differences. I don’t doubt that genetic factors can play an important role in a variety of traits. But the more complex the trait the trickier it is to sort out the relative contributions of genetics and environment. And the more likely it is that our own prejudices will swamp the evidence. We’ve been down this road before.