Books

Does DNA Make Us Who We Are?

(Pixabay)
A new book from a prominent scientist makes the case.

Ideally science is a progressive enterprise, drifting and twisting ever closer to objective truth. And yet science is also a human enterprise, shaped by its practitioners’ aspirations. In reality it navigates somewhere between its Platonic ideal and the darkest denunciations of the skeptics, who claim it to be just another ideology.

Psychology illustrates this perfectly. Despite their scientific pretensions, the dogmas of the 20th century — Jungianism, Freudianism, and Behaviorism — have given way to new scholarship more firmly grounded in rigorous method and fed by rich data. We now see that these movements were more reflective of intuitions and prejudices, not evidence and insight.

But the discipline of behavior genetics — the study of human differences in psychological traits, including their genetic “heritability” — has gained strength through an infusion of methods and data from new DNA technologies. The field once relied on statistics to illustrate how similar parents were to their children, and how similar siblings were to each other. Today it has added molecular biology to its toolkit.

Robert Plomin’s Blueprint: How DNA Makes Us Who We Are is a triumphal tour of a discipline that’s having its moment in the sun. While some fields of psychology are going through the “replication crisis,” with old results being proven false, behavior genetics is marching from strength to strength. Plomin looks back across his long career, with its successes and failures, and gazes hopefully into the future to come. A doyen of the field, he is in a position to make such assessments. But as an active participant he also has a definite viewpoint.

Blueprint is not a dispassionate textbook, and readers should be aware this is not a “view from nowhere.” As a scholar of some note, Plomin has staked his own positions in contentious debates, and he has strong opinions about which other researchers should or shouldn’t be cited – and here, writing for a general audience, Plomin shapes the narrative to his own advantage. The treatment of behavior genetics is Blueprint is defensible, but not necessarily indisputable.

The scientific meat of the book is neatly divided into two halves: pre-DNA and post-DNA. Behavior genetics in its conception goes back to the 19th century, specifically to Francis Galton — Charles Darwin’s cousin and, troublingly, the founder also of the field of eugenics. Galton and his intellectual successors in the 20th century were fascinated by patterns of inheritance of intellectual talent and mental debility. Methods of studying these patterns were developed long before Francis Crick and James Watson discovered that DNA was the molecule that encoded genetic information.

Plomin outlines the basics of the method that dominated much of the 20th century in behavior genetics: the study of twins. By looking at how closely identical and fraternal twins resembled each other, researchers could adduce the amount of the variation of a trait that was due to variation in genes. For example, the correlation between identical twins in height is quite high, if not a perfect 1.0. But between fraternal twins it is closer to 0.50, corresponding to the fact that fraternal twins share only about half their genes. From this pattern geneticists can infer that height is a very heritable trait, with most of the variation in the population due to variation in genes.

For intelligence, the correlation between identical twins is closer to 0.75, and that between other siblings somewhat less than 0.50. This means intelligence is quite heritable, but not nearly as heritable as height.

If you have read any behavior genetics in the past, most of the material Plomin presents in Blueprint is not new. Rather, it is the distillation of research that has been replicated many times over the course of half a century — and now often confirmed through genomics. If you didn’t believe before, the message of Blueprint is that you better believe now.

The overarching point one can take from Blueprint is that genetic variation is the one factor we know has a big influence over our psychological characteristics. This does not negate the importance of environment; indeed, genes explain less than half of the variation in most psychological characteristics. But we do not have a good grasp of what parts of the environment make a difference.

Blueprint retells the story that Judith Rich Harris unfolded in The Nurture Assumption 20 years ago: The impact of parental intervention on the outcomes of their children is minimal. The “shared environment” — the home and hearth — typically is shown to have very limited effect in studies of characteristics such as personality, intelligence, and social pathology. (If the shared environment had as big of an effect on a trait that genes have on height, for example, both identical and fraternal twins would be incredibly similar to each other. That is essentially never what we see.) Often the largest factor is “non-shared environment,” which is the dark matter of behavior genetics: the parts of the environment that make people different even when they’re raised in the same home. Influential, but unseen.

The reason that Plomin focuses on genes is twofold: They seem to explain more of the variation in behavior than shared environment, and as the science progresses we’re coming to understand genes much better than we understand the non-shared environment. Researchers using cutting-edge technology have transformed the operations of behavior-genetics laboratories in the past few years. Whereas 20 years ago researchers were working with sample sizes of a few hundred and inspecting correlations across these individuals, today genomics allows for an analysis of hundreds of thousands of genetic markers in hundreds of thousands of individuals. The largest study reported in Blueprint has more than a million individuals. Rather than measuring the impact of “genes” in the aggregate, researchers are identifying specific genes that affect these traits.

Why does this matter? First, researchers are confirming the heritability of some characteristics in a very concrete manner: pinpointing the numerous genetic loci at work. Though there is almost never a single “gene for” a psychological trait, researchers now know many of the genes, in the plural. The speed at which results are coming out is mind-boggling. In Blueprint Plomin reports that genomic prediction methods can now account for 17 percent of the heritability of height, a benchmark complex trait that is similar in some ways to psychological characteristics. He likely wrote that passage last year. I attended a scientific conference in October of 2018 where a researcher reported accounting for 70 percent of the heritability of height.

Plomin clearly sees this as vindication, and not without reason. The sibling-based work he pioneered in the 20th century came under fire around the turn of the century, when behavior geneticists’ efforts to isolate the genes responsible for behavioral differences generally failed. To state it simply, we did not have the technology. Today we do. Genomics has transformed behavior genetics into a biological science.

Eventually, Blueprint moves from basic science to its implications for social policy. The newest work on the genetic predictors of educational attainment, for example, now explains a greater proportion of the outcome than parental income does (though not more than the educational level of the parents).

It is one thing to be excited about explaining variation within the population, but it is another thing altogether to make individual predictions. Regarding the latter, one of the last chapters is titled “The DNA Fortune Teller.”

The title exposes the issues at work here, because the prediction of complex psychological characteristics is a matter of propensity and probability, not inevitability. Though Plomin’s enthusiasm for genetic methods may indicate to you that he is a determinist, the truth is that he is grasping at the only probabilities that he can provide: those of genes, because, as discussed above, the home environment is marginal in affecting long-term outcomes and the non-shared environment is mysterious. Plomin’s own analysis of his “polygenic risk scores,” his predicted characteristics based on his genetic profile, illustrate the primitive state of this enterprise. A man in the fullness of his years, he already knows who he is. Genes add little beyond entertainment value.

When it comes to individual predictions for psychological characteristics, DNA holds more promise and possibility in the future than actual gains for the present. Plomin admits this implicitly in Blueprint when he suggests that genetic prediction in the near future will be focused on medical conditions. Though much is made of intelligence in Blueprint, it is important to note that susceptibility to schizophrenia is by all measures more heritable than intelligence, with 80 percent of the variation being attributable to genes.

A limitation of Plomin’s research program, and of behavior genetics more broadly — from the sibling studies of decades past to the DNA analysis of today — is that it suffers from a focus on Western populations. This is simply a limitation of means and access. When Plomin regales you with heritability estimate in Blueprint, always remind yourself that this is the “best of all worlds,” where everyone is relatively healthy and has access to a basic education, not to mention food. Put differently, rich Western societies put a lot of effort into providing everyone the same environment to some degree, which reduces the effect that environmental variation has — and allows individuals’ genetic propensities to assert themselves. As is made clear in Blueprint, nature makes no sense without the context of nurture.

There is also one aspect of the science in Blueprint that may warrant some critique in light of new findings. A consistent result from behavior genetics, as Plomin accurately reports, is that “parents don’t matter” much in the ultimate outcomes of many characteristics. This was the finding reported to the world’s surprise in The Nurture Assumption, and it has been consistent down to the publication of Blueprint. But just as Plomin was no doubt completing his revisions to Blueprint, a new line of inquiry suggested that parents may actually matter more than researchers had concluded. New science shows that parents’ genes affect children’s characteristics even when those genes are not passed down to the child, implying that parents’ behavior shapes the environment of the child. Science surprises constantly, and does not always adhere to our blueprint.

Of course, genetic research also raises serious ethical issues, as it has since the days of Galton. The specter of eugenics hangs over the collection of disciplines that use genetic methods to understand humans.

Plomin tries to have it both ways: He admits that science is simply a tool and not does not dictate any course of action; but he is clearly enthusiastic and optimistic about the possibility of predicting individuals’ traits based on their DNA. Earlier I mentioned that behavior genetics matured in a period when eugenics was mainstream. To be frank, the same is true of human genetics, a field whose home departments are in the biological, as opposed to psychological, sciences. The specter of eugenics does hang over the collection of disciplines that use genetic methods to understand humans.

And that specter has additional salience in a technological society. In the United States of America there are now parents who give thanks to God that they were able to have biological children due to in vitro fertilization. This is not a controversial procedure anymore, with the term “test-tube babies” seeming quaint and retrograde. Additionally, most insurance plans now cover Non-Invasive Prenatal Testing (NIPT) for women older than 35. When these tests come back positive for conditions such as Down syndrome, abortion rates in developed countries are quite high.

In Blueprint there is discussion about estimating a child’s final height or intelligence. These are interesting questions for the future, but worries about eugenics coming out of this research are both premature and too late. Parents may make decisions about their future offspring based on probabilities about behavior, but the reality is that this is probably less of a concern than the inevitabilities, which is the domain of medical rather than behavioral genetics. Meanwhile, the reality is that biological technology is already affecting our lives a great deal, and that this is happening with little discussion or worry.

The social consequences of the genomic revolution to medicine and psychology are hard to predict. In developed Western societies the revolution is being driven by bottom-up consumer demand. Over 20 million Americans now have their own DNA results from genomic tests. Instead of top-down government fiat, the choices of hundreds of millions of individuals with their own myriad values and concerns control the field’s direction.

The fact is that there is no blueprint for how the science is realistically going to be used. But that would be a very different book.

Razib Khan — Razib Khan is a geneticist. His blog is Gene Expression (gnxp.com).

Most Popular

U.S.

A Home Run by Trump

In 2007, the Justice Department was in disarray. Though it was largely exaggerated, a controversy over the firing of some United States attorneys, the intrusion of politics into Justice Department hiring decisions, and White House contacts with Main Justice forced the resignation of an overmatched attorney ... Read More
Energy & Environment

Identifying the Problem

EDITOR’S NOTE: The following is Jonah Goldberg’s weekly “news”letter, the G-File. Subscribe here to get the G-File delivered to your inbox on Fridays. Dear Reader (including all passengers on Spaceship Earth), So, as often happens, a weasel crawls up your tailpipe (I mean of your car, sicko). It ... Read More