Twins: And What They Tell Us About Who We Are (13 page)

Read Twins: And What They Tell Us About Who We Are Online

Authors: Lawrence Wright

Tags: #Family & Relationships, #Siblings, #Science, #Life Sciences, #Genetics & Genomics, #test

BOOK: Twins: And What They Tell Us About Who We Are
5.66Mb size Format: txt, pdf, ePub
 
Page 74
the reared-apart twins who have ever been studied were middle-aged adults, whereas the IQ correlations for twins reared together were based largely on data that came from children and adolescents. If, indeed, identical twins grow to be more alike over time, then logically this "age effect" would explain the fact that the adult intelligence of reared-apart twins is more similar than that of reared-together twin children.
Devlin and his colleagues, Michael Daniels and Kathryn Roeder, thought that there must be another explanation. Age does not seem to affect personality, so why would it affect intelligence? Devlin also observed that twins who have been reared apart are said to have no common environment, but obviously that's not truethey shared the environment of the womb. Devlin claims that most studies of intelligence have presumed that the prenatal environment has a negligible influence on the development of intelligence, and yet a child's height or birth weight, for instance, can be significantly affected by the health of the mother during her pregnancy. Moreover, prenatal exposure to substances such as alcohol, lead, drugs, and cigarettes has been shown to lower IQ, just as good nutrition has been shown to raise it.
The Pittsburgh team postulated that this "maternal effect" accounts for the difference between the heritability correlations of intelligence between reared-apart and reared-together twins. They built a statistical model to analyze 212 previous studies of intelligence correlations that had been derived through kinship and adoption studies. They found that the maternal effect accounted for twenty percent of the variation of intelligence among twins and five percent among ordinary siblingsa rather extraordinary figure, which if true would lower the heritability estimates of intelligence by quite a margin. Indeed, the Pittsburgh team calculated that the
 
Page 75
broad heritability of intelligence was about 0.48giving a slight majority to the influence of the environment, including the prenatal environment. Such a figure would be far too low to support Jensen's argument, reiterated and expanded by Herrnstein and Murray in
The Bell Curve
, that the intermarriage of highly educated, intelligent people will lead inevitably to separate castes based on IQ scores. It is also considerably lower than the 0.66 figure that Bouchard cites as a consensus for all twin studies. "Overall, the study results have two implications," says Devlin. "A new model may be required regarding the influence of genes and environment on cognitive function, and interventions aimed at improving the prenatal environment could lead to a significant increase in the population's IQ."
Much of Devlin's information is arrived at by comparing fraternal twins with ordinary siblings. Despite the fact that on average fraternal twins are no more genetically alike than ordinary brothers and sisters, each having about half their genes in common, fraternal twins show much more similarity in intelligence. That increased similarity, says Devlin, is the maternal effectthe experience of sharing the same womb serves to make twins more alike than ordinary siblings. However, the experience of sharing the womb also tends to make twins different from each other. They compete for space and nutrition and even blood, which can create striking dissimilaritiesso much so that many twin researchers believe that their data on heritability actually underestimates the heritability for the population as a whole.
Bouchard disputes Devlin's analysis on several grounds. For one, he points to studies demonstrating that fraternal twins actually grow to be less alike in their intelligence as they agemore like ordinary brothers and sistersso that whatever effect Devlin is demonstrating disappears
 
Page 76
as the twins become adults. "If they were just drawing conclusions about the sample they worked with, which is heavily children, I'd say, 'Well, who knows?' But I know there's information about older adults, I know that these common family environmental influences disappear and that the postulate of maternal effects runs counter to all the work done by people who've studied prenatal effects. And so I just don't believe that this is the best model. They may be right in the long run. I just don't think so." The diminishing effect of prenatal experience is a quandary that the environmentalists acknowledge but have been unable to solve.
A more sinister charge is that the separated twins are making up stories in order to get into the press. "I don't blame the twins," says Kamin. "There's enormous implied pressure on them to exaggerate the degree of their separation. Nobody would be interested in them, they would not appear in the newspapers, they would not appear on TV shows and so on, if they said, 'Yeah, we saw quite a bit of each other and we went to the same schools.' If they convince other people and themselves that they saw very little of one another, then they're going to be valuable scientific resources and people will beat a path to their door. I think it is beyond cavil that these twins tend to gild the lily." Kamin points out that there is an economic interest as well: several of the separated twins, including the Jims and Jack and Oskar, have signed movie or book deals. Kamin claimed to have spoken to Jim Springer's mother, who he says admitted to him that the twins had met repeatedly in Florida when they were young but kept it a secret from Springer's father. "Now that appears, of course, in all the apocryphal literature about this as some mysterious things in the genes led the two of them to go each year
 
Page 77
to Saint Petersburg. I know she spoke to me about the fact that she had to keep these meetings secret from her husband." Kamin said he had notes of the conversation, but when asked to produce them, he found that there was no reference to the twins' meeting or the mothers' having known each other. "I don't know whether I was told or I deduced it," he says now. Jim Springer says such meetings never occurred.
"God could appear to me in a dream and tell me the outcome of a perfect twin study, and my question to God would be, 'Okay, now that I know that the heritability is 0.469327, what do I do with it? Tell me what that tells me,'" says Lewontin, one of Kamin's coauthors. He criticizes the statistical practice of correlation, which implies causal relationships between genes and behavior that may have nothing in common. "If I look at people who knit and people who don't knit, I will make the following discovery: that almost everybody who knits has two X chromosomes and people who don't knit have one X and one Y. Now, how can it be that having two X chromosomes makes you knit? Well, we know the answer to that. Having two X chromosomes makes you into a woman. In our society it is culturalpurely culturalthat women knit. If we had looked at exactly the same problem in eighteenth-century England, we would have found that all knitters have one X chromosome and one Y, because it was men who did knittingit was an economically important occupation. We wouldn't want to say there are genes for knitting on the X chromosome, but we understand that there are genes that make you into a female, which in the present historical circumstance has as a consequence that it's okay for you to knit."
It is certainly true that statistics can be used to associate unrelated matters. Daniel Patrick Moynihan, a
 
Page 78
United States senator who is also a distinguished sociologist, rather famously made the case on the subject of IQ, when Southern intellectuals were asserting white-race superiority. Moynihan found a high correlation between intelligence and proximity to the Canadian border. The further south one traveled, he demonstrated, the lower the IQ. As is the case with race, however, the variation within a region is so great that geography is virtually worthless as a predictor of individual IQ.
The statistical detective work that behavioral geneticists have done has already prompted an extraordinary reevaluation of the architecture of the human personality. And yet Lewontin has claimed that "nothing we can know about the genetics of human behavior can have any implications for human society," a statement that is perhaps best described as wishful thinking on his part. Lewontin and his coauthors Kamin and Rose are socialists who believe that the rise of biological determinism has led to the political ascendancy of the New Right. They are no doubt correct. Society will organize itself around its beliefs of how human nature operates in the world. "The consequences of determinism reach out beyond theory," Steven Rose wrote, perhaps despairingly, in
Nature
in 1995. "If the homeless or depressed are so because of a flaw in their biology, their condition cannot be the fault of society, albeit a humane society will attempt, pharmacologically or otherwise, to alleviate their distress. This 'victim blaming' generates in its turn a sort of fatalism among those it stigmatizes.''
As the views of the environmentalists lose favor, the politics that have been built upon their assumptions crumble. There has simply been nothing on the environmental side to counter the power of twin and adoption studies. "When I point to the weaknesses of behavioral genetics studies, I'm not saying there are a
 
Page 79
whole set of marvelous environmental studies," Kamin concedes. "We don't know what in the environment affects IQ. There's just not a good study on the family environment."
So far, the molecular evidence that would buttress the statistical studies of behavioral genetics has been slow in coming. As scientists map the human genome, they have been able to identify a few disorders that are caused by a single gene, such as cystic fibrosis, phenyl-ketonuria, and Huntington's disease. In several cases, scientists have trumpeted genetic markers for such things as manic depression or schizophrenia and have then quietly withdrawn their findings.
The brain depends upon enzymes for the neurochemical process we call thinking, and those enzymes are created by genes. From a biochemical perspective, there is an obvious genetic contribution to intelligence, but no single gene that makes one person smarter than another. Instead, more than half of the body's 100,000 genes donate in some way to the general fund of intelligence. Robert Plomin, a behavioral geneticist of the Institute of Psychiatry in London, who at that time was at Pennsylvania State University, examined two groups of children, one with high IQs and another with low IQs, to see if there were particular markers that were in or near genes thought to be associated with cognition. He found five markers that had significant associations with differences in intelligence, but four of them failed to be replicated in an independent sample.
Perhaps the most successful example of twin studies leading to molecular discovery is the example of Tourette's syndrome, a bizarre neurological disorder characterized by chronic tics, sometimes violent twitching, intrusive thoughts, and the uttering of repetitious, meaningless phrases. Tourette's runs in families and has
 
Page 80
all the characteristics of a genetic disease, and yet the search of the human genome has so far yielded no obvious genetic markers for the disorder. Daniel Weinberger, chief of the Clinical Disorders Branch at the National Institute of Mental Health in Baltimore, studied five sets of identical twins. In each set, one twin meets the so-called "300-yard diagnosis," which means that his symptoms are so apparent that an observer can diagnose them at that distance, whereas the other twin has less severe symptoms. When the researchers examined the brains of the twins, they found that the more affected twin had extremely sensitive receptors for the brain chemical dopamine. Dopamine affects the section of the brain called the caudate nucleus, the region where motor actions are planned. Because each set of twins has identical genes, the researchers reason that the cause of the difference in their dopamine receptors may have to do with birth traumait was usually true that the twin with the lower birth weight grew up to be the more severely affectedor else the stress of different life events. In either case, it appears that there is a genetic basis for the disease, but it requires an environmental trigger to awaken its catastrophic potential.
Alcoholism also runs in families, and identical twins are far more alike in respect to their drinking behavior than fraternal twins. Moreover, identical twins who have been raised apart are about as alike in their drinking as identical twins raised together. This suggests, as strongly as any other twin studies, that alcoholism is an inherited disorder. In 1990, researchers at the University of Texas Health Science Center at San Antonio claimed to have linked the D-2 dopamine receptor gene to alcoholismthe same location where scientists found the discordance for Tourette's syndromebut other scientists failed to replicate the finding. Even if
 
Page 81
there is an alcoholic gene, however, it is clear that environment influences much of drinking behavior. Children of alcoholics who have been adopted into families where drinking is not a problem rarely become alcoholics themselves; such cases seem to depend upon the stability of the adopting family. Twin studies correlating drinking behavior with religious affiliation found that alcoholism was five times higher in people with no religious affiliation than in the most fundamentalist group. Presumably Muslims and Amish people would carry at least some genetic vulnerability for alcoholism, if there is such a thing, but alcoholism is rarely a problem in cultures where people are simply forbidden to drink.
Similarly, divorce has a strong heritable component, at least in the United States. However, divorce is not tolerated in Amish societies; and in traditional Muslim societies men can have multiple wives and may divorce simply by renouncing the marriage. In such environments, genes that might influence divorce would have wildly different opportunities to express themselves. Many studies have shown that children of divorced parents have more emotional problems and perform more poorly in school, and eventually get more divorces than children of parents who stay togetherbut is that because of the divorce? Or do the children also inherit genes for personalities and behavior that may later lead to their own wrecked relationships?
To take another controversial example, homosexuality may be partly genetically driven. There is ample support for this from neuroscientists, who have found female brain structures in male transsexuals, and from molecular geneticists, who think they have isolated a "gay gene" on the X chromosome. Michael Bailey, a psychologist at Northwestern University, and Richard

Other books

When in Rome... by Gemma Townley
Once Upon a Lie by Maggie Barbieri
Wildwood by Janine Ashbless
The Trophy Wife by Ashley, JaQuavis
A Good Guy With A Gun by Steven Friedman