Thanks for reading! The best way to make sure that you don’t miss any of these conversations is to subscribe to The Good Fight on your favorite podcast app.
If you are already a paying subscriber to Persuasion or Yascha Mounk’s Substack, this will give you ad-free access to the full conversation, plus all full episodes and bonus episodes we have in the works! If you aren’t, you can set up the free, limited version of the feed—or, better still, support the podcast by becoming a subscriber today!
And if you are having a problem setting up the full podcast feed on a third-party app, please email our podcast team at leonora.barclay@persuasion.community
Steve Stewart-Williams is a professor of psychology at the University of Nottingham Malaysia, and author of The Ape That Understood the Universe (2018), Darwin, God and the Meaning of Life (2010), and The Nature-Nurture-Nietzsche Newsletter on Substack.
In this week’s conversation, Yascha Mounk and Steve Stewart-Williams explore the nature vs nurture debate within psychology, how much impact IQ has on an individual’s success in life, and whether evolutionary psychology can explain politics.
This transcript has been condensed and lightly edited for clarity.
Yascha Mounk: I really enjoy reading your Substack, which is very wide ranging from a psychological perspective. It's called The Nature-Nurture-Nietzsche Newsletter. So let's start with a question about nature and nurture. This is a question that I know a lot of people are very interested in, often without really knowing the latest. What is the best evidence today about the extent to which the most important human traits are caused by nature and the extent to which they are influenced by nurture?
Steve Stewart-Williams: I think there are basically two ways that you can tackle the nature-nurture issue. One is the nature and nurture of individual differences between people. The other is the nature and nurture of traits that we all have in common. I guess a good example to pry those two apart would be that all normally functioning human beings have the emotion of fear, for instance. So, the question would be, to what extent is that emotion a product of nature versus nurture? People differ though, in their proneness to fear and how fearful they get in different situations. The nature-nurture question there would be, what explains the differences between people? The first question I think is best answered by evolutionary psychology. Applying evolutionary biological principles to human beings and the human mind, the answer there might be that we all have the emotion of fear because fear encourages certain kinds of behavior that on average are adaptive—steering us clear at the edge of the cliff, for instance.
Mounk: If you're not afraid at all, you might tumble over the cliff or get eaten by the lion. So over time, people who lack that faculty are not going to be able to pass on their genes.
Stewart-Williams: Exactly, as are people who have that faculty too strongly as well. There's a sort of happy medium somewhere in between the two. One of the fields we’ll look at is behavior genetics, which examines the extent to which differences between individuals are correlated with differences in the genetic variants they carry.
Mounk: What would be a human universal—something that all humans have—that is not influenced by evolutionary pressures in our biology? Are there examples of things that every human being has, but is probably caused by nurture?
Stewart-Williams: I would say that there are, these days. Fear isn't the best example because I think that's a pretty obvious example where it is part of our nature. With other faculties like language, for instance, I think it's less obvious. I do think that we are by nature a talking animal, but that's more of a debatable issue than is the case of fear. As for the question of whether there are human universals, I guess part of the debate actually is for a given trait, whether it is a human universal.
Mounk: So once we believe it's a human universal, we're going to be assuming that it has something to do with our nature. People are going to go around saying, well, what about this tribe in some part of the Amazon that doesn't have a particular linguistic feature? or something like that. So perhaps it's not actually part of our nature. Is that the idea?
Stewart-Williams: Exactly right. But there are some things that probably would or could come quite close to being human universals that are not directly products of nature. I can only think of joke examples off the top of my head, like drinking Coca-Cola, having smartphones, doom scrolling, eating pizza and things like that. Maybe a more serious example would be literacy, which isn't universal, but it's close to universal in many nations and you can imagine that it could become a universal. Whereas I think with spoken language, there's a good case to be made that that's part of our nature—the capacity to learn a spoken language. Reading and writing are cultural inventions instead and could become universal ones.
Mounk: That's interesting. There's universals that are in the realm of something like technology that obviously are not part of our nature. Most human beings in the world own a t-shirt. But that's because those are technologies. You might think that the things that allow us to produce those technologies are human universals, but the particular technology that we employ today has to do with material culture—a particular stage of economic development, the universality of Western cultural influence in the world and other things that are much more contingent.
Stewart-Williams: Yeah, I see what you mean. I think maybe a better example would be something that could be a universal that's in our minds as opposed to t-shirts that we're wearing or technology that we're using. I guess reading and writing is implemented in the brain. So I think that probably would be the best example I can think of of a psychological trait that's nurture rather than nature that could become a universal.
Mounk: Before we move on to the very interesting question of differences between individuals, I know from a lot of people that they have these deep reservations against using evolutionary explanations for our nature today. In fact, a number of months ago I was teaching a chapter from one of my favorite books in political psychology, The Righteous Mind by Jonathan Haidt, and some of the students sort of rebelled against it a little bit and said that evolutionary biology is a discredited field. What they seem to have in mind, I think, is the fear that a certain kind of argument from evolutionary biology might commit them to political conclusions they don't like. There might be people in the manosphere, for example, who say, it's in the nature of men to want to sleep with as many women as possible and in the nature of women to want to find one man to commit themselves to. Therefore men should act like this and women should act like that… or something like that. Presumably, the answer to those concerns is to say that you can point out certain traits that we have as humans universally or certain differences between groups of human beings without thereby committing yourself in any way to a normative conclusion. So to those who may have this kind of skepticism about evolutionary biology, what would you say, first of all, to establish that there are in fact important insights to be gained from it, and secondly, perhaps to reassure them that they're not going to wind up getting committed to, for example, certain sexist views about the right division of roles in society between men and women?
Stewart-Williams: I think I'd start with the second part first. I would reiterate your point that the descriptive differences we might see between the sexes don't have any immediate political implications or ethical implications. To assume that they would is a fallacy of reasoning known as the naturalistic fallacy: the idea that the fact that something is natural therefore means it's good. Also, I think people are unnerved by the prospect that if something is natural, that therefore means that it's inevitable and can't be changed. Both of which I think are wrong.
I guess the first thing I would say is that even if that were true, even if your worst fears were true, it doesn't actually mean that an evolutionary biological approach cannot shed light on human nature and human behavior. We're heading into the territory of sex differences, which is where a lot of the misgivings come from. I think that a lot of the sex differences are actually not as big as people think. The differences are often quite modest. There's a ton of overlap between the sexes in most traits where you have average differences. But people do notice the differences. I would caution people that if we make it off limits to talk about these things, you're going to have more of a concern about people who think that there are obviously these differences between the sexes and that they're not being told the truth. That means it's more likely that they're irresponsible people. And the people who are lacking in nuance and talk about these things are more likely to take that ball, run with it, and draw inappropriate ethical conclusions from a garbled version of an evolutionary psychological perspective, as opposed to more responsible people with a more nuanced view.
Mounk: Yeah, let me make two points here: one where I fully agree with you and one where perhaps I'm not quite as sure that I agree with you. I think there's a strange fallacy of discourse control that pops up in big parts of our public intellectual life. I've written in the past about how journalists often overestimate the amount of influence they have over the views and the conclusions of their readers. One of the mistakes that many journalists have made over the last 10 years is to have the self-conception as wanting to save democracy. Now, I want to save democracy. I think that democracy is in peril and it's important to protect it. I think it's perfectly fine as a citizen to feel like one of your obligations is to do what you can to maintain this system of government. If journalists, particularly beat reporters and those who are reporting on what's actually happening in the world, allow that ambition to influence what they write, they're actually going to be doing damage because they think, for example, that if they don't talk too much about the fact that Joe Biden seems to have some amount of mental decline, that's not going to give sucker to Donald Trump. It's not going to make it easier for Trump to get reelected. What actually happens is the Democratic Party ends up being committed to Biden for far too long. They don't have time to have a real primary. You end up with Trump getting reelected. So there's all kinds of ways in which this desire to control the discourse ends up not working. What you actually mostly do is to convince people that they can't trust you and therefore to turn to other sources of information which are often less responsible and less accurate.
So I completely agree with you on that—if people can sense that there may be certain sex differences out in the world and you say there's absolutely none, that might get us into difficult territory. You're not gonna get the great and enlightened policies you want at the end of this. You're just gonna get people to mistrust you and possibly listen to people who have much more retrograde views. I agree with you on the naturalistic fallacy that just to say that something has a biological tendency doesn't mean that it's good or that we should encourage it or anything like that. But there may be an extent to which things that are hard coded in our nature make it very difficult to change. So for example, I believe that there are certain elements of human nature which make it very hard to live a fully communal life. As a result, every time that people have tried to live in communes (Nicholas Christakis has written about this)—in very different cultural contexts, different time periods, different ideologies, different religious aspirations, it works for a little while at the beginning when the founders are there and everybody's super committed, and then it always goes wrong in very similar ways. That does tell you how although it might sound lovely to live in that kind of commune, it just ain't gonna work. Now, perhaps some of it is true for sex differences as well.
One of the famous studies is that if you believe in a nurture account—jumping ahead here a little bit to the topic of sex differences—the reason why men are more likely to want to be engineers and women are more likely to want to be teachers is just the cultural norms in that society. But you should think that the more egalitarian a society is, the more women you’ll have that are going to want to be engineers and the more men are going to want to be teachers. What you actually find in a really fascinating international study is that in Norway, Sweden and other societies that may not be perfectly equal, where there may be some nurture effects nevertheless, but that are clearly among the most gender equal societies in the world, the difference in what professions people want is stronger than in some of the less gender equal societies, for example in Iran or other parts of the Middle East or societies in Africa and Asia that are much more traditional on a number of dimensions. So doesn't that mean that we do need to grapple a little bit with the fact that if our ideal is—which I don't think it needs to be—that 50% of engineers need to be women, then perhaps if some of those differences actually have a nature component, and we're just not gonna get there. We kind of do have to give up on that ambition.
Stewart-Williams: I do think we have to grapple with it. I don't disagree with you there. The fact that something has an evolutionary origin doesn't mean that it's necessarily unchangeable, but it does mean that it could be difficult to change. It could mean that it's impossible to change at least without draconian interventions that are likely to do more harm than good. Your comparison with journalists not telling the truth, trying to get certain outcomes, I think is really apt here. The best policy for journalists and for scientists talking about these kinds of issues is to just tell the truth—the whole truth, and to do so carefully and responsibly. To do that rather than trying to just put out the truth to try to accomplish a certain kind of aim. Our best bet for making the world a better place is to have as accurate a picture of the world as we possibly can.
I do think that there may be limits. And I think that trying to get 50% men and 50% women in every field is an example of where we may find that we come up against limits to what we can do without really coercive practices. That gender equality paradox that you mentioned is quite a firm finding. It's not just STEM fields, it's many different psychological traits. The prediction from a social role and socialization perspective would be that the more patriarchal a nation is and the less gender equal it is and the more men and women are treated differently and funneled into different social roles, you would anticipate that psychological sex differences would be larger in those countries. In some cases they are and in some there's no real association. But often and very surprisingly you do find the reverse. The differences get larger in the more gender equal nations. I’m not completely sure why that is, but one possibility is that in societies that are more gender equal, they tend to be wealthier, have more education, and they're more individualistic. There's a whole cluster of traits that mean that people are more able to pursue what interests them most and nurture their own individuality. When that happens, one possible explanation is that certain inherent differences between individuals flourish more and so too do certain average differences between the sexes.
Mounk: That seems like an intuitive explanation. Here's a different example that might go in the same direction. To what extent does psychological predisposition influence job choice among second-generation immigrants—whose parents came to, say, the United States from a much poorer country, and who are smart and have strong high school records—compared to people who’ve been in the country for 150 years and come from a very affluent background?
I would expect the psychological disposition of a WASP in this example to have a much greater influence on the job they choose because they probably have the financial security and the knowledge of the society and other things that make them say, if I want to go off and be a Broadway actor, I can afford to take the risk to try. They may know how to go and do something like that, which probably takes more cultural capital to succeed in. The smart kid of immigrants would perhaps also thrive on Broadway if they tried, but is under a lot more pressure to do a secure lucrative job that might also be more accessible because there's a more straightforward career path. So you’d expect that a certain psychological predisposition would make someone gravitate more toward something typical of Broadway actors than investment bankers.
The share among those of investment bankers with a recent immigrant background would be higher than among the members of a traditional society. In a similar way, you might think the same is true for whether those average differences in psychological disposition between men and women express themselves. A relatively affluent society like Sweden or Norway has many more women who might take the risk to pursue careers that are true to their dispositions, whereas in a society that's much less affluent, they might simply go for the most prestigious job, or the job that's most likely to give them a decent income, even if that's not what they dream of doing.
Stewart-Williams: I think that's absolutely right. That's a big part of the explanation. It also does seem to be that it's not just at a behavioral level that you get the gender equality paradox. It does seem to be that personality traits and preferences—internal stuff—does also seem to magnify in more gender equal, wealthier, better educated, more individualistic societies as well as just the behavior.
Mounk: That's very interesting. It's not just that in one society, people are more likely to want to do a certain kind of thing, to have that kind of predisposition and then to act on it versus not act on it. It's that in fact, the difference in basic preferences about what to do or how to relate to the world is bigger in more gender-equal societies. That's more surprising. What could the explanation be?
Stewart-Williams: My best guess is that because people have more scope to pursue their own individuality, this phenomenon just emerges and flourishes to a greater extent. Height is a helpful analogy here. In wealthier societies, people are better fed and so they grow bigger, but also individual differences get larger, average sex differences get larger just because they have more of an opportunity to flourish. So maybe that's analogous to what's going on psychologically with people as well.
Mounk: I think we've inadvertently started to move into the second part of the nature-nurture debate, which is the differences between individuals. Tell us a little bit more about what we know about that in general. This is one interesting example we've been discussing in terms of the gender equality paradox. When you look at differences in intelligence, life outcomes, predispositions, psychological traits—in some of the most important things you can think of—what’s our best evidence about which elements of the difference between you and me, between me and my brother, between you and your sister, are based on nature rather than nurture?
Stewart-Williams: Well, the best evidence we have—we've got a ton of evidence from twin studies, adoption studies and studies of twins adopted out at birth—paint a pretty clear message. At least in broad strokes, the message is quite clear. Almost any measurable trait that people differ on psychologically or physically is partly due to genes, but not completely due to genes. So there's a famous set of conclusions from this area—behavior genetics is the name of the area—which are the four laws of behavior genetics. The first law is what I just said, basically that every trait is partially heritable, though not completely heritable. Typical estimates are, I guess, roughly 50-50. It spans from about 30% of differences being due to differences in genes to 70%, depending on the trait. Nowhere near close to 100% though. I think the most intuitive way to grasp that it’s not completely down to genes is to look at identical twins—because if it were just about genes, they would be literally identical. But they’re not. They’re surprisingly similar, yes, but never identical. Even with traits like schizophrenia—for instance, if one identical twin has it, there isn’t a 100% chance that the other will too. The likelihood is about 50%.
Mounk: That's very interesting. Tell us a little bit more about what exactly twin studies do. The idea, presumably, is that you look at people where genetics is held constant. Obviously, this is not the case between siblings because even though you come from the same parents, if you're full siblings, there's going to be some random variation in which genes you inherit and which express themselves and so you're only about 50% equal. But with identical twins, you are going to be fully equal. Presumably, you look to some extent at identical twins within the same families, but particularly, when possible, at identical twins who have been raised in quite different families from each other.
Stewart-Williams: Yeah, that's the gold standard study—identical twins raised separately. That's the best example. A lot of the comparisons come from comparing identical twins with fraternal twins. If you have enough of them, you're right that with fraternal twins, the average is 50%, but it can be more or less. If you have a sufficiently large sample—and these studies often do—the average tends to hover close enough to 50% that you can use it as a reasonable approximation. Essentially, identical twins have about double the genetic similarity of fraternal twins.
Mounk: It turns out that the identical twins have much more similar life outcomes than the fraternal twins. Here, obviously, what you're doing is to hold constant in the same environment, because not only are they growing up in the same family, they're growing up in the same family at the same time. Whereas with siblings, it might be that there's some event that happens in the three or five years before the younger sibling is born, where the family gets much richer or much poorer, there's some psychological trauma, you're mostly ruling that out once you have those fraternal twins or identical twins.
Stewart-Williams: Exactly. I think fraternal twins do tend to be somewhat more similar because of that. There's no cohort effect. They are at the exact same age, and somewhat more similar than regular siblings who are born at different times. But identical twins are a lot more similar than fraternal twins. One question people raise is that maybe that is because if they're raised in the same home, people just treat identical twins more similarly to each other than they treat fraternal twins for some reason. You can measure that and typically they don't treat them much more similar anyway. But in any case, you can get around that by looking at identical twins who are raised apart. You can just compare those identical twins raised together or fraternal twins raised apart or together. Even when they're raised apart, identical twins are notably more similar than two randomly chosen people at the same age or fraternal twins raised apart.
Mounk: Give us a little bit of a sense of the magnitude of these outcomes. How much more similar are identical twins compared to fraternal twins? Or a different question, when you take these identical twins and they do end up getting reared apart, how similar or different do the life outcomes end up?
Stewart-Williams: Take a trait like IQ for instance. If you calculate the heritability level of that trait in early childhood—five, six, that kind of age—it's maybe 20% heritable, and there's a notable effect of growing up together in the same home. Two individuals growing up in the same home are more similar than they would be if they'd grown up separately in early childhood. But then one of the very counterintuitive findings from behavior genetics is that as kids get older, the heritability of IQ actually goes up rather than go down, as people have more experience and more nurture. So by early adulthood, a typical estimate might be about 0.5, e.g. 50% of the differences among individuals in terms of their IQ scores are due to genes and the rest is due to experience and non-genetic factors. Interestingly, though, the effect of the shared family environment, or the level of similarity caused by growing up together, wears off as people get older. Even though it's not all due to genes when people are adults, within the normal range of family homes, almost all the similarity between siblings including twins is due to genes rather than the shared environment. Identical twins who are reared apart are basically as similar in IQ and many other traits as identical twins reared together. The identical twins reared together are not much more similar by adulthood than the identical twins reared apart, even though it's not all down to genes.
Mounk: Unpack that for me a little bit, because if you're saying that something like 50% of the heritability comes down to genes, I would expect that at the age of 30 or 40 there would still be some very significant differences between them. But what you're saying is that the older they get, the more similar they end up becoming. Somehow, even the similarity that seems to be down to growing up in the same household appears to have this genetic component. Help me puzzle through this because these two things seem to pull in slightly different directions.
Stewart-Williams: It is a very odd finding. They do become more similar as they get older. They're not completely similar though. And it seems that by adulthood, the similarity is largely due to shared genes and the environmental contribution seems to be largely to create the differences between them. Bizarrely, sharing an environment and growing up together doesn't seem to make people much more similar than they would be otherwise. Often when people hear that, they think it's all down to genes. But it's not because identical twins are not literally identical. It just means that the environment, whatever it is in the environment that's shaping people, seems to make them different from each other much more readily and easily than it makes them similar to each other. The similarity seems to come mainly from the shared genes.
Mounk: Let me try and puzzle through this because I'm not sure I fully grasp it right now. Let's take two 50-year-old identical twins who have exactly the same level of IQ. That's not always going to be the case of course, but in this particular case, that happens to be true. There, you're saying that the reason why they're so similar is their genes. Now let's look at two 50 year-old identical twins that were reared apart and they have a significant difference in IQ. Let's say that one of them has an IQ of 110 and one of them has an IQ of 90. That is likely to be the case because they grew up in vastly different environments. Perhaps one of them grew up in an environment that really maximized the opportunities of this child and gave him access to all kinds of developmentally appropriate learning materials, etc. Perhaps the other one had some amount of genuine food deprivation or some significant influence on the development that really meant that they weren't able to reach their full potential. Is that thinking in the right direction?
Stewart-Williams: It is. It's just tricky though, because when talking about heritability, it applies to groups of individuals rather than specific pairs of individuals. So if you have identical twins who have the same IQ, the fact that it's exactly the same may or may not be due to nature or nurture. In that particular case, it could be something different. It could be more due to nature, more due to nurture. One individual is smarter and then had a head injury that took them to the same level or something like that. So it applies to groups of individuals. If the correlation between IQs is 0.7 in a group of identical twins, to what extent is that due to genes? It's about 50% due to genes, 50% due to other stuff.
In terms, though, of environmental factors that change IQ, I think the ones that you're suggesting are very plausible candidates for things that could affect individuals' IQs. Serious things like food deprivation. Opportunities as well. I should say that what I'm talking about now in terms of the effect of the shared environment is called the second law of behavior genetics. That's the idea that the effect of the shared environment is much smaller than the effect of genes. Sometimes there's no effect at all. Growing up together doesn't make you any more similar than you would otherwise be. It's worth pointing out that that mainly seems to apply when it comes to just within the normal range where people have decent opportunities, are fed well and the like. I think outside that range, all bets are off. If people are being abused, suffering food deprivation, or a total lack of opportunities, that's certainly going to have an effect.
Mounk: Once you've cleared a certain minimum threshold, you're within the normal range. But if you have a child of a billionaire and you have all of the possible opportunities, that doesn't make that big of a difference—or is that similar? Is it that on the one hand, if you have food deprivation, that's obviously going to ding your IQ a lot and on the other hand, if you have all the best tutors in the world, it's going to go up a lot? Or is it only on the downside? It's only on the risk side that makes a big difference?
Stewart-Williams: I think unfortunately, it's mainly on the downside that you have a big difference. You get diminishing returns. It just seems to be a lot easier to throw development off course than to massively increase it.
Mounk: I had Emily Oster on the podcast recently, who's a great economist thinking about some questions on parenting. I would love to get your feedback on one stylized takeaway that I had from that conversation. Roughly speaking, it was that the impacts of parenting tend to be much smaller than people realize. They agonize about how much screen time children are going to have, about whether they can afford to sign the kid up for that extra ballet class, whether they're being a bad parent if they don't engage with the kid for an extra half an hour at night because they're tired that day, etc. The studies that she's looking at say that these things really don't make much of a difference. Now, she doesn't get into the nature-nurture debate, but you might speak to it from either perspective. A huge part of the development is just predetermined by genes. Identical twins raised in completely different households are going to end up having pretty similar results. What you're doing in your household is not going to matter that much. And to the extent that that household matters, it's really going to be at the extreme.
So if you are a loving parent who's able to provide your child with basic sustenance, if you make sure—which is hard enough—that they're not going to get some horrible trauma in their childhood, that they're not gonna starve, you really should relax. Don't worry so much about it. Your choice about the extra ballet class or which school to send them to or how much cream you're gonna give them, is just not gonna make a big difference. Emily might have quibbles with how I presented this, but that was my personal take-away from the conversation. Would you roughly second this or do you have quibbles about that picture?
Stewart-Williams: No, I would definitely second that. I think that that is definitely what the second law of behavior genetics is pointing to. Within the normal range, little differences in parenting philosophy, the number of books that you have on the bookshelf, screen time and time that you spend with your kid, all of that kind of stuff does seem to have surprisingly little impact. I'm quite strongly persuaded by that and agree with the implication that you should chill out. Certainly with my own kids, though, even knowing that, I found myself agonizing about those things and thinking about small differences. So it's easier said than done, I think. But I do think that she's right and I think the implication is true as well.
I guess the only thing I'd add is that I think the fact that it doesn't make a big difference isn't because it's all just down to genes. It's partly because genes have a bigger impact than people think, but it's also partly just because we don't have a particularly accurate picture of what it is in the environment that does shape people. The environment is having a big impact, but I just don't think things like half an hour versus an hour of screen time or the extra ballet class are among the environmental things that make a difference. We have less of an idea about what does make a difference than we sometimes think that we have.
Mounk: By the time a baby is born, the genes are set in stone. A lot of the environment is also set in stone. You're going to be a parent in 2025, in whatever country you live, with a certain socioeconomic status, with a certain set of personality traits. Any sort of choice you make within that is just going to make a very marginal difference to those basic background conditions.
Stewart-Williams: That's very true. Yeah, I completely agree. Actually, I do think the country that you're born into does seem to have a big difference. The economist Brian Kaplan has argued that quite a bit. Within a given country and within a given socioeconomic sphere, we don't have a huge idea what the environmental contributions are. But we do know that moving, for instance, from a poor country to a rich country has a huge impact on kids' life outcomes. All the odds are that they're going to do a lot better in the wealthier country than they would have if their family had not moved and they'd grown up in the poorer country.
Mounk: That's very interesting. What kind of life outcomes are we talking about here? It's unsurprising that they might end up making a lot more money over the course of their lives because they're just in a context where even a similar kind of job is going to command a much higher wage. Is that also true, for example, for the kind of IQ they're going to have or the kind of psychological traits they're going to have or other things where that might be a little bit more surprising?
Stewart-Williams: It's a great question and I'm not completely sure of the answer. With IQ, I'm not so sure. With education, I believe that it gets elevated. Health, I think, is elevated. But in terms of other surprising things, I'm not actually sure.
Mounk: We've talked about two laws of behavior genetics. I believe you said there's four laws. Are there laws three and four that we need to cover before we move on to the next topic?
Stewart-Williams: We could do that briefly. The third law of behaviour genetics is the flip side of the second one. The second one is that the shared family environment doesn't have a particularly big impact. The third law of behaviour genetics is that there's a lot of variation left over that's not explained by genes or by the shared family environment. So there's what's called the non-shared environment, which is the environmental contribution that shapes individuals over and above genes and growing up together. I'm not completely sure what that is. The interesting finding there though is that it's been argued—pretty persuasively, I think—that just random chance, random noise, and development is a bigger part of what shapes who we are than we might want to think or than we might intuitively think. So a lot of the differences between us is due to genes, a little bit due to the family environment that we grew up in, and then a non-trivial amount is just due to random developmental noise, especially early on.
Mounk What kind of random developmental noise might there be? An interesting study popped in my mind when you said that. Younger siblings tend to have somewhat lower IQs than older siblings. You might think that's about nurture, that perhaps parents somehow spend more time with or invest more resources into older siblings. I read a recent study which seemed to suggest that perhaps older siblings are more likely to pick up colds and other kinds of mild viral diseases that they might then pass on to their younger siblings and that perhaps creates enough developmental noise to ding the IQ of those younger siblings. But with two or three percentage points, the difference isn't huge. Perhaps that kind of randomness at some crucial development stage—they happen to have a flu for a couple of weeks—somehow dings them by a couple of IQ points in a lasting way. When we're talking about random noise like that, what kind of factor are we talking about?
Stewart-Williams: Yeah, it is that. I think that's a really interesting idea, by the way. They're more likely to get the flu or something like that at a younger age than their older siblings would have been. But that would be systematic in that case because there's an increased chance that the younger sibling is going to be in that situation.
Nonetheless, if you do happen to breathe in a flu virus versus not breathing in a flu virus at a certain age, that's basically random. That's as random as getting struck by lightning. But it could have that effect. One way I think Steven Pinker put it in one of his books is that in setting up the brain a neuron zigs rather than zags for no particular reason. There's just some randomness in them—random accidents. Somebody happens to slip and bump their head for no particular reason, unrelated to personality traits or risk-prone-ness. Those would be the kind of things. Accidental stuff.
Mounk: Earlier the message was don't worry too much about what you're doing because the basic life outcomes of your children are set in stone. Now I feel like people could latch onto this part of the conversation and say, my god, whether or not my toddler falls over and bangs their head one more time could make this huge difference!
Stewart-Williams: Yeah, they might get a lot more worried about early life flus and keeping people away from that. That is tricky. I'm not sure what to say about that. I would say that the difference that started us down this track, like you mentioned, is a small difference. That IQ difference. I guess a couple of points is a big difference. Possibly, there are some things that have a bigger impact than others. But actually, I think one meta lesson from psychology is that pretty much nothing has a huge impact. Psychology is quite tricky, I think, in the fact that the vast majority of influences on us have a really tiny impact, but there's just lots of them.
Mounk: Interesting. Yeah. So if it's not enough to get the flu, you have to get the flu and get dinged on your head and seven other things. Then, perhaps you'll go to a university that's a little bit less selective, which might not make you a less productive or happy person either. So if you want to tell your parents to relax, you can always find reasons to relax. But I feel like this is going to make people a little bit more nervous. I believe we've now covered three of the four laws. What is the fourth law of behavioral genetics?
Stewart-Williams: The fourth law is the fact that complex traits are never just due to one gene, two genes, even half a dozen or a dozen genes. They're almost always due to hundreds or thousands of genetic variants, each of which has only a tiny impact. But collectively, they have a pretty big impact, accounting for the heritability of the traits. That's not just due to one gene or a couple. Maybe 10, 20 years ago, you'd see a lot of headlines that would say, we've discovered an IQ gene, gay gene, warrior gene or a longevity gene, etc. The vast majority of those have failed to replicate. They were false positives due to studies that had samples that were too small, basically. But what does seem to be replicated now is that hundreds of thousands of genes contribute to almost every complex trait that you care to name.
Mounk: What you're trying to tell me is that my biology teacher lied to me when they talked about a nice simple example of a dominant gene and a recessive gene—if my child gets the recessive gene from me and the recessive gene from the mom then that's presumably true for certain diseases or for certain kind of conditions but for most of the interesting character traits, that's not going to be the case.
Stewart-Williams: That's exactly right. With eye color, that dominance model does work, but most genetic effects don't work like that. They're additive rather than having that dominance recessive structure.
Mounk: So much for Mendel. We've been using the concept of IQ on and off in this conversation. I know that that is another concept that people who don't know the scientific literature well are often quite confused about. On the one hand, there's a popular fascination with IQ tests and the importance of IQ. On the other hand, there's quite a lot of skepticism about the idea that these strange tests where you're supposed to look at patterns and guess which next pattern is the logical continuation have a meaningful predictive capacity about the world or capture some really meaningful difference in intelligence between people. What are the psychological findings about, first of all, how predictive IQ is of different kinds of life outcomes and other things that we might want to care about?
Stewart-Williams: Yeah, well, it is surprisingly predictive though only one factor among many. It does seem bizarre. Doing a whole bunch of quirky little tests could give you one number that would predict lots of stuff. It seems a little implausible on its face, but it does seem to do it. Least surprising is that it does predict how well people do in education. At first glance, that does sound unsurprising. But then, I think it is a little bit surprising that a single number does a reasonable job of predicting how people are going to do across all their different school subjects. You could easily imagine that it wouldn't come out like that, that you would need several different numbers for different subjects. But actually it does seem that that one number does a reasonable job of predicting.
Mounk: Right, it's not obvious. It may perhaps be obvious why looking at this set of patterns and being able to predict the next logical step in the sequences might make you good at math. It's harder to see why it would predict whether or not you're good at writing an essay about history. But you're saying that, empirically, it just turns out to be true.
Stewart-Williams: Exactly right. But it also predicts other life outcomes further down the track. It predicts career success, job success, income, longevity, health, and mental health. Contrary to the stereotype, higher IQ is statistically associated with lower rates of most mental illnesses. Possibly anorexia is an exception to that rule. But for the most part, you look at the curves, rates of mental illness come down as IQ is going up. Those are some of the main ones. Those are things that really matter to people, right? How well you do at your job really matters to people. How long you live really matters to people. Health and mental health, these are all really important stuff, not just quirky random stuff that doesn't matter too much in the real world.
Mounk: One of the ways in which that runs up against my intuition from real life is that I can see how all kinds of intellectual skills might be related. But there's also ways in which people who are genuinely pretty smart are really bad at other things. I'm thinking of an acquaintance of mine who is literally a legendary figure in one field of mental activity. People might be able to guess who I have in mind. But, he has a terrible sense of orientation. You try to meet up with him on some street corner of New York and he will be standing at the wrong street corner. I don't want to compare myself to that person in any way, but I think I'm good at some things intellectually. I'm not the stupidest person in the world. I'm generally good at various kinds of things. I have mild face blindness—not formally diagnosed, but I've taken online tests provided by various universities, and I think I’m at the sixth percentile of the population for my ability to recognize faces. Not disastrous, not one of those cases like Oliver Sacks, who wouldn't recognize his own mom out of context. But we're looking at each other as we're having this conversation and if you sat next to me in the subway tomorrow and said hi to me out of context, I would have no idea who you are. So how do those two findings go together? If IQ is generally suggestive of all these things, why is it that people who clearly seem to be pretty good at a lot of mental activities can then have other kinds of mental activities at which they're just terrible?
Stewart-Williams: It's really interesting, isn't it? I think with the face blindness one, that is less surprising. I've seen research suggesting that people's facial recognition abilities are pretty much separate from IQ. It does seem to be a function implemented in the brain in some way that seems to be separate from however general intelligence is implemented. The spatial one is more surprising, I guess, and less common as well. It's just less common to have massive discrepancies between different faculties that make up IQ. Spatial abilities do correlate with IQ. First of all, with IQ, even though it is just one number and it's a pretty useful number, within that there are different cognitive abilities. It's not the case that everybody has the same level of those abilities. Everybody has strengths and weaknesses. One of the reasons that IQ isn't everything is the fact that it does obscure those strengths and weaknesses. It is interesting that you can have such a big discrepancy. It does seem like you can. I don't really know what else to say about that, but there is variability and I guess there's variability among individuals in terms of the extent to which you have those discrepancies. At the tails, you're just going to have a small number of people who have very large discrepancies.
Mounk: Yeah, and I suppose the other thing to say about it is that it's one thing to be strongly correlated. That doesn't mean that in each individual, all of those things will be correlated. It may be true for a lot of the people that I'm thinking of that general cognitive skill is correlated in such a way that nine out of 10 tasks are really good. There's always going to be one out of 10 tasks where they happen to be somewhat less good. Perhaps if you modeled that into 10 different faculties, you would still end up with a very strong statistical correlation, even if on one of the 10 they're not very good. What would be your best parsimonious model of trying to predict life success? You've made a strong case that IQ goes into that. High IQ predicts life success and perhaps to have extraordinary success at some field of endeavor, you really do have to have a high level of IQ. Presumably there's other things that go into that as well. When I think of people who might have a high level of IQ but who haven't turned out to be very successful, usually either they're really lacking in some form of self-control or have particularly strong mental health challenges or are lacking in some form of social facility, they're really lacking in an ability to cooperate with people. Those are a few other things that I would intuitively throw in where you have to have at least a minimum threshold of competence at those other things in order to maximize your chance of success. What would you say if you are designing a character in a sim game to see whether or not they're going to be successful? What qualities would you be trying to double click on to make sure that they have a certain level to guarantee that overall success?
Stewart-Williams: Definitely not just IQ. You're absolutely right. I think some people become IQ fanatics and seem to put everything on IQ. It's certainly not the only factor. Personality as well comes into it. I was just thinking about the literature on the Big Five and how they predict life outcomes as well.
Mounk: The Big Five personality traits, sometimes known as the ocean test: openness to experience, conscientiousness, extroversion, agreeableness, and neuroticism. And what did I think? Is it conscientiousness and neuroticism that are particularly predictive?
Stewart-Williams: Yes, conscientiousness is the most predictive of the Big Five. IQ is the single best predictor, followed by conscientiousness. That would embrace traits that you mentioned like self-control. Neuroticism is the next one after that, and that's a negative predictor of many life outcomes. It's a weaker predictor than conscientiousness though, in turn weaker than IQ. Agreeableness would capture some of the inability to cooperate, inability to get on with people. People who score very low on that are highly disagreeable. That might come under that. Some of mental health is captured by extreme neuroticism. Depression and anxiety and the like. That wouldn't capture things like schizophrenia and other severe debilitating things.
Extroversion is less predictive. It tends to be associated with better outcomes in a lot of ways, job outcomes, for instance.
Mounk: What about this idea that I threw in there, but I hadn't quite thought through? My intuition is that you can certainly be very intelligent and not successful in life. There's all of those examples of general losers in life who are always going on about how they’re members of Mensa. That presumably is because even if you have a high IQ, there's all kinds of things that can throw you off a course of success, including random chance, but also a lack of conscientiousness, a lack of realness, a lack of these other kinds of things that are necessary. What about the inverse claim that I threw in there, which is that people who are particularly successful in the world like inventors, writers, musicians, and athletes, need to have certainly a minimum threshold of IQ and perhaps a high level of IQ in order to have this special level of achievement in life. Is that true or do you think that somebody with an IQ of 100 or 90 can still have a real chance, not just of holding down a good job, having a decent life and a meaningful existence on Earth, but really of making the mark with some extraordinary level of achievement?
Stewart-Williams: That's a good question. I want to answer it statistically. I think that perhaps they do, but the further down that scale you go it just becomes less likely. Possibly there's a minimum threshold for a lot of forms of achievement where it gets close to zero at a certain point if you go far enough down the scale—zero chance of massive achievement. I wouldn't want to hazard a guess as to exactly where that is. But I want to double click on your point about how further down, it's perfectly possible to have a meaningful life where you're contributing something to the world. Extreme accomplishment is a different story. Only a small number of people even with very high IQs have extreme levels of accomplishment.
Mounk: Of course, but I know there's studies on that. Let's say we took the CEOs of S&P 500 companies. I know from conversations with people that there's really widely varying intuitions about what percentage of that group would have ordinary IQs. I think there's a lot of my friends who would say, oh, I bet that a half of them or a quarter of them have an IQ of 100. Others would say, I bet that perhaps one or two somehow inherited the company and have an IQ below 110 or perhaps below 120, but really the vast majority of them would have above 120, above 130, something like that. Do you have a sense from the literature of which of these two kinds of claims is more likely to be right?
Stewart-Williams: I think that the second group of people would win the bet. There's research for instance by Jonathan Wai and he has looked at CEO IQ levels. They do tend to be very high.
Mounk: One thing strikes me as interesting here, and I'm going to make some assumptions about listeners to this podcast, but I think it's an assumption that probably is justified. It's probably people who seek out intellectual content and have a real interest in that, which certainly means they have probably openness to experience and other kinds of psychological traits, but this probably also means that, on average, they have a high IQ. What I'm struck by certainly in my social milieu is that in a place like the United States in particular—a society that's incredibly stratified, probably much more stratified than most in the history of mankind by IQ level—to get into a good college, you have to do well on the SAT. The SAT is basically an IQ test. I recognize I'm saying something about my own social circle here, but it strikes me when people in my social milieu say, this person is kind of stupid, look, they're successful and they're kind of stupid, they're talking about people who to them seem rightly or wrongly, fairly or unfairly, strikingly unintelligent within a group that is highly, extremely selective in IQ. So the people who they would describe as being of ordinary intelligence may in effect still have quite unusually high IQ relative to the average or to the median of a population.
Stewart-Williams: That seems entirely plausible to me.
In the rest of this week’s conversation, Yascha and Steve discuss why evolutionary psychology can explain politics, how schools can tackle polarization, and the replication crisis in psychology. This part of the conversation is reserved for paying subscribers…
