As an adolescent, Robert wondered if he had been adopted. He recounts in his book that he grew up in a one-bedroom flat in inner-city Chicago without books (Plomin 2018). His father worked on the assembly line at a car factory, later becoming a layout engineer. No one in his family went to university, including his parents, his sister and a dozen cousins who lived nearby. However, his parents encouraged him to go to the public library and became an avid reader at an early age, bringing bags of books home from his local public library. He always did well at school, worked hard, was conscientious and he persevered. At that time, he often wondered where his interest in books and school came from, given that his family showed little attachment to these things. He found the explanation when he began studying genetics. As a teenager, he didn’t realize that although the first law of genetics is that like begets like, the second law is that like does not beget like. Genetics makes first-degree relatives 50% different as well as 50% similar.
His interest in behavioral genetics developed at graduate school in Texas, then a leading center for the subject. After finishing his Ph.D., he got his dream job at the University of Colorado at Boulder with a joint appointment in the Department of Psychology and the Institute for Behavioral Genetics, the only institute of its kind in the world. He decided to create a long-term longitudinal adoption study of psychological development. From 1986 until 1994 he worked at Pennsylvania State University, studying elderly twins reared separately and twins reared together to study aging, and he currently works at the Institute of Psychiatry of King’s College London. Plomin is presently conducting the Twins Early Development Study of all twins born in England from 1994 to 1996, focusing on developmental delays in early childhood, their association with behavioral problems and educational attainment.
Among Plomin’s numerous achievements, of particular distinction is the importance of non-shared environment, a term that he coined to refer to the environmental factors that reduce the similarity of individuals raised in the same family environment. Moreover, he has shown that many environmental measures in psychology show genetic influence, and that genetic factors can mediate associations between environmental measures and developmental outcomes.
In 1994, he was one of the 52 signatories of “Mainstream Science on Intelligence,” an editorial written by Linda Gottfredson (1994) and published in the Wall Street Journal, which declared the consensus of the signing scholars on issues related to intelligence research following the publication of The Bell Curve by Richard J. Herrnstein and Charles Murray (1994). This was a very significant event in his life. As he frequently points out, at the time when he began his research on behavioral genetics, it was practically scientific suicide to speak openly about the genetic conditions of individual differences, particularly regarding intelligence. As he emphasizes, for many years he kept his head down with his work. He found the courage to speak up only after The Bell Curve was published. Because of the general aversion to studies of behavioral genetics, he held off with publishing his famous book Blueprint: How DNA Makes Who We Are for 30 years (Plomin 2018).
For Robert Plomin, individual traits are largely a product of people’s DNA. He does not shy from discussing his own genetic predispositions as discovered from analysis of his genome. His polygenic score1 profile for psychological traits, which is the world’s first such profile, is detailed in Blueprint. Many people panic at the prospect of using polygenic score profiles similar to those described in the preceding passage. Plomin , however, sees more opportunities than threats. In his opinion, it will primarily facilitate preventative measures for many diseases, education better adapted to the potential of each student, etc. He views himself as a cheerleader for these changes and an incurable optimist.
Professor Plomin, in the prologue of your book Blueprint: How DNA Makes Us Who We Are, you wrote that you waited 30 years to publish it because you were counting on the volume of research results collected during the time would make it easier for readers to accept the theses in it. Nevertheless, I have the impression that when the book was published you found yourself in the middle of a firestorm of criticism at the point where ideology and politics mix with science (Comfort 2018). How do you feel in such circumstances?
I would disagree. I was, of course, very worried about how the book would be received. But I was quite pleasantly surprised by the response in contrast to the response I would have gotten even five years ago. Most of the reviews in the major media were very positive, whether in television interviews or in video blogs, as well as at around a dozen public events. I’ve been amazed at the positive reception. The only negative review I know of is one in Nature, but it hardly counts because it was by a historian rather than a scientist. He didn’t even address the book, he just didn’t like the message, and that review was so bad I think it actually helped because a lot of people were sympathetic and wondered how Nature could publish such a shabby piece of work.
Some of the critics claim that your theses represent the threat of a return to social Darwinism and eugenics or scientific racism. Could you respond to this criticism?
When talking to the public, I don’t see that at all. Not one person out of hundreds has ever raised those issues. It’s the media, and I think it’s laziness. Nature published something around a year ago saying that modern genetics shouldn’t be tainted with the history of eugenics because modern genetics in fact provides a much more subtle interpretation of genetics. Either way, I simply don’t worry about it. I don’t think people are concerned about a return to eugenics or determinism. If you understand the genetics, you won’t think that it’s leading to fatalism or determinism. These are issues I discuss in the afterword to the paperback edition of Blueprint that was published in June 2019. It presents my response to the reactions to the book and addresses these sorts of issues. As for eugenics, it bothers me a bit that people bring that up. As David Aaronovitch (2018) said in The Times, in what I think is one of the better reviews, it’s interesting how quickly critics from a certain political persuasion jump with alacrity from the science to talking about ethical issues because they’re not attacking the science. I don’t think they can. They’re just saying they don’t like the message. It’s a bit like the review in Nature where the reviewer says he doesn’t like the idea of genes being important. So they’re attacking on ethical rather than scientific grounds, which I think is an anti-scientific stance. I don’t necessarily have any trouble with that. What bothers me is that if you take it at its face, the Nazis were a totalitarian regime that killed lots of people with genetics as a rationale, but it was merely a fig leaf. There was no scientific justification for what they did, and most totalitarian regimes are environmentalistic. The science doesn’t conclude that environmentalism necessarily excludes a murderous totalitarian regime—think of Stalin’s Russia, Mao’s China, or North Korea today. Those are totalitarian regimes driven by an environmental hypothesis that people are all the same and the state makes people into what they want them to be. So, I think the basic message is that totalitarian regimes are bad for people, and the scientific justification they use is just a fig leaf with no basis in reality.
This is a good point because the results of your studies also call into question the work of many psychologists who have spent their entire lives studying the impact of environment on human behavior. How do these researchers approach you?
I think that if I had written Blueprint 30 years ago when I was first asked to, I would have been crucified. The reaction was so hostile and there was this view that genetics is bad, while environment is good. However, in the subsequent 30 years a mountain of data has emerged that has convinced most scientists of the importance of genetics. This isn’t just twin studies or adoption studies, but increasingly studies of DNA itself in unrelated people, and it’s very hard to argue with DNA. So I think there is much greater acceptance of a more balanced view that recognizes the importance of genetics as well as the environment. Furthermore, I would strenuously object to the view that “environment is good, genetics is bad.” I don’t think genetics could ever do as much harm as environmentalism has done. Think back to 30–40 years ago, when the textbooks said that schizophrenia was caused by what your mother did to you in the first few years of life: toilet training, breastfeeding, that sort of thing. That mother blaming is incredibly bad because—imagine that your child is turning 20, they become schizophrenic and then you’re told it’s because of what you did in the first few years of life. Most environmentalistic theories involve parent blaming, and this is wrong. There’s no evidence that early parental treatments cause schizophrenia. In contrast, there’s strong evidence that genetics is one of the major reasons why some people become schizophrenic and others don’t. I think the zeitgeist has changed and people are just more accepting of genetics. You hear it when people say “it’s in your DNA, it’s in your genes.” There’s no longer this knee-jerk reaction that says “genetics is bad, environment is good.”
Would you say that this very popular conviction, especially prevalent among psychotherapists, that early childhood is crucial for who we are when we become adults is a myth?
I think that used to represent psychotherapy 20–30 years ago, but clinical psychology has changed quite a bit. It’s become much more biosocial, recognizing biological and genetic influences. At first clinical psychologists were resistant for the same reason that educational practitioners are the biggest obstacle now, because they largely don’t believe in genetics. Clinical psychology mistakenly thought that if things have a genetic component that puts clinical psychologists out of business, because if it’s genetic then you can’t do anything about it. But if anything, that’s totally the wrong view about genetics—to say something is genetic doesn’t mean you can’t do anything about it; in contrast, it might actually mean you can do more about it. If you can identify particular genetic risks, perhaps you can work with that to improve treatment perspectives. This is because they realize it’s not genetics versus environment, that there’s a genetic component, there’s genetic risk, but you need to study treatments that take into account genetic differences so that you look for treatment by genetic interactions. I think that’s a hugely important and upcoming area.
Your research has also delivered results that demonstrate the uselessness of diagnostic manuals presently in use, such as the DSM. You yourself wrote in Blueprint that we cannot cure a disorder because there is no disorder, and so we can essentially toss our manuals into the shredder. Are clinical psychologists and psychotherapists taking advantage of the achievements of behavioral genetics and updating their methods of diagnosis and treatment, or are they still blindly groping around within the framework of the old model?
I see a lot of signs that clinical psychology, especially psychiatry, is much more advanced on this than some clinical psychotherapists who still believe in Freud and psychoanalysis, for which I see no evidence. I know there’s a resurgence of interest in psychoanalysis, which I understand at some level of just understanding yourself, but as a treatment I don’t see any evidence in support of it. In terms of genetics, I think it has huge implications for clinical psychology. The data very strongly suggest that, from the DNA point of view, we have polygenic predictors of schizophrenia and bipolar manic depression, and we will soon have DNA predictors of alcoholism, hyperactivity and all sorts of disorders. These polygenic scores involve thousands of tiny DNA differences into a multiple-gene aggregate. They’re normally distributed across a perfectly normal bell-shaped curve. There’s no evidence for any etiological break. So with schizophrenia, for example, there are thousands of tiny DNA differences responsible for the condition’s heritability. That means we all have thousands of genetic risk factors for schizophrenia. There’s no point at which you become schizophrenic or not. That probably has to do with the environmental stresses you’re under, and so when you take a view of the whole it kind of does tear up our diagnostic approach, the medical model that begins with diagnosis that says “do you really have schizophrenia or not?” But the problem is that a qualitative dichotomous approach is just simply wrong. It’s all quantitative. It’s not a matter of “either or,” it’s a matter of “more or less,”—all quantitative, not qualitative. I think that’s a very important concept to grasp. This is not to say that there are some people who have more problems than others. For example, in my area of educational problems: take reading disability, for example, which people try to dress up as a medical disorder by giving it a Latin or Greek name. You can call it “dyslexia,” which makes it sound like it’s more of a real disorder, and then the parent can say “doctor, does my child have dyslexia or not?” And it’s just all nonsense from my point of view. It’s all quantitative. Yes, kids have reading problems and we should work to deal with those problems, but let’s not pretend that there is this discrete disorder that we then have to cure, and partly those cures involve neuroscientists looking for a hole in the brain, looking for the one thing that’s gone wrong and made them reading disabled. Instead, it’s almost a whole new perspective to say it’s all quantitative. Some kids have reading problems more than others, and there’s probably a genetic component behind it, but it’s not a genetic component to have a reading disorder or not. It’s all quantitative. And that’s what I meant by “you can’t cure it,” because there’s no disorder to cure. What you can do is alleviate the problem quantitatively. If a kid has reading problems, you help them to have fewer reading problems, you hope you’ll help them to read better, but it’s not like you’re curing a disorder. You’re dealing with a behavioral problem that you’re trying to alleviate. It’s really an anti-medical model approach. I have no doubt this is true, and it is already beginning to take over—remember how DSM-5 tried to shadow these diagnostic disorders, this nosology of classification, with a quantitative approach by saying “here’s how you can measure these symptoms quantitatively.” But there was so much vested interest in believing in the medical model of disorders that they scuppered attempts to include a quantitative approach. This was because if they had done that, and people collected quantitative data as well as these ridiculous diagnostic criteria of “you must have two of these symptoms for three months, and then you must have another symptom for six months,” you could then ask empirically which approach works better in terms of prognosis and in terms of treatment: a quantitative approach or a diagnostic qualitative approach. So I think it’s very exciting how genetics is already transforming clinical psychology and will continue to transform other areas.
How should contemporary diagnostic manuals look in the light of the newest research results on behavioral genetics?
There’s such strong vested interest in these diagnostic manuals that there is very strong resistance against it, but I have no doubt in saying that in a decade’s time those manuals will be thrown out. People will realize they’re just pseudoscience. This is a difficult message because I’m not saying there aren’t problems. There are people who have thought disorders, and there are kids who have reading problems. So I’m not dismissing those problems. My question is, “what’s to be gained by pretending it’s an etiologically distinct disorder?” I would say “nothing.” In fact, there’s a lot of harm done because then people think “oh, it’s us normal people versus those schizophrenics,” and it’s not. We’re all schizophrenic to some extent, it’s just a question of how high our polygenic score is. This polygenic score of course isn’t everything, but that’s what’s driving schizophrenia genetically. However, genetics only drives about half of the problem. So it’s not to say it’s all genetic, but I think it’s a vital distinction that there are no disorders, there are just dimensions.
In Blueprint you wrote about generalist genes. What about the number of disorders? Do we need almost 400 as it is listed in DSM-5 now?
I think that’s another important example of the way in which the new genetics is transforming clinical psychology. What we’ve found, and what we’re discussing now, is that genetic effects are general across the dimension. They’re not specific to a disorder. It isn’t as though you have the genes for schizophrenia or not. It’s all quantitative. In a similar vein, there’s been another equally dramatic, almost unbelievable finding, that the genes for one disorder are largely the same genes for another disorder. The most striking example of this was with the first division you make when you’re diagnosing a patient with psychosis. Up until DSM-5, you could either decide that a patient is schizophrenic or that the patient has an affective disorder like bipolar or major depressive disorder. Until DSM-5, you couldn’t be both bipolar and schizophrenic, because that’s the major first division in the nosology of psychopathology. And you might ask “well, who says?” The answer is, it’s just people sitting in a committee deciding about these things. It wasn’t really empirically based. And so the shock came when the first genes were found for schizophrenia—they were the same genes that affected bipolar manic depression. In subsequent work it’s been found most genetic effects are general across all psychopathology. There are some genetic effects that are unique, but largely genetic effects concern one’s propensity to have mental health problems. But the specific route that propensity takes, whether it’s toward schizophrenia or bipolar, is probably mostly not genetic. I know that’s a lot to take on board, but one of the hottest areas of research now involves p, which is a little p that represents general psychopathology; just like little g represents general cognitive ability. The idea of generalist genes first came about in research on cognitive abilities where we found that the same genes that affect spatial ability also affect verbal ability as well as other cognitive abilities. Genetic effects are general, and the specificity comes from the environment. So the hottest topic in clinical psychology and psychiatry right now is general psychopathology p, which captures most genetic influences across all types of psychopathology. People are even talking about trans-diagnostic treatments. That is, instead of thinking that we have to find one specific treatment for schizophrenia and one specific treatment for bipolar, researchers are thinking “if the genetic effects are general, maybe we need treatments that work across many disorders,” because the problem isn’t the specifics of schizophrenia or the specifics of bipolar. The genetic problem is what they have in common, which is a lot to get your head around. But this is another example of the way in which genetics is transforming clinical psychology. Again, it’s going to tear up the diagnostic manuals that assume schizophrenia and bipolar are etiologically distinct disorders, because they’re not. Genetically, they’re mostly the same thing.
What you’re talking about will certainly impact the ways in which therapists have worked until now. When discussing child-rearing, you often stress that what we can influence is behavior, but not inherited tendencies (an aggressive child will not cease being aggressive but can control aggressive behavior). Should we take a similar approach to psychotherapy? Should we abandon the widespread faith in cognitive psychology that our thoughts and beliefs influence our behavior?
Cognitive behavioral therapy (CBT) and a lot of other modern therapies aren’t overly concerned with what caused the problem initially. They’re more concerned about developing healthy cognitive habits for dealing with problems that arise, so instead of ruminating, it’s basically about healthy or positive thinking. So I think that even without genetics, the forms of psychotherapy that work—not psychoanalysis, but therapies like cognitive behavioral therapies—seem more concerned about you simply dealing with your behavior. It could be your mental behavior as well, but it’s not trying to change you fundamentally. It’s trying to say “when you get anxious or your heart speeds up or whatever, and you start having panic attacks because you interpret those physiological feelings in a way that makes you think you’re about to die, you don’t need to think that way.” I think this is happening already without genetics, but I think genetics will help it along. It’s very important to emphasize that a trait or a dimension like panic attacks, anxiety, depression, even schizophrenia can show a strong genetic risk factor. That is, some people are more likely genetically to exhibit these problems than other people, but that doesn’t mean there’s nothing you can do about it. Causes and cures are not necessarily related, and I think that’s what clinical psychologists believe, at least many that I know at the Institute of Psychiatry. They believe there can be a strong genetic component, but that doesn’t mean you can’t do anything about it. And conversely your treatments can work even if you don’t know anything about the causes. For example, I have a strong genetic component tendency toward obesity, Body Mass Index. It doesn’t mean I can’t do anything about it. We all know we can lose weight if we just don’t eat as much and exercise more, but knowing my genetic propensity actually helps me in my battle of the bulge to realize it’s a lifelong fight to not put on weight and to try harder to lose weight. The main point here is that a connection between causes and cures isn’t necessary.
Are we witnessing the DNA revolution in clinical psychology and psychotherapy?
Yes, I think it’s already happening. As I say, a lot of the big studies, like the big randomized control trials of depression and CBT versus drug trials, they’re all looking at DNA. If you have a large study and you’re not getting DNA, I think this is a big mistake because the DNA is relatively cheap compared to everything else you do. For around seventy-five euros you can gather DNA, genotype it and bring a genetic perspective to bear on whatever it is you are studying. I know of a huge study in the USA that is looking at people with depression and how they respond to either CBT or drugs. There’s been quite a bit of work on that already, but by collecting DNA they can now ask whether some people respond better genetically to drugs or to CBT. About hyperactivity, there are studies asking whether DNA can predict which kids will respond better to methylphenidate, and eventually you might be able to predict which children will suffer if they’re given methylphenidate, which is the other side of the coin. Some children might respond very badly to drugs, and if you could predict who those people are, you wouldn’t administer drugs to them. Once some of these findings come along, I think everybody is going to demand that they get their DNA genotyped. That’s why there’s a major discussion in the UK now about making genotyping available on the National Health Service.
We can improve health care in so many ways. That sounds optimistic.
I don’t think it is optimistic, because as the Health Secretary of the National Health Service says, all of medicine is moving toward prevention, and to be able to prevent problems like severe heart attacks, alcoholism or obesity, you need to predict. DNA is by far the best early warning system that we have because you can predict from birth, which nothing else does. So I don’t think it’s optimistic at all. One severe heart attack costs the National Health Service about €750,000. You can predict that some people are at much greater genetic risk. The argument then becomes that it’s actually unethical not to do this.
But some people are also afraid to know about their future.
Indeed, and there are some people who smoke two packs of cigarettes a day too.
Let’s go back for a while to how the results of genetic testing are adopted by the scientific community. The resistance you experienced for years regarding acceptance of your research results, is perhaps a classic example of how the scientific establishment defends the official paradigms. It’s something Thomas Kuhn called a “paradigmatic war,” and which he considered a barrier to the development of science. Looking through the lens of his philosophy, would you say that psychology has undergone a revolution and a new paradigm has replaced the old one?
I think it’s more a matter of science winning out in the end. There were some fields, say, 30–40 years ago, such as sociology, about which we could worry whether they would remain an empirical science. What I mean by this is that data are the ultimate arbiter—you have to accept the data. You may not like it at first, and it may be hard to accept a change to your old view that everything is environmental. That’s a big shift to say “well, no, it isn’t all environmental. It isn’t just what your parents did to you. There’s a strong genetic component.” That’s a lot for people to take on board, and that’s why I avoided arguments with people. I just wanted to collect data, feeling that psychology will ultimately be an empirical science whose fundamental credo is that data rules. You have to look at the results of the data. You can argue against those results, but then more and more data arise leading to the same conclusion, which, as a scientist, you’ve got to accept. I think this is coming to a head with the replication crisis in science. Daniel Kahneman says that neuroscience is the poster child of failures to replicate. The results in behavioral genetics replicate. Nobody denies that. We have these huge findings about heritability being roughly 50% and about genetic influence increasing over one’s lifespan, supporting the idea of a general genetic influence on environmental measures. These are massive findings, yet they replicate time and time again. You started asking about the resistance to genetics. I’ve written a paper on the replication crisis and why behavioral genetics results replicate (Plomin et al. 2016). Psychologists used to be so resistant to genetics. They made it hard to publish or to get grants about genetics. In a way, they were doing a service to the field of behavioral genetics because it made it much harder to convince people. As a result, you needed more and better data to do that. This explains why behavioral genetics results replicated, because you had to work very hard to convince people that these results were true.
Would you say that this crisis we are experiencing in psychology is also a product of insufficient attention being paid to genetic factors now researched?
I wouldn’t say that in particular. I think it’s the problems that people have been talking about forever: p-values, chasing probability values, and samples that are too small. Genetics only comes into play in terms of the etiology of individual differences. Much of psychology is what I call normative, asking about what the human species does; for example, asking when kids develop two-word sentences, a focus on averages. Most experimental psychology is also about averages; you take a group of people and you randomly assign them to an experimental group and a control group, then you do something to make them different on average. Individual differences in the ANOVA design are called error. What genetics is studying is that error. It’s studying why people are different, and when you study individual differences you have to pay attention to effect sizes. Whereas if you study mean differences, there’s a strong temptation just to look at p-values and not to ask about effect sizes. I think a fundamental reason for the replication crisis is the focus on averages and p-values rather than on effect sizes—how much variance is explained.
What should we do first to start overcoming this crisis?
We should start paying attention to individual differences and effect sizes. When you publish something and you find, for example, that boys and girls supposedly differ in verbal and mathematical ability. In these large samples you do get significant mean differences, but the effect size is less than 1%. So if all you know about children is whether they’re a boy or a girl, you don’t know anything about their verbal or maths ability. I think there’s much to be said for focusing on individual differences rather than just treating them as a nuisance.
When discussing the crisis, there is frequent mention of the frauds that have been perpetrated by scientists in our field.
Indeed there is.
Considering your knowledge and experience, you can offer a credible assessment of a certain historical incident that remains a source of controversy today. Specifically, I’m referring to the case of Sir Cyril Burt. Considered a scientific fraudster, he arrived at very accurate conclusions with respect to inheritance of intelligence. Was he a master fraudster, a scholar too lazy to do research, or perhaps came up with his theories by chance? What do you think about his work?
I think that is a tremendous misrepresentation of the whole issue. That’s old hat which has been discussed in three books. It’s not at all clear it was fraud—that’s a simplistic media trope to use. As we’ve shown in papers, you can forget Cyril Burt’s data but still get the same results from the world’s literature. This Cyril Burt story diverts from the science of what we’re doing. Fraud contributes to the replication crisis, but I think it’s a rather minor issue. My colleague, Stuart Ritchie, is writing a book on the replication crisis. Fraud captures a lot of attention, for example the current issues with embryos and in vitro fertilization. I think, however, that most of the replication crisis is at a more subtle level. It’s not where someone just sets out to make up data, and for what it’s worth, most scientists don’t think that Cyril Burt made up his data. I don’t think that’s worth talking about anymore though. Let’s forget about all his data and we still get all the same results. What I think is more important is how, for example, I hear scientists in large consortia where people combine their data saying things like “well, if we take the results for that measure, they look a lot better. Look, the p-value is less than 0.05. So why don’t we just cut out this one sample and go with that other one? The results look a lot better.” That’s fraud. And that’s where the failures to replicate come from, because once you start picking and choosing which data and which analytic methods to present, probability goes out the window. You shouldn’t be allowed to talk about p-values anymore because you’ve destroyed their validity.
A good point. The growth in the importance of genetics in psychological studies has been accompanied by several other issues, such as epigenetics, which are becoming more popular among practitioners. With the help of epigenetics, outworn New Age concepts are being revived, some therapeutic practices are being justified, and attempts are even being made to revive Lamarckian ideas. Various contemporary gurus commonly employ the concept of transgenerational epigenetic inheritance to sell us on practices that involve the therapy of inherited traumas, or to work on changing our own genome. How much truth is there in any of this?
There’s hardly any truth in it. People like anything that’s anti-Mendelian and pro-Lamarck, and Lamarck was wrong. We don’t inherit acquired characteristics. The notion of transgenerational inheritance is incredibly exaggerated. The few examples in mice (Dias & Ressler 2013) and in the Dutch famine study (Painter et al. 2008) do not prove that we inherit epigenetic marks. In fact, there are major processes involved in conception that get rid of any epigenetic marks from the mothers’ chromosomal material. Gene expression is important, but it’s a response to the environment. While I think epigenetics is valuable for understanding environmental influences from a biological perspective, it has nothing to do with inheritance. What we inherit is DNA sequence. RNA, which is what expression is involved with, has evolved to respond to the environment. DNA evolved to faithfully transmit evolved adaptations across generations. I find it amusing how people jump on the epigenetics bandwagon, but it’s wrong to think epigenetics supplants genetics. We want to know everything that goes on in between genes and behavior, but this business about epigenetics somehow undercutting the importance of inherited DNA differences is nonsense from my point of view.
In discussing the genetics of individual differences it is impossible to avoid the highly controversial subject of gender studies. In the light of your research and your knowledge, would you say the claims of gender studies about the cultural sources of sex differences are justified?
Gender is fundamentally biological in that females have two X chromosomes and males have a Y chromosome, but that doesn’t tell you what the cause of the average differences between, say, boys and girls are. In the afterword of the paperback edition of Blueprint, one of the things I respond to is the criticism that I’m not talking about average differences between groups, like gender, class or ethnicity. I have good reasons why I’m not talking about it. It’s very difficult to understand the causes of average differences between groups, whereas we have very powerful tools for understanding the causes of individual differences within groups. So I choose to study the tractable problem of individual differences. The other thing that’s important to note here is that the causes of individual differences are not necessarily related to the causes of average differences between groups. For example, you can find that males are substantially taller in almost all cultures than females. This might lead you to think “well, that has to be genetic.” But it doesn’t have to be. In fact, the genes that are involved in height are not on the X or Y chromosome. They’re on the autosomes, the 22 pairs of chromosomes that are not the X and Y chromosome, which are called the sex chromosomes. I think it’s likely that the average difference between males and females in height is genetic, but conceptually it doesn’t have to be. It could be that the genes for height or hormones work differently in a girl’s brain than in a boy’s, and that could be driven environmentally. Perhaps height is a bad example to try and come up with an environmental explanation. The larger issue is that in developmental psychopathology we find some of the biggest differences between males and females—think of autism, hyperactivity, or even reading disability. It seems like anything with a neurological basis is worse for boys than girls. While you might think it’s genetic, every attempt to find a genetic basis has failed. It could well be something in the environment, or that the genes involved work differently in a boy brain and a girl brain, given that they differ, for example, in hormones. The issues around gender studies and some of the politics that are involved in it are interesting, but I refuse to talk about average differences between groups because, as I’ve said, I don’t think we have very powerful tools for nailing down the causes of the differences. I think that’s why there’s so much heat and so little light. Also, as I say in the afterword of the paperback edition of the book, I don’t have to study everything. I’m increasingly comfortable saying that if you want to talk about group differences, that’s fine. I’ve got enough on my hands with individual differences and I take enough flak for studying individual differences that I don’t need to stick my neck out and study average differences between groups.
Another hot topic is artificial intelligence. Is the growth of this field also opening up new research perspectives on behavioral genetics?
Not to my knowledge, because artificial intelligence is a normative approach, it’s not talking about individual differences. It might eventually be able to do that, but right now artificial intelligence means a lot of different things. For some people it involves analytics, a sort of approach to problems, like using advanced analytic techniques such as mathematical modeling. And those analytic approaches are being used in genetics, for example, to try and find genes when you’ve got billions of DNA bases involved and you want examine interactions between genes. But if what you mean by artificial intelligence is robots, I don’t see where that’s going to help us in behavioral genetic research, although robots have greatly increased the speed and lowered the cost of genotyping. I hope behavioral genetics will be open to new approaches. If you can show me something that will help me try to find genes related to behavior, I would welcome it. For example, machine learning algorithms are increasingly being used to solve the very complex problem of trying to find all of the thousands of DNA differences and their interactions that affect behavior.
We have devoted a lot of time to analyzing controversial issues related to genetics and psychology. Let’s talk about the achievements of the field. What great questions has psychology as a science managed to answer so far?
To be fair, I’m a one-trick pony. So I immediately think of genetics and how drastically psychology has been transformed. It’s been a huge step to go from Freudian thinking that everything is environmental, that we are what we learn, to recognizing that inherited DNA differences account for more variance than anything else we know about. It accounts for something like half of the differences between people. This is a big advance, and then the DNA revolution came along and showed that we can identify some of those DNA differences and begin to predict behavior from DNA alone. And that is an earth-shattering development in psychology. It’s what I love about behavioral genetics, something Peter Urbach and other philosophers of science talk about as a progressive science that builds on previous results rather than hopping from one fad to another. But if you look at recent issues of psychological journals, a lot of it doesn’t seem very progressive. It seems like a lot of fads where something is of current interest and people rush that way, and then a few years later people get bored with it. It’s not that they’ve solved the problem, but rather they just head off in another direction. I think a lot of psychology hasn’t been progressive. Although it’s self-serving and it’s my area of interest, I don’t think any other field in psychology has caused such a transformation over the last few decades as genetics has done.
Are there any great questions which remain unanswered?
From my very limited perspective, the big question everybody’s facing is not how we find some of the genes that predict the heritability of behavior, but how we find all of them. That’s going to involve new technology. Instead of dealing with SNP chips, which are DNA arrays that measure a few hundred thousand single nucleotide polymorphisms across the genome, we’re now moving toward whole genome sequencing, which identifies the sequences of all three billion nucleotide bases of DNA; this captures everything that you’ve inherited. The rest of the DNA variation must be there, and along those line there have been some big developments in recent months. I think the next big thing is whole genome sequencing, which I believe will make it possible to predict most of the genetic variability in behavior.
What questions are you trying to find answers to these days?
Most of my work has shifted from what we call quantitative genetics, which involves twin and adoption studies, to exploring the extent to which inherited DNA differences are important. For the last 20 years, and especially in the last decade, I’ve been much more interested in using DNA itself to make these predictions. The neat thing about polygenic scores is you can use them to make genetic predictions for any sample of unrelated people, without needing twins and adoptees. That’s going to make genetics much more accessible to psychology. And as I’ve said, if you have a very large and valuable sample and you’re not collecting DNA, you’re making a big mistake. You don’t have to become a geneticist. You can simply use DNA to put a different perspective on whatever question you’re asking. It’s really just the beginning of genetics in psychology, and I strongly believe that genetics is going to transform not only clinical psychology but also society in general and how we understand ourselves. 25 million people have already paid about 150 euros to have their DNA genotyped because there’s a real interest in this, and I think that interest is going to increase as we improve at predicting psychological traits from DNA alone.
Probably our conversation will also be read by students and young psychologists just getting their careers started and who are wondering what specialization to explore. What areas of investigation in psychology do you think are particularly promising and would recommend exploring?
Psychology is the most popular undergraduate major in the UK, and I think that’s because the wide range of questions that we address in psychology are questions that are of interest to all people. And even though I’m in a medical school now, so many people in medicine are coming around to the idea that much of the medical burden on society has to do with psychology. People come to the GP and sometimes they have a very specific problem, but very often it has to do with depression and psychological sorts of issues. That’s why I think psychology is going to remain an increasingly important field, not in itself as much as across all the life sciences. So I naturally think that psychology is important, but I would say that anyone starting out in psychology who doesn’t keep an eye on genetics is making a mistake, and psychology departments that aren’t teaching their psychology students about genetics are doing them a disservice. Genetics is going to become increasingly important and I would urge young people to learn about genetics even if they’re not taught about it, because it’s going to change all aspects of psychology. It doesn’t matter whether you study perception or the nervous system. Whatever you study in psychology, especially clinical psychology, you’ll make bigger advances by considering genetics.
At the beginning of your career you yourself choosing the subject of your research at the same time decided to swim upstream against the dominant tendencies of the time. Would you encourage young people to adopt a similar attitude?
People shouldn’t just be contrarian and say “everybody’s going this way so I’ll go in the opposite direction.” The reason I do what I do was that I was so impressed with the power of genetics that I couldn’t understand why I’d never heard about genetics in my undergraduate psychology, or even in graduate psychology, up until the point I had a course in behavioral genetics. If you do find something that interests you, go with it, even if it isn’t currently in favor in psychology. Particularly if you have the right disposition, by which I mean you have to be strong to go against the flow and be prepared for people not liking you. That’s a small price to pay for being right.
Who would you point to as an example to follow?
I had a lot of heroes, but for me it wasn’t about just following someone else. I always did my own thing, but there were people I admired for the way they wrote and for the cleverness of their ideas. My religion is science, not individual scientists. For example, when I write, you may notice that I never use anybody’s name. I’ve never said “Jerry Kagan said this,” as some typically do, talking about people who are famous because it gives credibility to what they’re saying. I believe people belong in the parentheses, in the brackets at the end of the sentence. You can say something and you can cite a study, but you’re focusing on the ideas rather than the person. That has always been important for me—I believe in science, but I try to resist the temptation to believe in scientists and to hold them up as heroes because I don’t think the people are important. I don’t think I’m important. I think my research is important, and in the end that’s what counts. If you’re too focused on people and personalities, I think you start worrying about awards and getting credit. I see this happening to a lot of people as they get older—they’re concerned about not getting proper credit for what they think they discovered. Well, they’re not important, what’s important is the discovery. That makes us feel part of a community of science instead of a cult of personality of scientists. But that goes against a strong evolutionary trend—I think we do have this tendency to want to have heroes and to make it about people rather than about ideas.
That goes strongly against our individualistic culture. What should young people be careful of? Can you give them any other tips?
I’m not so good with giving advice. I like to get the brightest students I can, those who are hardworking and really interested in being scientists, and often we get the best students from all over Europe. One criterion I use in selecting students is that they’re feisty—they don’t agree with everything I say. They don’t take my advice, and I think that’s a good thing, to be a free thinker and to argue with people. But then if you’re wrong, you have to be gracious and say “yes, I was wrong about that.” I’m not really one for giving people advice except to say “don’t get locked in too early with a very specific area.” My general strategy is like in chess, to play the board broadly at first and keep as many doors open as possible. In psychology, that would involve embracing new techniques that come along, because the field is moving so fast. If you become too specialized too early, it’s like Darwinian evolution. If that niche that you’ve very specifically evolved toward continues to be important then you might be a big fish in a little pond, but chances are the field will move along and you’re left in an evolutionary backwater, specialized in something that’s no longer of interest. I would encourage people to go at it broadly, which is why I like the American model of graduate school in psychology, where it’s assumed you don’t know anything. Even if you hold a psychology undergraduate degree, you’re taught everything again at a graduate school level for the first two years of training. Whereas in the UK and I think in a lot of Europe the undergraduate psychology degree is viewed as the final degree, and then graduate school is viewed as a specialty. You’re supposed to come in and say “what I want to do for my Ph.D. is this very specific project.” I think that’s a big mistake because you don’t know enough after an undergraduate degree. What I like is for my graduate students to end up doing a Ph.D. on some topic they didn’t even know existed when they got started. So, while I can say that I don’t like giving advice, that’s the advice I do tend to give my students, even though they often don’t pay attention to my advice anyway.
Is there anything I haven’t asked you about during this conversation that you would like to mention?
There’s quite a lot to discuss regarding the new developments in DNA, but I think from a psychologist’s perspective we’ve probably touched on the biggest topics. I suggest that readers of your book read the afterword to the paperback edition of Blueprint, because while we have touched on most of the issues that have come up since the hardback was published, there are a couple we haven’t.
Would you agree to answer one of the 30 questions that my readers would like to ask the most eminent psychologists in the world?
Yes.
Please draw one.
I’ll take number 12.
Interesting question: What do you believe is true, even though you cannot prove it?
I don’t believe anything is true. If I can’t prove it, I could say it’s a hypothesis. And I do have hypotheses, such as we will succeed in finding all the DNA that accounts for the heritability of behavior. But that’s just a hypothesis. I would say one of the phrases I most overuse is “it’s empirical.” You can say anything and hypothesize anything you want as long as it’s empirical, because in the end it’s the data that will determine whether it’s true or not. You can’t simply believe something’s true.
Do you apply the same philosophy to your private life?
Yes. That’s why I stopped being a believer. I became an atheist at the age of 12 for that reason. I don’t believe in religion, because you can’t prove it.