#1312 Kevin Mitchell: How to Understand Individual Differences in Psychological Traits

00:00 Ricardo Lopes: Hello everyone. Welcome to a new episode of The Dissenter. I'm your host, as always, Ricard Lops, and today I'm joined for a 5th time by Doctor Kevin Mitchell. He's associate professor of Genetics and neuroscience at Trinity College, Dublin. And today we're going to talk about individual differences in psychological traits, how we can study them, how we should interpret them, and also a little bit about group differences. So Kevin, welcome back on the show. You're already a regular, so yeah,

00:32 Kevin Mitchell: thanks Ricardo. It's great to be back. Thanks a lot.

00:35 Ricardo Lopes: OK, so, um, let me start by asking you this. Of course, some of the things we're going to talk about today, we've already at least touched on them, perhaps not into the level of detail that we're going to talk about them today when in our first interview about your book in Nate, but Um, what are individual differences in psychological traits? I mean, what are people talking about psychologists, neuroscientists, cognitive scientists when we refer to that? Yeah,

01:11 Kevin Mitchell: I think, I mean, I think we can just start with like what are we all talking about when we describe people in different ways. We're just recognizing that they have some stable patterns of behaving, um. That apply or are manifested across contexts, right? So, the idea is just, you know, the fact, I mean, English, for example, has 8000 words for what we could call personality traits, you know, shy, outgoing, lively, stubborn, conscientious, open-minded, what, you know, we've got so many, so many words for these things. And, and What they're tracking is that when we see someone else's behavior or our own, um, we notice certain patterns. So if we say someone's shy, that means that that over many different kinds of circumstances and scenarios, they're behaving in a way where we think, OK, there's something stable about the person themselves. It's not just the situation that they're in. It's that it's that person in the situation that produces the behavior, right? And so it's just a recognition that between all of us. We have these dispositions or I, I like to think of them as sort of tunings of, of underlying parameters of decision making that influence um what we do in any given situation without like, they're not controlling what we do because they don't have any information about anything, right? They're, they're information free in, in, in the sense that they're not about anything out in the world. They're just a tuning of this individual parameter, like, you know, how risk averse you are. It could be just something that's sort of stable between people. And that makes sense because we know that's stably different between species, right? Some species are very risk averse and very skittish, um, others are, um, you know, braver in a sense, right? Um, SO the idea that there can be biological differences in those parameters isn't, um, anything, it's not saying anything, you know, controversial. It's basically saying, look, humans have some human nature generally, but we all have variations on that nature and Those are the things that we call these, the psychological personality words, right? Um, THE question then is like how do you study them and how do you put some, you can't go around studying 8000, uh, dimensions of, of personality between. Between your subjects

03:42 Ricardo Lopes: also because, also because some of those words might be used as synonyms.

03:46 Kevin Mitchell: Yeah, well, that's exactly the thing is that many of those words are synonyms. So, you know, if you say that someone is shy or withdrawn or um well there's a bunch of other words uh for that, then. All of those things are clearly tapping into the same underlying idea, right? The same underlying parameters. And so what people did actually in personality psychology was, was cluster analysis. So they took all those words and they, and they saw which ones of them are really just um synonyms or are in some way related to each other or correlated with each other. And that was the way that they tried to actually boil down those, you know, 8000. Different terms into some set of dimensions where they could say actually underlying all of this stuff, there's variation along these two major dimensions or 5 or 6 or 16, uh, you know, there's a bunch of different um schemes that people have come up with for major dimensions of personality difference.

04:52 Ricardo Lopes: Yeah, uh, but by the way, when it comes to how we uncover scientifically those personality traits, um, when you say that, for example, in English, people have around 8000 words to refer to personality traits, I mean, So, uh, the uh, the way that people come, came up with, for example, the big five personality traits, and I have back in 2018, Doctor Louis Goldberg, who was one of the main people behind that. Uh, HE'S unfortunately already passed earlier this year, but, um, yeah, of course, he applied the lexical hypothesis, which is doing exactly what you described. There in terms of cluster analysis and all of that, but, uh, when people arrive at, uh, the big five or the exao or other ways of, uh, grouping different traits that people might vary on, uh, is that supposed to ever be an exhaustive list or is just simply what comes up when people, uh, cluster. Those traits in particular statistical ways.

06:10 Kevin Mitchell: Yeah, uh, more of the latter, I think. I don't think anybody would claim that they're, that those things are, are exhaustive, either that they exhaust all the dimensions of variation, or that if you know, you know, someone's scores on those, say, the big five or hexao or something, that you have like total predictive power over what they're going to do, right? They're, so they're very much partial, um. Statistical constructs that capture some trends about the ways that people behave. And, you know, one way to think about that is to is to actually Ask people how they, how they tend to behave in different scenarios and then look at, look at the answers to these various questions. I mean this is basically how personality scores are generated, right? You're just literally asking people how do you behave, but um. You know, you could ask people like, do you like to travel, uh, do you like to socialize? Do you like to, uh, do you like to have sex? Like how much do you like to have sex, those kinds of things, and you would think, um, that maybe those things would be independent of each other, but they're not, they're correlated with each other. So people who, who like to travel also tend to like to socialize, right. And so then you can do some statistics and basically there's sort of a couple ways of looking at that. One could be like that those different pairs all have some correlations between them because they share some common causation. And the other would be is like if you see a bunch of those that have those correlations that may be underlying all of them is one shared construct. There's one shared thing which in this case we would call extraversion. And when you do that, like it gives you a statistical construct, this latent construct, which is contributing to something of the variation in all those um behaviors, but it doesn't explain it fully, right? So extraversion contributes something to how much you'd like to travel, statistically speaking, but there's also a little bit of extra variation left over, right? So for me, one of the interesting um questions is what do those constructs represent ontologically? Like you can say it's a statistical measure and it's a way of modeling things, that's great. But does it actually point to a singular thing like in the brain that's, that, that can be tuned up or down, which is the extraversion dial, right? Or is there an extraversion circuit in the brain that is just fires more in some people versus others. Or is it still quite diverse uh set of underlying parameters that all, you know, can contribute to this domain of, of behaving that we recognize as more or less extroverted where it's a little bit of us imposing, um, trying to impose order on something that's quite messy and in doing that reifying the construct. By thinking I've imposed order with this statistical construct, therefore that's a thing, and the reason I'm saying this is that you people have looked in the brain, right? They've looked in the brain for circuits that vary with how extroverted somebody is or conscientious or any of these other traits. And they very consistently failed to find anything. Basically, there aren't any brain markers of those personality traits.

09:31 Ricardo Lopes: Uh, CAN I just ask you, when you refer to ontology, are you alluding in any way to that distinction that some cognitive scientists and philosophers of cognitive science make between substance ontology and process ontology in the sense of thinking about personality traits as being sort of fixed or being, or uh or varying over time in terms of, uh, I don't know, their. DEGREE, I mean, maybe someone in a particular context is more extroverted, less extroverted when their life circumstances change, maybe they are more prone to depression, less prone to depression, or more or less neurotic.

10:18 Kevin Mitchell: Yeah, I, I wasn't referring to that, but those are all really important points. I was referring more generally to the idea of ontology as the, you know, the study of what is, what exists, and the idea that when we When we generate a statistical measure like an extroversion score or a neuroticism score that we then infer that extraversion, that is a thing like that exists in in the brain somewhere. Um, BUT you're, you know, you make the wider point here that's really important, which, which is that, um, people's personality traits are just these underlying dispositions and. Their actual behavior in the world, first of all is going to be dependent on the situation that they find themselves in as the person that they are in that situation, um, and of course, you know, people's actual, you know, patterns of behavior can change if they've had, I mean, grief is a is a good example where, you know, if you've lost a loved one. Um, IT may be for a long time afterwards, uh, you might not be as extroverted as you were before, right? You know, that's, that's a very typical kind of a thing. Um, BUT the other aspect of that is that, that I find really interesting is that, um, The way that we behave in any moment is a reflection not just of these underlying dispositions and the current situation, but of everything we've learned along the way, right? It's this very constructive process, and that's where the process ontology thing, I think, becomes interesting because I would see our character. As a reflection of these underlying dispositions and all of the experiences that we've had through our lives and all of the choices that we've made and all of the ways that we've built up our, our habits and our policies and our knowledge of the world and just constructed our identity basically through our actions, through our lives, um, which is a, you know, it's a much more agential way of looking at, um, How we come to be the person that we are and how we express that identity in our behavior in the world.

12:29 Ricardo Lopes: Mhm. Yes, uh, uh, I mean, uh, I, I, I asked you that because since around 2023, I've been having talks with people who do work in, uh, dynamical systems theory, for example, I mean people like uh Inezi Palitu and uh Naomi Reuter. And I mean, everyone that, uh, basically learns cognitive science in the Netherlands is a very interesting, uh, subject because they actually learn dynamical systems theory that is non-representationalist, non-computationalist approaches to cognitive science. And psychology which is very different from what we have, for example, as a bases here in Portugal, I, I would imagine also in Ireland so because that, that really brings a completely new way of approaching and understanding uh. A personality traits and psychological traits more broadly as not really something that we are predisposed to and so it's fixed in time and space, but it's something that is always in a dynamical relationship with, uh, I don't know, the rest of our body, our environment, our culture, and so on.

13:57 Kevin Mitchell: Yeah, no, I absolutely think that that's right and the um. So if you're trying to understand anyone's actual behavior in the world, right, you have to think of all those things. You have to think of their history, you have to think of their, their motivations, their current states, what they know, what they think they know, um, what situation the world is presenting, what their goals are, um, and their dispositions. And their dispositions, I would draw a distinction with their predispositions, right? I mean, I, I would say we're born with some predispositions. That doesn't mean Those are our dispositions at every moment of, of our lives, right? The dispositions are a little bit more fluctuating, can change over time, can change dynamically, um, within short time frames, but, you know, underpinning that, there may still be some of these predispositions that are more stable tunings, right? But it's just that they don't, they don't control us in every moment. And I think, you know, the dynamical systems view for me is just the right way of thinking about biology, generally speaking. Like if you're missing, if you're missing the dynamics, you're missing the liveliness of life, the interactiveness of life, um, and the situatedness of any organism in its environment. And those, those dialectics are what life is, right? Is that those relations are, um, How life expresses itself. And actually for me, you know, the dynamical systems theory is, is really key, um, control theory and cybernetics, I think are really interesting too, especially when you're talking about personality where, you know, You could say, so I, I said I, I, I give, you know, lectures on personality and genetics, but I set my students a little challenge at the start of it. I say, OK, so you have to design a robot, it has to get around in the world. Um, WHAT are you gonna equip it with, right? And they quickly get to the idea of, well, it, it needs some sensors, uh, it needs some motors, and then, uh, maybe it needs to monitor its own internal states. Maybe it's, you know, it's, it's low on fuel or whatever, but it maybe there's a few internal. States where it might have to make a choice between certain things like here, yeah, I need fuel, but there's danger or there's a, an attractive robot mate potential over there, whatever it is, right? Um, AND then, you know, they, they, they sort of come to the idea, well, maybe then it needs a control system to adjudicate between these different things in any scenario, um, to say, OK, uh, I'm gonna weight these things a certain way and make a choice between a possible threat and, and an opportunity. But the next step is to realize that I, like, you need to tell it what the relative weights should be or it needs to learn. Um, AND I might tune my robot with a slightly different weights to yours. Like I might tune my, my novelty seeking dial a little higher and my risk aversion dial a little lower, and, you know, whatever my reward sensitivity a little higher. And, and so, once you think about those, you realize like there's no right answer. To that. There's no set of tunings that's going to be the best for every scenario, and that I think reflects, you know, the diversity across human beings and actually not just human beings, like every animal species has something we could call personality, some variation on the species typical nature. And so if you think about that in In control system terms, it's like, yeah, I'm I'm gonna need some underlying um parameters that feed into this control system that are going to allow me to um determine my own behavior and govern my own behavior in over complex scenarios and multiple time frames and so on.

17:41 Ricardo Lopes: No, it's interesting that you also referred to, to other animal models because uh in my conversations with primatologists on the show, uh, I actually got to learn that even when thinking about, so let's say we're talking about chimpanzees. There were, I remember perfectly well that there was this one interview where I asked the primatologist, so, Um, in terms of hierarchy, how do chimpanzee societies function, and the way he answered was like, OK, what chimpanzee society are you referring to? Are you referring to chimpanzee societies in Gombe and then he mentioned other sites because The ecological circumstances vary, the way that they structure their societies also varies. So I mean, even when it comes to other animals and until at least very recently, and unfortunately some people still use those terms, terms like Um, instinct, I mean, I guess that fixed action pattern is more, uh, rigorous scientifically, but even when people think about something that is part of the evolved behavioral repertoire of a particular species, they think that it's just a fixed and that. Particular individuals of a particular species, regardless of the ecological circumstances they inhabit, they will respond exactly the same way to similar stimuli, but that's just not the case. It's

19:21 Kevin Mitchell: not, and, and I think two things can be true, right? There can be a species typical nature. Maybe this is gonna be 3 things. Um, THERE can be, in fact, there, there, there must be variation around that nature in the same way that there must be variation around the physical nature of any species, right? All these chimps don't look the same. Uh, BECAUSE there's genetic variation that affects how bits of their bodies grow. Well, there's genetic variation that affects how bits of their brain grow as well, and, and it's basically inescapable. There's no way that there couldn't be some variation around that. So if there's any sense in which the nature of a species is encoded in its genome in the and in the broadest terms, then we have to expect that there should be individual variation around that. But at the same time, We also have to have to realize that there's this embeddedness of the species in a context and that is where these underlying this underlying nature will be expressed. And so, you know, the challenge for People studying personality is, you know, to not just throw their hands up in the air and say, look, this is all too complex. It's always context dependent. Um, I can't get a grip on anything, right? And so what they're doing, what they're trying to do is literally just get rid of the context or look for patterns that apply across context in a statistical sense. And I think sometimes what people think they're trying to do. Is explain every bit of every behavior all the time, and they're not, right? They're they're just abstracting away from the context dependence and all of the details of any given situation to see are there underlying patterns of variation, and the answer is there are, and those can be studied without it, without anyone involved in that enterprise claiming that they explain everything about behavior all the time.

21:19 Ricardo Lopes: So, uh, let's talk a little bit about how we study genetic predispositions and of course there can be other disciplines and other scientific tools, but, uh, talking first about behavioral genetics and the gold standard in behavioral genetics tends to be. Uh, THE study of, uh, twins reared apart, monozygotic twins, zygotic twins, and sometimes people also study, uh, adoptees, uh, and then, of course, compare, uh, twins, um. Monozygotic dizygotic twins and uh adoptees, but what can we learn about genetic predisposition from behavioral genetics? I mean, uh, what do you think are its limitations maybe?

22:11 Kevin Mitchell: Yeah, um, well, there's lots of limitations. Um, LET me start, I guess, with the, with the basic observation is that you have some trait that varies across the population. So if I measure extroversion, I give people these questionnaires, um, I'll see some people are high on extroversion and some people are low. And I have a, I have a kind of a pseudo quantitative score that I'm giving people, right? Which is just a, it's just a statistical proxy for something that we recognize in our real lives, um, day to day anyway, right? OK, so I've got variation across, uh, across the population. And like any, like most traits, um, that's going to be a normal distribution, right? There's gonna be a lot of people who are sort of in the middle, and you've got some really, really shy people and some really, really outgoing people at the extremes. And then the, the question is, what causes that variation? What are the sources of variation in that trait? And of course, the, the, the sort of the age old idea that, well, maybe you were just born that way because of your genetics, or maybe you are that way because you were raised that way. So the classic nature versus nurture dichotomy. And one way to tease that apart, I mean, the problem is you can't just look at at similarities between relatives and answer the question, right? If you see that relatives are more similar to each other than non-relatives, which is true. Then you don't know the answer to the question whether it was nature or nurture, and that's where twin studies, family studies, um, adoption studies, and so on have been so useful, but they are absolutely limited in lots of ways. But the basic idea is that You can tease out the contributions to the variance in a trait. Without um saying that you can disentangle the causal contributions to to an individual person's the way that they actually are, right? So you can't for me disentangle the effects of genes and environment or experience in the way that I am right now, but you can across a population say how much does variation in genes contribute to variation in the trait, and it's a, it's a different thing. Um, AND I, it's really important to, to emphasize that what behavioral genetics is interested in is explaining differences and, and variants, not explaining everything about the way that somebody is or, or, you know, everything about their phenotype in a productive way. It's a, it's a difference making kind of causation that we're after. So when you do that, you can do these various designs and say, OK, look, if it was nurture that was making people the way that they are, then anyone raised in the same family should be more similar to each other. And when you look for personality traits and you look at adoptive siblings. They're almost not, no more like each other than any two strangers on the street. Like it really just that observation by itself suggests that being raised in the same family does not in fact make you more similar to each other. And actually, I mean, I should, what I've done here is I've gone from talking about variation across the population to similarity, which is the flip side, right? So what we're doing whenever we're doing behavioral genetics is actually studying similarity between people and what makes people more similar to each other, but it's, it's, it's the inverse of a source of variation, right? So you can also look at twins who are reared apart and like you said, this is the the the sort of gold standard is you take identical twins, monozygotic twins who share the same genome, but who've been adopted separately and raised separately in different families, and then you can ask, well, Uh, you know, a family environment is important in generally in making them similar to each other when they're raised in the same family, then they should be less similar to each other when they're raised in different families, and the answer is they're not, they're not much less similar actually, they're quite still quite similar now. You know, some of the limitations of those particular experiments is just the numbers, right? It's hard to find twins, monozygotic twins who are reared apart. That's not the the

26:23 Ricardo Lopes: sample the sample is usually low,

26:27 Kevin Mitchell: right? Yeah, it's low and it's usually reflects some particular. Um, HISTORICAL circumstances, sometimes, I mean, you know, after the Second World War, for example, the number of children had to be, you know, were adopted and, and, you know, shipped off to different families and so on, but thankfully that's a rare, um, kind of condition. So when those when we do those studies, those are a couple of the designs, the adoptive designs, you know, twins reared apart, um. Sorry, I should have, I should have said that actually all of this, what we're using or what we're interested in is the correlation in the phenotype between pairs of, of individuals who have different degrees of relatedness or who have been reared in the same family or not. And by doing that you can dissociate these different sources of, of variation. Um, AND the other kind of classic design is to take identical twins who are obviously born at the same time and reared in the same family, and then see, well, how similar are many, many pairs of identical twins versus um fraternal twins, right? So dizygotic twins who, who are basically just like siblings, right? They share 50% of their genes. But who are also born at the same time and raised in the same family, so you would say, OK, this is as close as we can get to an ideal control comparison. And then, um, you know, the idea is if the monozygotic twins are much more similar to each other than the dizygotic twins, that shows really a genetic influence. So you can do all those kinds of studies and then um you can put some numbers on this and say, here's how much of the variance that we see in this trait across the population that we can attribute to genetic differences versus something else. And the bit, the proportion that you can contribute to genetic differences is what we call the heritability. And there's a whole bunch of, of kind of mathematical assumptions that go into actually generating a number. Um, MY own feeling is that the exact number doesn't really matter. It's not like a biological constant. Um, YOU might find that, you know, extroversion in your population is 40% heritable, and that means 40% of the variants that we see across the whole population is attributable to genetic differences, which means the rest is something else. But it could be, you know, in your study, it might be 40%, in my study, it might be 50% or 30%. It doesn't really matter. I tend to think of them as in sort of in buckets of like low, moderate, high.

29:05 Ricardo Lopes: Mm, yeah, you know, it's funny because er in Portuguese the common words that we use to refer to monozygotic and dizygotic twins is true twins and false twins, so I mean the dizygotic is, it's like, oh they're not really twins, they're like just normal siblings.

29:26 Kevin Mitchell: They're just like normal siblings who happen to share a womb, yeah.

29:29 Ricardo Lopes: Yeah, yeah, yeah. So, uh, let me ask you about heritability because this, I think that this is a concept that many people have misconceptions about and I, I, I, it's, it's not for, uh, it's for good reasons because I, I mean it's. The concept that it's really hard to grasp for most people, but what does it mean, for example, for a trait to be 70% heritable? I mean, does that mean that it is 70% genetic?

30:03 Kevin Mitchell: It means that again, heritability just applies to variance, and variance is. It's something that most people are not, don't, don't think about as a mathematical defined thing, right? But you can measure the variance of something across the population, which is just, you know, you say, take the height of people and you can say, well, um, the variance in height across the whole population is this much and it's a number, right? Um, WHEREAS if you take a height of just NBA basketball players, the variance in height is going to be less, right? Cause you've just sampled this bit of the population, so the bell curve is going to be narrower. And if you took just monozygotic twins, it would be really narrow for height, right? There would be very little variance left. And that kind of gives this intuition, right? So we have the whole population, here's the, here's the spread of values of, of height. Now, imagine everybody was a clone. Uh, LIKE the, not just pairs of twins, but everybody was genetically identical, then the distribution would be like, come, you know, very narrow. So, um. In a sense, the heritability reflects how much narrower the distribution would be if everybody was genetically identical. THAT makes sense. I haven't said it that way out loud before, but that in my mind that that's a sort of an intuitive way to, to think about it is like, OK, because we're genetically different from each other, because we're not clones, then the distribution of this trait spreads out. And that's true for physical traits and it's true for um cognitive behavioral traits like personality measures and so on. So if something is 70% heritable, that means that 70% of the variants. In the trade that you see the spread of values in the trade that you see in the population that you're studying. Is attributable to genetic differences in the population that you're studying. Right and It doesn't mean that 70% of a person's extroversion comes from their genes and the other 30% from something else, right? That's not, it just, it doesn't apply to an individual. It just applies to a population because it's about how people vary across the population.

32:16 Ricardo Lopes: Yeah, but this is really a very common misconception, right, because it's not that uncommon for, for you to read even sometimes in, uh, I don't know, uh, just popular uh publications that sometimes, uh, divulge a little bit of science, do a little bit of science communication that, oh, a new twin study found out. That, uh, I don't know, personality traits are 50% genetic, and they put it in those terms, but that's not really the correct way of thinking about it.

32:55 Kevin Mitchell: Yeah, yeah, and the problem is like the word heritable, um, you know, it sounds like, uh, it, it sounds like a colloquial term, but it's a super, super technical term, and actually even within that there's multiple types, there's different. There's narrow sense heritability, there's broad sense, like it's a very technical term from quantitative genetics and really, you know, it came from animal breeding and in animal breeding, like it matters what the number is, because you might have a trait in a, in a population of cattle or something like that. You say you measure milk yield in a population of cattle, and then you want to know it's like I've got some here that are really give really good milk yield, or if it's bulls, like their daughters give really good milk yield. Um, HOW much can I increase the milk yield of the population in the next generation if I just breed from those, uh, those cows and bulls at the extreme, right? And, and you can put a number on that. You can get over how much is the population going to respond to selection. And if the trade is highly heritable, then that means it's worth doing this breeding to push the population mean. And that's how we have done, like that's what we've done when we've domesticated animals and we have pushed them by breeding, um, and the same for plants. So, There the the actual value becomes important, like whether it's 4% or 40% is going to predict how much of a gain you're going to get in the next generation in this trait that you're interested in. Um, SO that's a, a different, more operationalized perspective on what heritability is as a tool that people use, like a mathematical tool to say, yeah, uh, this much of the variation in the trait that I see is due to genetics and therefore I can breed for that trait, and this is the response that I, I can expect. Whereas if none of it was due to genetics, then it wouldn't matter which individuals you bred from, they'd all, the next generation would all be the same.

34:53 Ricardo Lopes: So, uh, still on the topic of heritability, let's tackle what might be another common misconception. So, uh, if a trait is highly heritable, what kind of implications does that have? I mean, does it necessarily mean that it is less permeable to environmental inputs, to environmental influences? Does it mean that it is harder to change or not,

35:21 Kevin Mitchell: um, not, it doesn't necessarily mean any of those things. So, again, it's very population specific and you might find in one, in one population that something like height is actually very highly heritable, like 80 to 90% heritable. Most of the variation we see is due to genetic differences. And yet, If you compare across populations, there's mean differences in average height. And they actually track with um like protein intake in the first few years of life. So it's really infant nutrition. And, and so you can make sense of those two observations. It seems like they're conflicting, right? One is saying, look, this is mostly genetic in origin, the variation, and the other is like there's clearly an environmental source of variation that makes a big difference. And you can, you can reconcile those by, by realizing what it is you did when you did your study on the heritabilities. You took one population that just may not have a lot of environmental variation in it. And then you ask, given that set of circumstances, what's the proportion of variation that I see this due to genetic differences. Now, if you expanded that over many, many countries and you did the same study at the same time, then what you would find is that the heritability would go down. So it's not a fixed number, it's a proportion. And it can go down as other sources of variation increase. So if there's more environmental differences, your heritability will be lower just by definition. It doesn't mean whatever those biological influences don't exist anymore. It's just you've added a bunch of other ones, right? So when we do, um, you know, twin studies or family studies and so on, we're almost always just Trying to control for all that other variation. We're not doing it worldwide, right? We're doing it in like the Minnesota twin study or the, you know, the Scotland twin study or something. They're very local defined populations and, and so that can give a misleading view that In a population, just because there isn't a lot of environmental variation affecting the trait, it can lead you to infer that environmental variation can't affect the trait, and that's just not correct. It's a very much a contingent kind of a measure specific to the study that you've done. And then the other, the other question is like, is it um. Is the eventual trait amenable to any kind of um non-genetic or environmental factors or or so on, and there I think the answer varies by trait. It's not necessarily true that just because something is 50% heritable, that it's also, uh, you know, amenable to, to variation by any sort of systematic factors in the environment that may or may not be true. Um, IN many cases it, it is, I mean if you think about something like intelligence, um, like intelligence as the ability to think well is clearly trainable, right? I mean, that's why we do education. So the idea that there's just this fixed biological innate IQ score that is then not um uh affected by any of your experiences is just silly in my mind because of, of course it is, right? That's why we have schooling. So, um, for personality traits, it's really interesting. I mean, it's kind of an open question whether in fact there are any kind of systematic factors that vary in the environment that are correlated with, say, making people more neurotic or more extroverted or more conscientious, and of course people have done loads and loads of work trying to find those and, you know, one of the Inferences from um behavioral genetics is that actually these traits are only partly heritable and that already leads people to think, well, the rest must be environmental, right? It must be something outside the organism. And maybe we can find what those factors are. And so people have looked for, for any kind of systematic factors, you know, things like whether your parents got divorced when you were young, like does that have a systematic effect on people's personality traits. So you just take a bunch of people whose parents did get divorced and a bunch of people who didn't, and you ask if their scores on, say, neuroticism are higher or lower. And basically they've not found anything that's really systematic, there are just aren't any clear sort of life experiences, maybe like with the, the exception of really like severe neglect or abuse, which can obviously have very, very long lasting deep psychological effects, but um which thankfully is, is fairly rare and um so in the, in the more general realm of, of, of experiences that aren't as extreme as that. There don't seem to be any good correlations. I mean, birth order was a is a a famous one that people have thought, you know, was tied to different personality traits, and it just, it just isn't. Um, YOU know, those, those effects have just kind of gone away as people have done larger and larger studies. So, yeah, for personality traits, it's interesting that Yeah, they're only partly heritable. Um, EVEN if they were fully heritable, that wouldn't mean in principle that they couldn't be malleable or subject to experience. And yet when you look at them, in fact, for those traits, it doesn't seem to be good um evidence that the, that the really underlying biological disposition, disposition. Is affected by experience. Of course, how it's how it's expressed is of course affected by experience, right? That's how we come to. Adapt ourselves to the world and accommodate to to experience and so on. So it's um, yeah, it's subtle because there's different. There isn't just one simple answer to to the question, it just varies by different traits, I think.

41:25 Ricardo Lopes: Uh, WHAT would you say are the most rigorous scientific ways of studying environmental influences on, uh, personality traits or other kinds of, uh, psychological traits?

41:41 Kevin Mitchell: Um, WELL, I should preface this by saying I'm not like I'm not a personality psychologist and I'm not a sociologist, right? So, so those, those people could answer that question better, um, better than I could. But, you know, a simple design like what I just talked about is, is just. Looking at uh groups of people who had exposure to one type of experience versus another or who've been brought up in one kind of environment versus another and seeing if there's systematic differences in, in how they behave or their scores on personality traits. Um, THERE may be systematic differences in how they actually behave without there being differences in their underlying personality traits, so. You know, the challenge when working with, um, human beings obviously is that we can't do interventions, right? We can't do controlled interventions in in the way that we can do in in animals, um. So, you know, if we were thinking, you know, a really rigorous way to do this, if we accept that animals also have personality differences and in fact that sometimes the dimensions of their personality differences. Map onto the dimensions of human personality differences, then, um, then thinking about, you know, doing work in animal models is another angle that you can take in there you can control the environment, you can control the um The experiences that animals have. Yeah, Again, um, I think the answer sort of is kind of the same is that there really are innate differences between individuals. Um, THOSE underlying differences are pretty refractory to change, and yet the way that they're expressed, the thing that we're actually interested in, is very much subject to experience and um. I is, is, you know, an expression of the actualities of the experience of an individual as opposed to the sort of cloud of potentialities that that we could think of.

43:37 Ricardo Lopes: So when it comes to the environment and how we should think about, uh, I mean, the role of nature and nurture in shaping our psychology, another concept coming from behavioral genetics is the concept of gene-environment correlation, and we have passive gene-environment correlations, active gene-environment correlations, and reactive gene-environment correlation. So, I mean, It, uh, what people might think is being environmental influences might just be, for example, uh, your parents sharing some of the the same genes with you and having, uh, similar personality traits or people just, uh, treating you differently because of your own personality traits or for example, you because you have a particular kinds of Interests seek actively seeking environments where you can fulfill those interests. I don't know if I love reading, maybe I spend more time at the library or something like that or have friends who are also intellectuals or something like that and that further reinforces my, my personality traits, my interests. So, but, but I mean, when can. We say, or, or even do you think that it is a proper question to ask uh whether we can tell when a personnel or when a personality or when the environment changes a trait or when it reinforces it? I mean, do you think that that's a proper question to ask or

45:18 Kevin Mitchell: I think it's a reasonable question to ask, um, again. You can think about it in terms of sources of variance, right? And, and what I talked about earlier was this idea that we can separate the, the genetic sources of variance from environmental sources. And of course, that's, that's one of the assumptions underlying the calculation of an actual number in, in like twin studies and so on, is that there isn't any interaction between those two terms, that they're independent from each other and they're just acting separately. And of course, it may well be that under, you know, people could vary. In a way where, for example, they're vulnerable to some kind of environmental stressors, where um, in the absence of those stressors, these people would end up behaving just exactly the same as each other, but when those stressors appear, some of those people um will show a phenotype and and and others won't, and you know this comes into play with things like um psychiatric. Disorders or mental, uh, mental illness where, um, I mean, we saw this during COVID, for example, where, you know, the circumstances were so brutal during, during lockdown and so on where people were really isolated, their support systems were gone, that many people actually developed mental illness in a way that they would not have done otherwise. So if you were just, if you were just drawing back from that and, and looking at Mental illness as a phenotype and saying, you know, is there a genetic effect or is there an environmental effect? Clearly in that case, there's a, there's an interaction between those two. So with personality traits, it's perfectly reasonable to suppose that there might be an interaction between those two things, um, broadly speaking, but then the question is, how do you actually study it? It turns out to be really hard. To study in, uh, in a global sense because you just, you, you just don't know how to model it. You almost have to say, you almost have to put the terms into your model mathematically in order to get them out. If you don't put them in, you won't get it, you won't get them out basically. Or you have to do very specific experiments. They're not, they're not global experiments. They're much more local saying, OK, I'm looking at the interaction between neuroticism as a personality trait and Um, some kind of life event that, yeah, like divorce, say your parents got divorced, um, and then asking, OK, do kids who are more neurotic, um, you know, have more negative consequences when their parents get divorced than kids who are not neurotic, that kind of thing, right? So that's a much more local design that you can, um, that you can do, but. You know, more generally, I think the. You, you also touched on, so here, when I was talking just then, it, it was like framing things as experiences as things that have happened to you, where you're not actively involved in that, it's just you're, you're passive, and those are factors from the outside affecting you. And I think that that's true, those, those can exist, but I think it's also true that a lot of our environment and our experiences is crafted by us. Or initially crafted by our parents who may share uh genes with us and genetic predispositions as you said, so, you know, you may have, um, uh, say a child who's maybe anxious, uh, you know, biologically speaking. May have parents who are also share that trait a little bit and they may be a bit overbearing or overprotective, and then you can get a kind of an amplification in the, in the behavior of the, of the individual, or at least in the relationship between the the child and their parents that reflects that, right? Um, YOU talk about

48:57 Ricardo Lopes: and the, and the thing when it comes to that example, the thing is that when people see a situation like that, they might immediately assume that the child is anxious because the parents are creating an anxious environment, but it's, it might just be that they share some of the same, the same genes that predispose to anxiety. Yeah,

49:19 Kevin Mitchell: exactly, or there's the other factor that you suggested, which is that um. Some of the behavior of the parents is actually evoked by the personality of the child, right? Children have different personalities, and you like to think as a parent it's like, oh, I'm treating everybody exactly the same, and it's like, like hell you are. You're in a relationship with this child and you're reacting, you know, as much as anything else. So, so there's that sense in which the child can evoke, um, some patterns of behavior in their parents or. Siblings or friends or peers or teachers that then can feed back onto their um Onto their own development and you know, the most obvious kind of case of that is a child who is very disruptive for whatever underlying reason. Um, Can can engender negative interactions or interactions that other people perceive negatively that then can feed back onto a pattern of, of disruption, right? Um, SO what, you know, at early stages we've got those two things going on. You've got, you know, there you are as a baby, you're not really in control of anything, but you're in an environment, at least if you're biologically related to your parents, that, um, Some of your underlying genetics may be expressed through them and the environment that they create. Which is, that's a challenge in itself when you look at heritability because actually heritability of something like intelligence actually partly reflects the fact that some of the genes that you're carrying are shared by your parents and they're creating an environment which influences your intellectual development. So there's those complexities are all um can be all kind of woven together. But then like as as individuals get older, they're not just evoking reactions, they're actually creating their own experiences. They're seeking experiences, they're creating environments, they're choosing who to hang out with, they're following their, their interests in, in what they're doing. Like you said with reading, you know, if you look at how well people read, that's a factor of both potentially, you know, some underlying um aspect of cognitive ability. Generally, maybe some ability that's more particular to reading, but then also just practice, right? I mean, that's just a skill that people learn and they're not going to learn it if they don't practice it. And so people who like it will do it more and it just amplifies and same with like sporting uh skill or musical skill or whatever is this interaction between an initial interest, maybe a talent. And the reinforcement that a that a child can get by being good at something and then wanting to practice and so on. What that means is that you can start out with um small differences between individuals that through that experience get amplified. So nurture is now amplifying um these initial differences that you see between individuals and that kind of dynamic. You can see, you know, in, in humans, but you can see that in animals as well, actually, people have studied that in, you know, in, in mice or fish where small differences at, at birth lead to this, you get this magnification across time and I, I guess the, just the final aspect of that that I find really interesting is that that. It's not just an individual thing. It's, it's like uh there's a almost a competitive process there where it's like there's some niches for somebody to fill, um, and if I'm filling them, there's no space for anybody else to be that type of person, you know, um. That was a little, uh, a loose way of saying that, but the idea is that there's the actual behaviors of any individual depends on the behaviors of everybody else in the group, right? It's a very interactive relational sort of thing, and those dynamics emerge over time in a way that's informed by initial differences, but that can actually amplify or crystallize them.

53:13 Ricardo Lopes: You know, I, I'm, I'm always uh fascinated by these topics surrounding whether in a particular case, someone has been influenced by another person or by their environmental circumstances or where those circumstances or that person is just reinforcing underlying traits because I mean just thinking about myself when I was Very small, I mean, I was, I don't know, 566 years old. I mean, my mother always stimulated me to read, but I, but I mean I, I went beyond that because I mean, my mother is a is a teacher and her goal was like, oh, you have to read to be a straight A student, but I'm like, oh no, I don't care about being a straight A student. I, I want to read just because I want to learn new things. I mean, the straight A student for me is just a byproduct of that or something, and then, and then I was like 5 or 6 and also watching her entire collection of VHS tapes with old movies. I mean, I was 5 years old and watching movies about the Roman Empire, which probably I shouldn't have done, but anyway, but uh I mean, just to say that, uh, it's like. In that particular case, of course, probably if my mother didn't have the books and the tapes at home, I mean I probably wouldn't have developed those habits or wouldn't have been exposed to reading, watching movies and so on, but at the same time it er she stimulate. ME, but it was not like, uh, oh my God, this is so boring. I'm going to read like 5 minutes just for my mother to leave me alone, but then I'm going to drop the book and go play football or something like that. No, I actually went beyond what she wanted me to do because I really liked learning from a. Very early age, so I mean, was it really my mother influencing me or was she just reinforcing it or was more of a sort of, oh yeah, I have this environment at my disposal, but I'm going to actively seek those kinds of activities, you know, that

55:34 Kevin Mitchell: I

55:34 Ricardo Lopes: sometimes

55:34 Kevin Mitchell: I

55:34 Ricardo Lopes: think about that. Yeah,

55:35 Kevin Mitchell: I think it's probably, you know, some combination of all of those, but I think that, you know, what's interesting. With that particular example is that, is that you have the opportunity to do that, right? Where, um, and that, and this is, this is a key thing where you have kids who have different interests and different abilities and their personality traits align them more, more or less to certain activities that just sort of suits their personality, right? Um, AND then, you know, we talked about this idea of this amplification and children seeking their own environments and building their own experiences and, um, you know, that's, That's like in an ideal situation almost, right? That's a that's a luxury situation and it's a luxury that a lot of kids don't have in the world is they don't like, it's great for you. You could go and read a book because you didn't have to go work in the mines or uh go to the sweatshop or go out in the, you know, go out in the fields and, and, and whatever, right? So, um. I think that, you know, the idea that that individuals can flourish under certain sort of circumstances where they have the luxury of becoming themselves, right, in that exploring those things um and following their interests because they don't have all these other pressures on them. And that's where, you know, if you look at broader sociological factors and things like poverty and socioeconomic um stresses and conditions and so on. Um, YOU know, those are where you can see the effects of those kinds of things is like not on very particular things, it's just like, well, do kids in general have the opportunity to um follow their interests or is there just so much, there's just so much outside pressures that they're not um able to be, um, you know, sitting in a room and, and, um, reading all these books or whatever, whatever the case may be.

57:22 Ricardo Lopes: Yeah. So, let's talk a little bit about the group differences because I think that it's very important here also, I, I mean, I, I won't, I won't say this is a limitation of behavioral genetics. I think that it's just a misconception that people have about what uh uh what information they can take from behavioral genetics studies because I've talked with behavioral geneticists and all of them told. Me that uh no behavioral genetics is not doesn't have proper tools to study group differences. I mean if people uh pick information from behavioral genetics to uh make claims about I don't know, race differences, uh sex differences, ethnicity differences, whatever, uh they're using it wrong so uh but but but I I I mean but do you think that. There's uh anything at all apart from simply we uh uh knowing from behavioral genetics that there's at least some genetic basis to uh people's psychological traits. Is there anything there that you think can be reasonably at all used by people to make claims about group differences?

58:45 Kevin Mitchell: Well, I mean, let's, if we just think about it in in principle, right? If we think about the the the um the ancestry of any people in the world. What that's going to reflect is this very sort of, you know, a deep ancestry of, of, um, migrations of, of humans, you know, out of Africa, across um across Europe, across Asia, into the Americas, into Oceania. Um, THE various populations of hominins that they interacted with and interbred with, migrations back into Africa, you know, it's really, really complicated sorts of, of sets of, of interactions. And then in more recent times, um, you know, you had these Paleolithic, uh, people across, say, Europe, and then they were largely replaced by migrations of Neolithic farmers in, in the subsequent migrations with some interbreeding. And then, You know, a few 1000 years later, replaced again, or with some hybridizations from, you know, invasions by steppe pastoralists and, and so on with back migrations and all of which is the reason I'm saying all of that is that, The idea that you have this sort of clean branching of different populations, which then just became the race, the continental groups that we think of as classical races that were defined by, you know, Linnaeus in the, in the 70s, 100s as part of his classification of, of different animals and species and so on. Um. Those are just arbitrary groupings from a genetic point of view, like there's much more tangled kind of ancestry. It's not that they just diverged and then didn't see each other for 10,000 years or something like that. Um, THAT'S not the case. So first of all, the, the groupings that people use for the sort of classical races are arbitrary. It's not that they don't reflect anything. It's just that there's no reason to privilege that level because if you look at like genetic. Markers, you can find genetic markers that will distinguish, you know, as a as a whole profile, people from Europe versus people from Asia, for example. Um, BUT you'll also find that that's those differences are continuously changing, right, as you move for it's not like there's big monolithic blocks where everybody in one group is the same and everybody in the other group is the same, and there's big differences between them. It's more of a gradual kind of thing. And that reflects, you know, ancient ancestry, but then of course, uh, each of us has genetic variants that come from more recent ancestry. I mean, any genetic variant in a population arose as a mutation in somebody. Uh, SOME of them arose thousands of years ago, and some of them arose in just the sperm or egg that generated you, right? I mean, that's the, that's the continuum, right? Or some of them are just brand new, and some of them are very ancient and have spread through the population. And then there's other ones like ones that arose a few generations back. They're particular to one population versus another, but really local population. So, you know, you can use, you can use genetic profiling. To look at, you know, you could find um distinctions between the profiles of people from Portugal versus Ireland, and that's what like 23andMe will tell you that information. But in Ireland you could go more local, right? You can find differences between people, you know, from Dublin versus Cork, or, you know, even neighboring counties like in England they had a whole map that all of the little counties, so. All our any individual person's genetic profile reflects this. Ancient vari variants that they inherited, um, much more, you know, more recent ones from, from, from local, um, populations and then very, very recent ones that, um, arise. So all of that is to say, But if you're thinking about group differences, the first thing you have to think about is like, how am I defining my groups? And are they actually natural kinds, or am I making some arbitrary cuts that I could, I could cut it at a finer level if I wanted to, right? So, once we say that, OK, that's the sort of a, a background to the just getting clear in our mind, what's the question that we're asking. Then you can ask or say I've got two populations, and within a population I find that some trait is quite heritable, and within the other population, it's also quite heritable. So there's some variation in each of these, um, populations and a lot of the variation is due to genetic differences. And then say I find a difference in the mean. Between those populations, it feels like a really natural inference to say, well, I've shown that this trait is genetic in these populations, there's a difference between them, that must be genetic too, right? And that's the fallacy that people um that people make. And if you look at something like body mass index, right, within any population, body mass index is really highly heritable. And I mean even surprisingly if you look at like adoptive siblings. They're not at all correlated in body mass index, right? You would think that growing up in the same family, you'd have the same diet, you'd you'd have the same tendencies, and there's just no correlation in body mass index between adoptive siblings. So clearly there's big genetic influences, right? But at the same time, Between countries, there's massive differences in mean body mass index. And those are environmental and cultural, right? They're due to um lifestyle differences, social differences, and so on. The like nobody looks at those and thinks, oh then those must be genetic because we know, right? It's obvious that they're um cultural and there's good evidence that that's the case. When So when people look at something like intelligence, where of course this has been done, right, these claims have been made, um, my first inclination is to say, well, look, the actual performance on IQ tests is a product of underlying biological ability, but also all of the opportunities that people have had, all of the schooling, um, you know, availability of education. But also, you know, social factors that are pressures that, that contribute to whether people actually have the luxury of, of engaging in education. And, and then there's other factors like infant nutrition and maternal nutrition and, and so on. So, there's a whole load of social factors that we know, um, within any, within any country contribute to those kinds of, um, Scores, right? And, and so we know that those things vary between countries and even between subpopulations or socioeconomic, um, sort of divisions within countries. So, um, you know, hypothetically speaking, It could be, that could be due to genetic differences. There's just no evidence that it actually is due to genetic differences. Um, AND especially when we know that cultural factors can have such a big influence on it, it seems like that's the more parsimonious explanation.

1:05:43 Ricardo Lopes: Yeah, no, uh, and I think that, uh, I mean, it's very important to, to keep in mind that, uh, I, I mean, of course these kinds of conversations, um, at least in the minds of certain people very quickly, um, get into a very nasty political things like racism and eugenics and white supremacy. And so on, because it, it really bothers me quite a bit and not only, I mean, even if we put just the political and moral aspect of it aside for a second, even from a scientific perspective, why is it that certain people are so insistent in claiming that because supposedly in the US you find an average of one standard deviation difference in terms of average IQ between black people and white people, why they make the claim that that must be genetic. I mean there's no evidence, there's no solid evidence of that and even more so if you take into account the fact, uh, like socioeconomic factors. That distinguish uh black people, uh, the majority of black people from the majority of white people and so on and so forth. And then, for example, uh, when I hear claims like, uh, oh, black people are more predisposed to developing sickle cell anemia to protect them from malaria. I, I mean that's also. Actually not true. It's very specific populations in Africa or descendants of very specific populations in Africa. I can't remember exactly that's in sub-Saharan the location, yes, sub-Saharan Africa that actually might have a predisposition to developing sickle cell anemia. So I, I mean when people talk about. Uh, BLACK people, white people, East Asian people, it seems to be like a, seems to me like a very crude scientific analysis

1:07:58 Kevin Mitchell: of. No, absolutely, so there's, there's just multiple sort of, um, issues all tied up there, right? So the first is that just these racial groups, these classifications that we use, um, are just really, really arbitrary and crude, and the idea that you would say as a class. White Europeans are just this monolithic group, and they're all the same, uh, for the purposes of, you know, what we're interested in, and they're all like this big differences with this other group, and they're all the same. It's just a nonsense, right? And that's what I was talking about earlier, when there's like, you can just go down the, the, the, the resolution of genetic differences and profiles between people and groups of people, and you'll just keep finding differences, right? I mean, in any one of us, you can do 23andMe. Genetic profile and get it to tell you which major ancestry groups you're related to, but you can also do a paternity test with it. I mean, it's right, that's how fine it is, right? So, um, so there's that, uh, and then, yeah, there's also just sort of ignoring socioeconomic conditions where, you know, it doesn't have to be like in the, in the US those socioeconomic conditions because of the history of the country. ARE often correlated with race, right? In other countries, they're not correlated with race, right? They, in Ireland, they're not correlated with race, um, because we haven't had that kind of history, but we still have socioeconomic divisions between areas that are more or less socially well off. Um, AND you see differences in, in terms of school performance and how many students go to college and all of those kinds of Um, all of those kinds of measures that are readily explained by, um, access to education and, um, and other sort of social factors that influence, you know, whether people are just more inclined to go get a job or whatever, right? So, yeah, and I think, you know, when you're asking why do people make these claims. Um, I think it's a very obvious answer to why people are inclined to make those claims is because they fit with the view that they already have in their mind, which is a view of racial differences that, and not just of differences, but of, of, uh, some being better than others, right, of superiority, and then they can, they can look to some scientific, um, Results or findings that they think support that view, and so there's a sort of a motivated reasoning and a kind of a confirmation bias, but also just a misunderstanding of what those results actually mean and what they don't mean. But it's very, it's very challenging, you know, as a geneticist to Uh, talk publicly about this stuff and, and try and make sure that it's not misunderstood, um, and that people aren't getting the wrong end of the, of, of the stick, and, you know, because the, the problem is, like as we've been discussing all of this stuff through our whole conversation, it's like, it's nuanced, it's contextual, it's interactive, it's not, and those are hard stories to tell. Um, IT'S much easier to say, no, it's all due to this one thing or that one thing. Um, AND so that's the challenge is to, is to try and um. Talk about these things in a way where you're getting conveying some of the complexities without obviously, you know, overwhelming. People with them. Um, YEAH, but avoiding or actually I think anticipating or preempting the, the misunderstandings that can come.

1:11:24 Ricardo Lopes: Yeah. So, let me talk about another, uh, let's talk about another factor here that might play a role in, uh, development and psychology. Uh, WHERE do you think we should place randomness in relation to nature and nurture? I mean, do you think that We should consider it to be part of either nature or nurture or should it be thought of as a as a third factor in, in the, in development and psychology. Yeah,

1:11:58 Kevin Mitchell: so this is, um, as you know, one of my favorite, um, topics to think about and, and, and yes, I do think it is a third factor. So when we think about. You know, I talked about earlier, we'd say some trait is 70% heritable, and, and you would say, OK, 70% is due to genes, that means genetic variation, then the other 30% must be environment, and people just have that kind of dichotomy. It's like either genes or environment and so there's a, a, a, a sort of an easy inference that the limits of heritability that behavioral genetics has shown is actually proof that environmental factors are affecting these things like say personality traits. And by environmental, I mean, in the typical understanding of that word, which is something exogenous, something outside the organism itself, right? That's a normal colloquial understanding of the word. It, it unfortunately, there's a statistical um definition of what's called the non-shared environment, which is like not family environment, but anything else. So you can kind of partition. Effects of, of shared genes, of a shared family, and then you've got something left over, and that's called the non-shared environment, which again makes it sound like it must be something outside the organism, but it isn't necessarily. It could be something endogenous inside the organism. And really this just it it it comes from the idea that the genome is not a blueprint for an organism. It doesn't encode everything about the final product. It really encodes kind of the settings of rules for development, and, but it encodes them rather loosely, right? Because it can't control once it starts making some proteins of different levels and different different cells. Those proteins are just going to jitter around in the cell, they're going to diffuse around and bounce into each other and bounce off and and and like at the molecular level, it's incredibly noisy. So the genome that can't control that. All it can do is say, OK, I'm going to make this much of protein A and this much of protein B and this much of protein C, and let the self-organizing dynamics of these cells make a limb and let these ones make a heart or a brain or whatever it is, right? So, um, when we look at the developmental outcome from clones, right, in animals. Um, WE see loads of variation between individual clones. We see that in the, in, in the anatomy, we see it in neuroanatomy, we see it in the connections of individual cells, even with it like down to the level of individual neurons in the fly nerve cord, for example, which is, you know, what I worked on as a PhD student. Those individual neurons that you can look from segment to segment, they have names, right? And you could say, here's the RP3 neuron, and it innervates this muscle here, and it does it 100% of the time in a wild type animal and I can see it from segment to segment, but in a, you know, if you start adding genetic variation. Maybe it only does it 70% of the time, but that's really probabilistic. There's nothing from outside the system that is determining that because on two sides of the embryo or from one segment to another, it either makes the connection or it doesn't. So there's genuinely probabilistic stuff that happens during neural development. That can influence um the way that your brain is wired, literally, right? And if you look at the brains of monozygotic twins from like as as soon as you can image their brains, they're already different from each other. Of course, they're much, much more similar than any non-related people, but there's still a lot of, of, um, variation left over. And, you know, what's interesting in neural development in particular is that you could say, OK, maybe, you know, I've made a, I, I've made a few more neurons here than, than, than there, uh, or then in my, in my imaginary twin, um, I have a few more axons going from this region to that region, whatever. It's kind of like quantitative differences. But you can also get qualitative differences, and the reason is that neural development is a series of, of contingent events where for each event to happen properly or, you know, I, I shouldn't say properly, but along the typical um way. The previous events have to also have happened. And so you can get noise that happens at one time during development that then has cascading effects over the subsequent trajectories of development. And to me that can explain things like um like the inheritance of handedness. So handedness is like say on average 10% of the population is left-handed. If you have, uh, you know, both of your parents are left-handed, your, your chance of being left-handed is about 25%. If your identical twin is left-handed, it's also about 25%. But it's still just probabilistic, right? If you cloned someone left-handed. And you made 100 clones, 25 of them might be left-handed and 75 might be right-handed. So, um, And the idea is that like at some point during development, there's some sort of noisy thing and then development gets channeled down one trajectory or another, which are in a sense competing in that in that example. So, um, yeah, for me, developmental variation, this endogenous variation, which, like you said, it's just based on the randomness, the stochastic nature of the molecular processes that are actually going on in the developing brain plus. These potential cascading contingencies um is a major source of differences in how our brains get wired that can be expressed in behavioral tendencies.

1:17:39 Ricardo Lopes: So I have one last question then. uh, I, I want to ask you this because since you also wrote an entire book on uh the biology and the neuroscience of free will, I want to hear your thoughts about this. So, um, I mean. Do you think that uh perhaps this sort of, of, of course it might be also because it is very relevant, it's just part of uh the uh the culture surrounding the scientific community and all of that, but do you think that perhaps There's also a tendency of uh behavioral scientists to endorse determinism because in a way that would make uh their predictions or their ability to predict behavior uh stronger. I mean, because I mean if we just try to eliminate these, I mean, messy things and uh. Othersome things related to free will and things we cannot really predict, we cannot really account for in our models, then, I mean, this study or that study that I did, uh, whether it's, whether we're talking about genetic factors, environmental factors, then, uh, I mean, it has more, it, it is more authoritative, let's say if I Say that, oh I can predict with 50% probability that someone will do this in these kinds of circumstances or someone will have these personality traits if they have these genes and so on and so forth. I mean, do, do, uh, do you think that there's, uh, uh, that sort of a tendency among, uh,

1:19:29 Kevin Mitchell: I'm not sure, um, I'm not sure whether. I would sort of identify with that motivation to be um claiming more predictive power than than you should, maybe that's true for some for some people. I think more generally there's an intuition at play here where um you know I this this came to me when I was watching my son play a video game where he was interacting with this non-player character. And he said something to it, and it had certain responses and it's certain actions and so on, and I was thinking, well, OK, it's like he's doing something and playing that game, but the non-player character is not doing anything because they're just playing out their programming, but they've been programmed to act in certain ways and they're just playing that out, and I think the. The, the intuition that you're talking about comes from thinking, well, shit, maybe we're like that. Like, maybe I've been preprogrammed, uh, you know, first by evolution because we we're all humans and, and, uh, you know, we've spent the, the conversation here talking about differences, but The similarities, you know, are, are huge, right? Everything that makes us human and different from other animals, um, so that human nature is there and we can't change that. And then we have our individual nature it's like, well, I didn't choose to have these genes that made me more extroverted or more neurotic or whatever. Um, SO there's this sense like I can't change that and now that's affecting my behavior. So am I really in control? Do I really have free will or am I just playing out my programming? And I think it's a, it, it's like it's not an unreasonable intuition, but it's wrong in the sense that uh it gives too much weight to these predispositions in like as if they determine your behavior on a moment to moment basis. And they don't, right? What determines your behavior on a moment to moment basis is this lifelong trajectory that we talked about earlier of how you have come to be the person that you are with the knowledge that you have, with the beliefs, with the, um, with the habits and policies that you have, with the commitments that you made, with the ongoing projects, with all of your, the relationships that you have with everyone and all of your memories. As well as these underlying biological predispositions which are in play, but they're like a minor, they're not trivial because they've affected. The trajectory of your life, right? But they haven't governed it directly in a deterministic fashion. And so, you know, once you think that At any moment you have some agency because, uh, basically evolution can't pre-wire in the right thing to do in every situation. For certain organisms, they, they get around in a, uh, their, their world is not that complex, and they don't have to be very flexible because they're only responding to a few different things in the world, right? And their evolution can kind of wire in some fixed control policies. That allow the organism to adapt to the conditions, the small range of conditions it encounters. But the whole point of having a nervous system is that you can adapt to all kinds of new stuff, right? Because evolution can't say, give you just a lookup table for every possible scenario you're going to encounter. Instead, it hit upon the idea of like, why am I doing all the work? Let me give these things of their own onboard learning system and let them figure it out themselves, right? So when you encounter a new scenario. Well then you're gonna take in the information, assess it with respect to what you know about the world, with your current states, your motivations, what are your goals, um, any other pressures and information, and then decide what to do, and that's a genuine deciding. It's not, it's not just, uh, looking up some algorithmic thing that already pre pre-exists. It's really open, right? There's bunch, you could do a bunch of things and you're going to try and figure out what to do and at every moment of your life. Exercising that agency contributes to you becoming the person that you are right now, right? So, yeah, so my view is that there's, there's nothing in the statement that we're not blank slates. That undermines free will, right, in the sense that my free will. And in fact, Ricardo, this is exactly why I wrote Free Agents is because I had written innate and some, and, and some people were taking that to my mind the wrong way, like as if we're, we're just robots, and I wanted to actually explore the idea that no, we, we, we do have predispositions, but they underpin our agency. They don't undermine our agency. And, you know, in one way you could say, The ability for you to act freely is to act like you, right? I mean, that's, you don't want to freedom is not the, the, uh, you know, ability to do just anything. That's not, that's not a freedom, that's just randomness, right? What you want as an autonomous agent for you if you think of freedom reflecting autonomy and your, your ability to control things is that you can express yourself, right? You can express your identity, you can do you. You can be Ricardo, right? Um, AND you can continue to do that through time, and that's what it means to be a self, right, is to have that continuity and integrity where, um, you're bringing your identity and your character and your knowledge to the world and continuing to act like yourself, right? Um, AND so I don't see anything in that. That undermines the idea that that we're acting freely. Unless you start with a weird conception of freedom, which is effectively like freedom from any prior cause whatsoever, which is just kind of nonsensical. It just doesn't make any sense. You just be random, right? I don't want to be free from any prior causes. I want to be a cause of things in the world and I think that I can be.

1:25:27 Ricardo Lopes: Well, uh, let me just tell you before we go that, uh, I should warn you to not use the word non-playable character too much or NPC because, er, without even noticing you might start also using main character and then you might get into a rabbit hole on the. Internet and you might become a conspiracy theorist and thinking that everyone out there is an NPC and the main character so or at the very least people might confuse you with uh Gen Z so they might think you're Gen Z so

1:26:05 Kevin Mitchell: we had, we had NPCs before the Gen Zs came along. Yeah.

1:26:10 Ricardo Lopes: OK, great. So, uh, look, I, I'm going to leave again a link to your latest book Free Agents in the description. Apart from that, where can people find you and your work on the internet?

1:26:23 Kevin Mitchell: Yeah, so I have a, a website. It's, um, KJ Mitchell.com, and I'm on Blue Sky these days at, at wiring therain, um, I guess it's tbsky.social. Um, YEAH, so, generally there and, um. Uh, AND on the dissenter, generally, so.

1:26:44 Ricardo Lopes: Well, maybe next year you'll be back. Let's see.

1:26:48 Kevin Mitchell: Yeah, great. Thanks a million.

1:26:50 Ricardo Lopes: Yeah, thank you. Hi guys, thank you for watching this interview until the end. If you liked it, please share it, leave a like and hit the subscription button. The show is brought to you by Enlights Learning and Development done differently. Check their website at enlights.com and also please consider supporting the show on Patreon or PayPal. I would also like to give a huge thank you to my main patrons and PayPal supporters, Perergo Larsson, Jerry Muller, Frederick Sundo, Bernard Seyaz Olaf, Alex, Adam Cassel, Matthew Whittingberrd, Arnaud Wolff, Tim Hollis, Eric Elena, John Connors, Philip Forst Connolly. Then Dmitri Robert Windegerru Inai Zu Mark Nevs, Colin Holbrookfield, Governor, Michel Stormir, Samuel Andrea, Francis Forti Agnun, Svergoo, and Hal Herzognun, Machael Jonathan Labran, John Yardston, and Samuel Curric Hines, Mark Smith, John Ware, Tom Hammel, Sardusran, David Sloan Wilson, Yasilla Dezaraujo Romain Roach, Diego Londono Correa. Yannik Punteran Ruzmani, Charlotte Blis Nicole Barbaro, Adam Hunt, Pavlostazevski, Alekbaka Madison, Gary G. Alman, Semov, Zal Adrian Yei Poltontin, John Barboza, Julian Price, Edward Hall, Edin Bronner, Douglas Fry, Franco Bartolotti, Gabriel Pancortezus Suliliski, Scott Zachary Fish, Tim Duffy, Sony Smith, John Wisman. Daniel Friedman, William Buckner, Paul Georg Jarno, Luke Lovai, Georgios Theophannus, Chris Williamson, Peter Wolozin, David Williams, Dio Costa, Anton Ericsson, Charles Murray, Alex Shaw, Marie Martinez, Coralli Chevalier, Bangalore atheists, Larry D. Lee Jr. Old Eringbon. Esterri, Michael Bailey, then Spurber, Robert Grassy, Zigoren, Jeff McMahon, Jake Zul, Barnabas Raddix, Mark Kempel, Thomas Dovner, Luke Neeson, Chris Story, Kimberly Johnson, Benjamin Galbert, Jessica Nowicki, Linda Brendan, Nicholas Carlson, Ismael Bensleyman. George Ekoriati, Valentine Steinmann, Per Crawley, Kate Van Goler, Alexander Obert, Liam Dunaway, BR, Massoud Ali Mohammadi, Perpendicular, Jannes Hetner, Ursula Guinov, Gregory Hastings, David Pinsov, Sean Nelson, Mike Levin, and Jos Necht. A special thanks to my producers is are Webb, Jim Frank Lucas Stink, Tom Vanneden, Bernardine Curtis Dixon, Benedict Mueller, Thomas Trumbull, Catherine and Patrick Tobin, John Carlomon Negro, Al Nick Cortiz and Nick Golden, and to my executive producers, Matthew Lavender, Sergio Quadrian, Bogdan Kanis, and Rosie. Thank you for all.