https://youtubetranscript.com/?v=yOJR-nEhNMk
A couple of announcements first. It looked like the personality assessment program got overloaded on the due date. I presume that was because a large number of people in the class were doing it in the last two hours. So I’ve recalibrated the due dates. The earliest due date is a week from the previous due date, which I think is the 23rd. And then I’ve divided it up by your last name. So you should have got an email from me telling you when the personality assessment is in fact due. So those of you from A to H or something like that have to hand it in first, but you still had an extra week, so that shouldn’t be too bad. There’s also no class on Thursday. I have to be out of town. So that’s the announcements. Okay, so today we’re going to talk about a personality trait, which is agreeableness in one end of the distribution and something sometimes called antagonism on the other end. It’s a difficult trait to conceptualize, in my estimation. It’s always been the one that’s puzzled me the most in some ways. And I also think it’s the one that gets most obviously troublesome at both ends of the spectrum. So extremely agreeable people are empathetic and compassionate and compliant. But the downside of that is that they are not that good at standing up for themselves, and so they’re often manipulated and pushed around. For example, they’re not as good at negotiating for their own salaries, and that’s something we’ll discuss in a little bit. And my experience clinically has been that agreeable people, the consequence of their compliance is that they tend to be resentful. Just because you’re agreeable and compliant doesn’t mean that at some level of your psyche, you’re not interested in a fair deal. And if you’re not particularly good at negotiating for yourself, especially in the presence of disagreeable or antagonistic people, then you’re going to be left with the short end of the stick. I think the typical complaint of someone who’s very agreeable is, I do so much for other people and they seem to do so little for me. And so if you feel that way, there’s a reasonable probability that you’re agreeable. There’s some probability that you’re too agreeable, and maybe you should stop being so easy to get along with. And that’s one possibility. On the other end of the distribution, the antagonistic end, well, if you’re antagonistic enough, you end up in prison. It’s the best personality predictor of antisocial behavior and seems to be associated with sort of cold-blooded aggression rather than the reactive form, which would be more characteristic of negative emotion, neuroticism. And I think that people who are lower agreeable, or higher antagonism, are probably predatory. Now, you know, human beings are predators, and we have been for a huge swath of our evolutionary history, maybe for all of it, you know, maybe not when we were one-celled animals, but for a huge chunk of it. And during that time, our capacity to be predators has definitely been key to our survival. We perhaps started to cook two million years ago. It’s a very peculiar human activity, you know, the use of fire and cooking, and there’s no other animals that do that. It made meat much more digestible for us. It shortened the length of our digestive system a lot. So if you look at human beings compared to chimpanzees, you know how a chimpanzee is shaped, they’re sort of shaped like this, and it’s because a chimpanzee is mostly intestines, and the reason for that is because they mostly eat leaves. They do hunt, though, they will catch monkeys, but they mostly eat leaves, and so most of what a chimpanzee’s life consists of is sitting around chewing leaves and then digesting them, because leaves are not particularly nutritious. Whereas human beings, we figured out how to hunt very effectively, and then we figured out how to use fire, and fire makes almost everything more digestible and increases the caloric efficiency of the food. And in particular, it makes meat more digestible, and so that meant that we could decrease the length of our gut and increase the length of the size of our brain, which seems like a pretty good trade-off fundamentally. So anyways, before we can talk about this, I’m going to go over something perhaps I should have gone over before, which is a bit of elementary statistics. Well, they’re not so elementary, because people dispute them continually, but I want to talk to you about the normal distribution, because you’re going to run into this pretty much in every science, but particularly in psychology, because most psychologists will tell you that most things in the world are normally distributed, and it’s not exactly obvious why that is. It’s also not obvious, by the way, that it’s true. There are lots of things that aren’t normally distributed. Money is not normally distributed. Productivity at work is not normally distributed. The number of books people publish is not normally distributed, nor is the number of records people record or the songs that they write or the paintings that they make. Probably anything that has to do with productivity is distributed in a preto distribution, which we’ll talk about more towards the end of the class. A preto distribution is a distribution where most people, at least in the human case, where most people have or do very little of the productive work, and a very small minority do virtually all of it. So it’s a vicious distribution, and it’s much more common than you’d think if you went through the average psychology training program. I didn’t really learn about preto distribution until about 10 years ago, and then it was because I had a client who was a financier and a psychologist at the same time, and he taught me a lot of things that I didn’t know, which was quite convenient. So anyways, back to the normal distribution. Psychologists will tell you that things are normally distributed, and almost all the statistics that you will encounter are predicated on the assumption that the data that you’re manipulating is normally distributed. Now you’re supposed to check that assumption before you engage in the statistics, and if the data isn’t normally distributed, there’s various things you can do to it to make it approximate a normal distribution before you do the analysis. Although sometimes you can do, I think, real conceptual and empirical damage to the data by forcing it into a normal distribution, especially when it’s one of the things that actually isn’t normally distributed, which theoretically aren’t supposed to exist, but which do. Now let me show you something about a normal distribution. And this is a Galton board, that’s what this thing is called, and Galton was this incredibly intelligent polymath from the late Victorian times who made contributions in a whole variety of scientific disciplines, psychometrics being one of them, the measurement of human behavior. Galton is often regarded as the first person who studied intelligence, but that’s not true, actually. Galton studied what he called eminence, and eminence was sort of the probability that you’d be in the same class structure, say, as an aristocratic Englishman, and he thought of that as an inborn trait, you know, which made the British aristocrat superior to the British commoner, and then, of course, superior to everyone else as well. And people confuse that with intelligence, but if you read Galton carefully, you find out that he really, he didn’t conceive of intelligence, it was more like talent or ability, well, or eminence, which is what he called it. So now Galton was interested in measuring things, and so he made a lot of measurements of people, various aspects of people, physical aspects, and he started to develop some of the presuppositions that we use in our statistics today, and one of the things he noticed was that a lot of things that you can measure, like height, are normally distributed, which is to say that most people are average, and there’s a distribution around the sides, so that most people are average, a few people are really tall, and a few people are really short, and it’s kind of a continuous distribution across that entire span, and so that’s a normal distribution. Now, the theory is that things are normally distributed when a lot of different factors contribute to them, so you can imagine in the case of height, well, what might contribute to it. Prenatal conditions might be one, nutrition might be another, genetic heritage might be which you could also divide into the genetic component you get from your mother and the genetic component that you get from your father, posture, gender, those are a few, maybe ethnicity, the Dutch, for example, are extremely tall, so there’s a variety of factors that might contribute to height, and when a variety of factors contribute to something, then it tends to be normally distributed, that’s the hypothesis, and that’s because the variation is random, so let me show you what it means for variation to be random. Now, so Galton figured this out, now this is a Galton board, now what a Galton board would normally be would be a board with nails, so each of those little dots there, you can imagine those are nails, and then it would be slanted a little bit, and then there would be a spout at the top, right in the middle, that would drop marbles down, and then at the bottom there’s a bunch of containers where the marbles could land, and then you can see how the marbles distribute themselves if you let them fall randomly, now this is random because at each nail, imagine a nail is a choice point, right, the marble can go right or left, and it’s entirely random, and so here’s what happens if you run something randomly. Now, you can see as the balls accumulate, or the marbles accumulate, that they’re starting to be considerably more in the center, and considerably fewer out on the edges, now the technical reason for this is that there’s lots of ways to get to the middle, and hardly any ways to get to the edges, right, because if you want to go all the way to the right, for example, then you’d have to, there we go, if you wanted to go all the way to the right or the left, then every time you hit a nail, you’d have to go either right, so right, right, right, right, right, right, etc., or left, left, left, left, left, etc., and you can imagine since there’s only a .5 probability of each of those, that once you have 10 or so of those stacked together, the probability that you’re going to take that particular path is vanishingly small, whether we’re in the middle, there’s all sorts of different ways you can get there, right, left, right, left, right, left, or right, right, left, left, or, well, you get the picture, there’s all sorts of ways, so there’s more paths to the center than there are paths to the edges, and so that’s why you get a normal distribution, and the normal distribution does some pretty weird things, so for example, we’ll go to this next one, this is another kind of Galton board, and we’ll run the first part of it, and you’ll see that you get your typical normal distribution, so you can imagine that each of those little nails is a factor in some sense, a random factor, determining where you end up in the distribution. Okay, so after we fill this up, and we have a normal distribution, then you see the little yellow and black line there, we’re going to take that out, and then the balls are going to fall again, what do you think will happen when they go to the next level? Any guesses? Yeah. Okay, so that’s one hypothesis, the balls will be more likely to go to the edges. Anyone else? What do you guys think? Be more flattened? More flattened? People will agree with that? Okay, so we better withdraw it quickly, it’s starting to stack up here, so okay, let’s see what happens. Low and behold, if we kept running this, you’d get another normal distribution, so it actually doesn’t change it a bit. So the normal distribution is a pretty robust phenomena. So, now if you’re doing basic statistics, let’s go back to that for a minute, what you’re really trying to do is to figure out what the probability is that two distributions will overlap. So for example, let’s say you took the distribution of height among women, which is about 5 foot 4 on average, and so there’d be a normal curve around that, because there’s some very short women, and then there’d be a normal distribution around men, and men are on average 5 foot 9. And then you could calculate the probability that those two curves were the same by chance, and that would depend to some degree on the number of people in your sample. So to do the statistics, you need the two normal distributions, you need the number of people in the sample, and then you need a way of calculating the probability that those two distributions could have been drawn from the same population. So that’s your hypothesis basically, that’s the null hypothesis, is that the two samples are drawn from the same population, in which case there’s not going to be any difference between them, or there’s going to be random differences between them. And the cutoff point is that if the two samples you draw have normal distributions that overlap in such a limited manner that that would only happen in one of 20 cases, right, so you’re drawing random samples from the population, then you presume that that’s a significant difference. Now it’s arbitrary, that’s p is less than 0.05, it’s arbitrary. Now you might say, well why did psychologists pick p is less than 0.05, and the answer to that is there is no reason except that there’s two errors that you can make when you’re calculating your statistics. One error is you find a difference when there isn’t one, right, you find a difference when there isn’t one, and the other is that you don’t find a difference when there is one. And those are type one and type two errors, and what you’re trying to do is to minimize both of them simultaneously, and that’s, you can’t say where that’s most suitable, because by reducing one error you increase the other. So psychologists had to pick some cutoff and 0.05 seemed to be as good as anything else. So now here you want to look at this because actually you really need to get familiar with what a normal distribution looks like and the ways that it’s parsed up. So you see right at the center, that’s the mean and the median and the mode, if the distribution is perfect and normal, those three things are going to be the same thing. At the next level there you see percentage of cases. So at three standard deviations below the mean, there’s only 0.13% of the population that’s three standard deviations below the mean, or three standard deviations above the mean. So we could say, we can think about IQ, because people are sort of cognizant of IQ to some degree. 100, which would be the mean in terms of IQ, is the average IQ of someone who’s your typical high school graduate. And then 115 is the IQ of your typical state school graduate in the US. So there’s a difference of 15 points and that corresponds, if you look here in terms of percentages, which is the bottom line, is 50% of people have an IQ above 150% have an IQ below. And it’s because it’s defined that way, by the way. And then if you go up one standard deviation, you see that pretty well, we’ll say 85% for the sake of simplicity, 85% of people have an IQ below 115. And then if you go up two standard deviations, you see that 98% of people have an IQ below two standard deviations, and three standard deviations is 99.9. So when you get up to 130, there’s only 2.5% of the population has an IQ that high. That’s about where you guys are, somewhere between 120 and 130. So you look around and you think, well, this is an average class of university students, but you’re not average in terms of intelligence. And that makes a huge difference because intelligence is one of the best predictors of career success, for example. If you go up to three standard deviations, that’s an IQ of 145. And there’s undoubtedly a couple of you in here. Well, it’s one in a thousand in the population. There’s a couple of you in here that have an IQ of 145, or maybe more than a couple. Now, one of the things you want to notice is, because people get very confused about this with regards to percentages. So let’s say that you were assessing two people to hire them, or you were assessing them for their intelligence, which is something that might be done, for example, if you were screening them to go to university, because the SATs do that in the US, or if you were screening them using the graduate record exam. And you have one candidate who’s 98th percentile and then other candidate who’s 99th percentile. Okay, so think about that for a minute. 98th, 99th. Where is there more difference? You have three candidates. One’s 85th percentile, one’s 98th percentile, one’s 99th percentile. Where’s the biggest difference? Is it between the 85th and the 98th? Yeah, in terms of of improbability. So it’s the same distance. Now, that’s what you have to remember when you’re thinking about percentiles, because 99 and 99.9 percentile don’t sound very different, but they’re the difference between one and a hundred and one and a thousand. So it’s a huge deal. That’s why you see that the normal distribution is divided up into these standard deviations. And those are basically identical differences. And those are often identical differences in terms of ability. So, for example, you know, you might think that once you hit 130, going above that isn’t going to matter, but that’s not right. There’s just as much difference between someone who has an IQ of 130 and someone who has an IQ of 145 as there is between someone who has an IQ of 115 and an IQ of 130, even though the percentiles start to switch around. So you need to know these sorts of things. If you’re when you’re starting to think statistically, because otherwise, well, you can’t understand what you’re doing. And that’s that’s generally not a good thing. So although it’s very frequent. Okay, so agreeableness is one of the five dimensions of the big five. We all know that. And it can be broken down into these two aspects, compassion and politeness. And so you can kind of figure out how agreeable you are by looking at these items. They’re all positively scored for the sake of simplicity. So if you intensely feel others emotions, if you’re always inquiring about how others are doing, and you actually care rather than doing it for the sake of show, if you if you’re capable of sympathizing with other people, if you respect authority, if you don’t like to seem or be pushy, and you don’t like to impose your will on others, then you’re an agreeable person. And a disagreeable person or antagonistic person, obviously, is the opposite of that. So they don’t feel people’s emotions very profoundly. They’re not really all that concerned automatically about how other people are feeling. And they’re not that sympathetic. It’s funny, because in my clinical practice, I can tell the difference between the agreeable people and the non agreeable people. Because when the agreeable people come and they have a cup of coffee, they always bring me one. But when the disagreeable people show up and have a cup of coffee, they only bring a cup of coffee for themselves. So it’s quite comical, actually. So now you might think that, you know, because our our our what would you say, our society is at the moment tilted towards regarding agreeableness as a virtue, you know, because you should be kind and you should be empathetic and you should be compassionate and all those sorts of things. But our research, this isn’t published yet, because it’s complicated to communicate. And we haven’t figured out how to do it yet. But what we’ve found is that if you push any of the traits too far, they fall off a cliff fundamentally. So if you get too agreeable, then you’re dependent and you can’t make decisions and so on. And maybe you’re kind of a Freudian Oedipal mother, too, right? You’re so you’re so concerned with your with your child’s well-being that you can’t you’re not harsh enough to send them outside or or make them do anything they don’t want to do because it might upset them. And so that’s not so good. And then if you’re too disagreeable, well, I already said what that was like. If you take antisocial people who are in prison, you know, they’re they’re aggressive and violent. And, you know, the criminal types and their their predatory are low in agreeableness. So you can see how that’s a bad thing. But then in the middle range, it gets more complicated, because then agreeableness, where you’re located on the normal distribution, how good that is, depends to some degree on what it is that you are going to do. So one of the ways to think about how to maximize your success in life is to attempt to match your personality to the environment. And so now what we’re going to do is we’re going to look at how agreeableness plays out as far as we can tell in the actual world. This is, I should tell you too, this is this is the sort of discussion that really people really don’t like everything we’re going to talk about pretty much from here on in rubs people the wrong way, because our culture is very likely to assume that most of the differences between people are cultural in origin. And almost none of the data show that what they show is that as at least as your culture becomes more egalitarian, which is hypothetically what you’re aiming at, the differences, gender differences, for example, not only become more biological, because you’ve eradicated all the cultural difference, but they also become more pronounced. So for example, the gender differences between men and women are largest in the most egalitarian societies, like Norway and Sweden and so on. And so that the theory there is that, well, once you’ve eradicated all the environmental differences, all that are left are the genetic differences. And so that sort of echoed in a weird finding in families. So, you know, you might think that if how you raised your children mattered, then children who were raised well would be more similar than children who were raised badly, because that would reflect the effect of your parenting, right? So you’ve done a good job as a parent, so your children are more similar, where someone who just ignored them, those children are different. But that’s actually the opposite of what happens, is that the children that are neglected are more similar, and the children who are attended to well are more different. And the reason for that, I think, is the same reason that we see more pronounced differences in egalitarian societies, which is if you care for your children, it means that you make a very individual relationship with each of them. So you’re not necessarily using the same strategies for each child, unless you think of them as meta-strategies. And like, a meta-strategy might be, well, I get to know you, and I get to know you, and because you’re different people, I react differently to you. And what constitutes my philosophy of parenting would be act differently towards the children’s differences. And so maybe if you’re particularly good at that, your children turn out maximally different from one another, and that’s because you’re allowing their genetic differences to manifest themselves. You’re not oppressing them, you know, and trying to cram them into the little box that you think that they should fit in. Now, the downside of that is that you’re going to get differences, you know, and some of them might be quite surprising to you. So, all right, so let’s look at some gender differences to begin with. So women are half a standard deviation more agreeable. So we looked at the standard deviation. So half a standard deviation would represent, let me see if I can figure it out here. So, 50, it’s about 34. So, 1767. The average woman would have to be less agreeable than 67% of women to be as disagreeable as the typical man. So that’s the basic size of the effect. The difference, now you see how the two distributions overlap, right? So, what the first thing you can see is that in the middle of the two distributions, there’s a space, and that’s a space that’s filled with men and women. So, most of the space is occupied by men and women. But if you look off to the right, say, the most agreeable women, the most agreeable people, say three standard deviations from the mean, they’re all women. Well, that might not be that relevant. But if you go to the left, all the people who are three standard deviations below the mean are men. Now that really starts to matter. Here’s one way it matters. Who’s in prison most? Men. Now, you don’t throw anyone in prison unless they’re like three standard deviations below the mean and agreeableness, which would put them at about one in a thousand. In fact, more people ended up in prison than that. But the point is, once you get out on that edge, then the gender differences become absolutely staggering. And that’s a weird thing about normal distributions, right? You can have a very little mean difference, because the mean difference here isn’t that big. It’s only half a standard deviation. It’s not that massive. But as you move out towards the extremes, the effects accumulate, and they have huge societal effects. So, the CL here, I can’t remember what that stands for. The probability that a randomly drawn woman will be more agreeable than a randomly drawn man is 63%. Now, women are also higher negative emotion, and the difference isn’t quite as big, but it’s roughly comparable. So fundamentally, the two big gender differences are in agreeableness. So women are more compassionate and sympathetic and empathetic and less confrontational. And then negative emotion, and negative emotion is anxiety, emotional pain, frustration, disappointment. They all kind of clump together into one axis. So the agreeableness dimension seems to be stable across the lifespan, the gender difference, whereas the neuroticism difference only seems to emerge in puberty for women. And then it stays constant pretty much for the rest of the lifespan. So, one of the things this means is that the probability that a given woman is more agreeable than a given man is about 63%. And then also, in terms of negative emotion, say it’s approximately the same thing. But then you might think, well, what’s the probability that a given person will be higher negative emotion and more agreeable? And then you see how, because these dimensions are orthogonal, right? They’re unrelated, essentially. So that means you can add them together. And so once you start to add the personality differences together between men and women, they actually start to become quite big, far more than half a standard deviation. Women are also more extroverted than men. So that’s funny, eh? Because extroversion is a positive emotion dimension, whereas neuroticism is a negative emotion dimension. But women are more extroverted too, which means they experience and display more positive emotion. And that seems to be the aspect of extroversion that seems to carry most of the gender differences in enthusiasm rather than assertiveness. And so women laugh more than men, for example. So there’s an old cliché that women are more emotional than men, and that actually seems to be the case if you look at negative and positive emotions specifically. Conscientiousness also seems to be linked to emotion, to disgust and to shame and to guilt. And there’s very little gender difference there. So it isn’t like women are more emotional than men across all the emotions. But the gender differences are larger in more egalitarian cultures. Well, the thing is, this is starting to become difficult to dispute because the size of these studies is starting to become extreme. So this one study, Schmidt et al., they studied 18,000 people across 55 cultures. So that’s starting to become a pretty impressive, you know, a pretty impressive sample. UN indices of gender equality and economic development were associated with larger gender differences in agreeableness, but not in extroversion or neuroticism. So that’s interesting. So the effect of egalitarianism doesn’t seem to be spread across all the gender differences just in terms of agreeableness. And gender predicts more powerfully than gender equality. Now, these are interesting studies. And, you know, I don’t think that the theorists in the humanities have really caught up with them because there’s a huge stress in the humanities and modern universities on the idea that gender differences are cultural in nature and can be eradicated by cultural transformation. But the evidence seems to suggest that the only way you could eradicate them by cultural transformation would be by working very, very hard against the biological substrate. And you’ve got to wonder if you really want a society to do that. You know, like, what kind of society do you want? Do you want a society where everyone is allowed to be who they are, so to speak, and to be successful at that? Or do you want a society that does everything it can to make people the same, regardless of their individual, you know, of the individual differences that are intrinsic to them? It is by no means obvious that that’s what you want. So you’d have to do some pretty wicked social engineering to eradicate these differences. And it looks like the social engineering that you’d have to do would run counter to egalitarianism, right? Because since egalitarianism seems to heighten the differences, it only makes sense that you’d have to run counter to egalitarianism to reduce them. So the LIPA study here, where there’s an association between the UN indices of gender equality and gender differences and agreeableness, I think that was 500,000 people. So the internet has made it possible to run huge samples. Now here’s some interesting things, too. The occupational preferences between men and women are actually a lot larger than the personality differences. And the occupational differences are large. And like the gender differences in personality, the occupational preference differences increase as you move towards more egalitarian societies. So the big difference between women and men seems to be that women prefer working with people and men prefer working with things. Now, it’s an interesting way to look at it, because for a long time, you know, if you’re thinking about sexist stereotypes, say up to the 1950s, maybe a little later than that, maybe up to the mid-60s or so, the idea would have been that whatever gender differences there were in occupational skill, in occupational distribution would have had more to do with skill. That was so that, you know, men were more skilled, say, at engineering, and women were less skilled at it. But what seems to have been, what seems to be the case is that if you look at the basic predictors of success, like intelligence and conscientiousness, there’s virtually no gender differences at all. So what that seems to mean is that the reason that there are occupational differences, at least in egalitarian societies, is because the occupational differences are driven by choice, not by ability. Now, that’s something to think about, you know, because again, that seems to indicate that if you’re going to push hard against gender preferences, what you have to do, in some sense, is restrict the degree to which people’s choices determine your occupational outcomes. And, you know, once again, it’s not exactly clear that you’d like to live in a society that would do that. I mean, unless you think that precise equality across every dimension, in terms of occupational, what, the number of men and the number of women in each occupation is such an important index that it actually trumps what people, as individuals, actually want to do when you allow them free choice. So the big difference is, and it’s one standard deviation, so to be as interested in things as the typical man, the typical woman would have to be more interested in things than 85% of women. And to be more interested in people than things, to be more as interested in people as the typical woman, a man would have to be more interested in people than 85% of men. Now, that’s starting to become a very large difference. Now, there’s other differences too. This is using the Holland Holland is a very well-renowned industrial organizational psychologist, and he divided jobs into these six categories. Realistic, investigative, artistic, social, enterprising, and conventional. And what you see is that men show stronger realistic and investigative interests. The realistic category is particularly big. It’s eight-tenths of a standard deviation. And women show strong artistic, social, and conventional interests. And there was no gender difference at all in the degree to which people preferred enterprising occupations. So here’s from a different study, from Psychological Bulletin, looking at there’s this phenomena that people are quite upset about, especially in the US, although it’s also an issue in Germany, that women are underrepresented in science, technology, engineering, and mathematics dimensions. But if you look at the interests of men and women, you see under men that men are 1.1 standard deviation more likely to be interested in engineering, and three-four-tenths of a standard deviation to be more interested in math, science, and three-tenths of a standard deviation more interested in math. Now, again, that has nothing to do with ability. So, you know, there is some evidence, although it’s really highly disputed, and the evidence, I think, keeps going back and forth, that men are better at spatial rotation. And that seems to be linked to testosterone. But like I said, this idea has been attacked very heavily. There’s also some evidence that if you look at the intelligence distribution of men and women, that the women’s distribution is a little higher peaked and a little less wide. And what that means is there’s far more men who fall into the extremely intellectually impaired category. And that sort of seems to go along with the fact that the vast proportion of children in school who are in trouble are boys. But they’re also overrepresented at the extreme upper end of the continuum. So the male distribution is flattened. And the people who believe that believe that, well, it’s typical in animal communities, mammalian communities, for there to be more variability among the men. So because men are expendable fundamentally, and so in some sense, the species can take a bigger chance with variability, even though the means remain the same. Now, that was what Larry Summers was referring to. He was a president of Harvard when he suggested that there might be intellectual differences between men and women that in part accounted for the differences in STEM prevalence. And he just got roasted for that. In fact, I think they fired him, which was really unfair, I thought, because he was just referring to a study. He wasn’t making a claim that that was his opinion. But we can put that aside for the time being. It doesn’t look like you need to invoke an intellectual capacity explanation to account for the underrepresentation. You can just see that the interests diverge. Now, you could attribute that to stereotypes and so forth. But then you run up against the problem that the differences are bigger in the countries that have done the most to eradicate the stereotypes. So I don’t really see how that hypothesis can survive. As far as I’m concerned, it’s just been proved wrong in the last 10 years. The gender differences are not primarily cultural. So then you see this underrepresentation here. Completing degree in employment in math, 50% fewer women in math, 40% in the physical sciences, 25% in computer science, and 25% in engineering. So here’s other occupational differences that are quite extreme. And there’s other reasons for these. I don’t think these are only interest related. So let’s say, what are differences between men and women? Obvious ones, right? There’s the primary and secondary sexual characteristics. What else is different? Yes, well that goes along with this sort of male expendability issue. So yeah, they are more dangerous jobs and the probability that a man will be killed at work is extraordinarily higher than the probability that a woman will be killed at work. So what other differences are there between men and women? Yeah, they’re about four inches taller. They’re not as much heavier as you would think. I thought it was bigger than this. I’ll show you. 164 versus 194. Those are the American averages. So here’s differences between men and women. The gender differences, these sort of physiological gender differences, the psychological gender differences to some degree too, can be lumped under a biological category called sexual dimorphism. And in some animals, sexual dimorphism is how much difference there is in the morphology between a male of the species and a female of the species. And there can be a lot, like with elephant seals for example, the males are like ten times bigger than the females. Usually in a situation like that, the male has a harem. So there’s extreme male reproductive success on the part of a couple of males and then extreme male failure among the rest of them. So there seems to be some relationship between lack of sexual dimorphism and monogamy among mammals. So the less dimorphic a species is, the more likely they are to pair bond. And human beings are relatively non-dimorphic. So we’re a little dimorphic than the typical monogamous animal, but a little less dimorphic than the typical polygamous animal. And gender differences, these sorts of gender differences with the male being larger and more aggressive than the female, are not characteristic of all mammals, although they’re characteristic of many of them. So the major exception are hyenas. And the hyena females are very testosteroneized and they’re very, if that’s a word, they have very high levels of testosterone and they’re slightly larger than the males and they are more aggressive. But the price they pay for that is that their reproductive organs mimic the penis and that’s what they have to give birth through. So it isn’t exactly clear that that’s an… I’m not sure that’s an advantage. Okay, well so here are some of the differences between men and women. So you can see there’s a five inch difference in height, a 30 pound difference in weight. Men are about 40 to 60 stronger in their upper body and about 70 to 75… sorry, that’s wrong. Women are 40 to 60 percent as strong as men in upper body strength and 70 to 75 percent in terms of lower body strength. Men have thicker jaws, heavier jaws. Women who are ovulating prefer men with thicker jaws. So for example, if you show women pictures of the same men across their menstrual cycle and all you do is widen the jaw, the narrow jawed guy is more preferable when there’s less probability of conception and the wider jawed guy is more desirable when the probability of ovulation is high. And jaw dimensions are associated at least to some degree with aggression. Why would that be? Why would that be? Because it’s for biting. You know, like think about animals, they bite, you know, and human beings can bite too. So the jaw, the relationship between jaw dimension and disagreeableness, say, seems to have… it’s a somewhat of a hangover from our revolutionary past. Women have a higher proportion of body fat. Women are more resistant to diseases. That’s a good deal for women. They live longer. That’s a good deal too. There’s way more male homicide victims, which seems spectacularly unfair until you notice that there are many more male homicide perpetrators, which then seems to sort of even things out. Men kill non-intimate acquaintances or strangers 80% of the time, where women kill intimate partners or family members 60% of the time. Of course, they’re doing it a lot less frequently. So, and you never know how many of those are in something approximating desperation and self-defense, right? Excluding the odd psychopathic woman. So, let’s see. Okay, so back to these construction forestry mining utilities. Well, you can kind of see, maybe, that these are jobs… you can tell me if you disagree, but these seem to me to be jobs that might require more physical strength. That would certainly be the case in forestry and mining and construction and likely in utility work too. So, and then they’re also thing-like jobs, right? So these aren’t, you know, there’s not a lot of value on fostering close human relationships in these sorts of occupations, you know? So, here’s some interesting things. Ten most male-dominated occupations in the US. So, I guess one of the things that you might ask yourself is, would you be interested in a job like this? So, hypothetically, men are more interested in jobs like this than women. Brick mason, block mason, stone mason, so, you know, carving up stone, cement masons, electrical power line installers and repairs. That’s a tough job, Abe. It’d be very hard to do that, like during the ice storm. That’d be a hard job. Carpet floor and tile installers, heating, air conditioning, refrigeration mechanics, structure iron and rebar workers, bus and truck mechanics and diesel engine specialists, miscellaneous vehicle and mobile equipment mechanics, tool and die makers, and roofers. So, and those are overwhelmingly male, right? If you look on the right hand column there, there isn’t even 1% of, the total number of women in those occupations isn’t even 1%. You know, it’s funny, you don’t hear a lot of, like, gender affirmative action calls for structure iron and rebar worker gender equality. So, which, well, I think that’s interesting, you know? If it’s merely a matter of gender inequality, then why does it differ, what difference does it make what the category of employment is? That sort of argument shouldn’t be reserved for only the upper echelons and the most desirable jobs. You know, otherwise, that’s not a coherent argument. It either applies across the spectrum of jobs or it’s incoherent. Those are the alternatives. So, here’s the 10 most female dominated occupations in the US. It’s still secretaries and administrative assistants. Now, you know, that was the most common female job in the 1950s, and it is still overwhelmingly the most common female job. So, that’s quite interesting, although it’s incredibly dominated by women. So, you know, is that a good thing or a bad thing? It’s like, it’s not that easy to tell. Childcare workers, well, that doesn’t really come as much of a surprise, but it is notable. And one of the things you see in the Scandinavian countries, for example, it’s pretty much related to this, is that there’s a 20 to 1 difference, men versus women in engineering, and there’s a 20 to 1 women versus men difference in nursing. So, and the Scandinavian governments every now and then go through a fit of enthusiasm about attracting more women into engineering, and so they, you know, crank up their social engineering and their information about how desirable profession that is, and then the proportion of women who go into engineering rises slightly, not very much, and then as soon as they stop, you know, pushing it hard, it just drops right back down to 5%. Teacher assistants, registered nurses, bookkeeping, accounting and auditing clerks, maid and housekeeping cleaners, nursing, psychiatric and home health aides, personal and home care aides, and general office clerks. So, that’s where the bulk of women are in the, that’s where the bulk of the female dominated occupations are. So, let’s see, now did I cover all those things? Yeah, I did. So, here’s, I’ve been trying to figure this out for a long time, like, what this agreeableness thing is, and this is sort of what I’ve concluded, because I’m always interested in tying this down to the biology, and Jaak Panksepp, who’s a very good neuroscientist, has identified a number of fundamental biological, emotional systems, some of which we’ve alluded to already, and some of which we really haven’t. So, for example, it was Panksepp who identified the existence of a play system, for engaging, for example, in rough and tumble play, and that seems to be something that males prefer more than females prefer, but he’s also identified a care system, and you can think about that as the basic, there has to be a care system, I mean, for God’s sake, we’re mammals, right? I mean, the definition of a mammal is, well, first, the mammal is warm-blooded and has hair, but the next part of it is that the mammal is fed breast milk, and so, obviously, there’s a dependent relationship there, and it’s not only a, like, it’s a dependent relationship, and it’s a difficult dependent relationship, you know, like, children are very, not able to care for themselves for a very long period of time. I don’t remember if I mentioned to you this or not, but a mammal of our size should carry an infant for two years, gestation, so he should be pregnant for two years, and then the baby would be born and would be, you know, good for something, walk around and so on, like many mammals can as soon as they’re born, but because our heads are so big, and because women’s hips have to be narrow enough so that they can run, there’s an evolutionary arms race between these structures, in some sense, and so what’s happened is that, you know, women compromise by having wider hips, and babies compromise by having heads that can be compressed during childbirth, and also by being born much earlier, but the price that’s paid for that is that they’re incredibly dependent, and so, so you think, well, there has to be, well, it’s foolish to even argue otherwise, people fall in love with their babies to a staggering degree, and the typical response of new parents is, well, I never knew it would be like this, because even if you’re not interested in someone else’s baby, you tend to be unbelievably interested in your own baby, and that’s a good thing, because there are a lot of trouble, and so there’s a system, a maternal system, primarily, a care system, it’s a system that’s shared by men, because men are also very maternal for mammals, and so, I believe that’s on the one end of the agreeableness spectrum, it’s like the maternal care system is what’s driving empathy and sympathy, and all of those things, and, you know, I think actually that it was the expansion of the maternal care system to include men, that actually resulted in human beings’ capability to pair bond, because one of the things that’s really weird about people is that they share food, because animals aren’t very good at that, like chimps will do it a little bit, but they’re not too happy about it, and wolves sort of look like they share food, but that isn’t really what happens, the dominant ones just eat till they’re full, and then they don’t mind if, you know, the subordinate ones eat what’s left over, but human beings will share food, now, of course, female mammals often share food with their infants, but they don’t generally share food with, you know, their compatriot, and, you know, you think about all the, think about the language that couples who are in love use to each other, you know, there’s a real infantile element to it, it’s baby and deer, and, you know, they make little cooing noises, and they give each other little pet names, and, like, there’s something that’s like, there’s a juvenile element to it, that seems to me to be a consequence of the activation of this care system, you know, between adults, rather than from the adult to the child, and so, anyways. Then, on the other end, well, there’s a predatory aggression system, and I think that they’ve come together in evolutionary history to balance each other out, so, both men and women are caring and predatory, women more caring and less predatory, and men more predatory and less caring, but both genders have that, both capacities within them, and I think they’ve evolved so that they’re mutually inhibitory, so that a predatory male is not that likely to kill infants, whereas, and fairly likely to care for them. Now, there are exceptions to this, you know, and they’re not trivial, so, for example, stepchildren are much more likely to be killed, especially if they’re under two, you are at 100 times more risk for abuse from a stepparent than you are from a natural parent, and so, you see echoes of this in the animal kingdom, too, so, for example, if gorillas have harems, and then, of course, one gorilla has, you know, offspring with a variety of female gorillas, and all the adolescent gorillas are sort of outside that little circle, just waiting and tapping their feet for the guy in the middle to get tired or hungry or, you know, exhausted, so they can come and run in there and, you know, overthrow them, fundamentally. When they do that, they kill all the infants, and so do lions, so, the predation and care systems, you know, they’re very tightly balanced, and they are influenced to a large degree by genetic similarity, rough as that is. So, in caring professions, 5% males, let’s look at predation. 8% of the male population hunts, which is actually quite a large number, when you think about how many people are urban, right, because it’s hard for, what are you going to hunt, raccoons, you know? I mean, once you’re in an urban environment, it’s not that straightforward to hunt, but in a rural environment, it’s, you know, right there for you, fundamentally. So, it’s only 8% of the male population in the US that hunts, but, you know, you have to factor the urbanization issue in there. 91% of those people are male, so hunting isn’t something that generally appeals to women. So, now, what other reasons might there be for the gender differences? Well, this is more speculative than what I’ve discussed with you so far, so you can think about it. I mean, these are things I’m trying to get straight, and so I’m going to share with you some of the pathways of my current thinking. So, there is evidence that more disagreeable people are more likely to be successful as managers. Now, why? Well, Baudreau, who wrote a paper called Effects of Personality on Executive Career Success, said the following, agreeableness associates with being trusting, submissive, and compliant, which could be perceived as naiveté, docility, and a tendency to follow rather than lead. All right, so that’s his opinion. But then, here’s what he measured. So, these are effects of big five traits on career success. Now, you know, career success can obviously be defined a number of ways. It could be career satisfaction, or it could be like external markers of career success, and they did both. So, we’re going to look at direct, because that’s the external sort of objective markers. What you see is that if you’re high in neuroticism, that’s not so good for how much money you make. There’s a negative correlation of 0.3, which, by the way, that’s a big correlation. So, you know, you’ll hear people say that 0.5 is a large correlation, and 0.3 is moderate, and, you know, 0.2 is small, and that’s wrong. And that was clarified four or five years ago. I’ll get the paper for you. I can’t remember. It was an American psychologist, but the guy who wrote the paper, what he did was he looked at a whole bunch of social science studies and then calculated how frequently different effect sizes showed up. And what he found was that 0.5 was unbelievably large. You know, that 5% of social science studies ever got a correlation of 0.5. It’s like if you get a correlation of 0.5 in your study, you’ve either made a dramatic error or you’ve replicated something that’s already well known, or, you know, you’re in science because it never happens. 0.3, that’s a pretty good correlation. So, the fact that neuroticism is negatively correlated with how much money you make, how likely you are to ascend, and then how close you are to being CEO. Obviously, the effect size decrease. So, neuroticism is also, or sorry, extroversion is a reasonable predictor only of how much you’re ascending, and it’s pretty small. Openness has a correlation with how much money you make, but that’s probably because openness is highly associated with intelligence. And so, openness is not a good marker for intelligence. IQ tests are much better markers, so that’s an attenuated relationship. But then you look at agreeableness. It’s negative 0.32 in total, negative 0.24 for direct in terms of how much money you make. So, you know, that’s an interesting thing because one of the things that determines how much money you make is how willing you are to say no. Right? Because if you’re negotiating with someone, then the only thing you have at your back is your ability to say no and to push it, or even to ask for a raise. And, you know, pushy people are much more likely to ask for a raise, and of course those who ask for a raise are much more likely to get it. So, the guys who are hard to get along with, because most of the people who are hard to get along with are guys, are more likely to be paid more. They’re more likely to ascend the corporate ladder. They’re more likely to be close to the CEO in terms of proximity, and they’re even more likely to be rated as employable. It’s funny, eh, because you’d think that, you know, if you’re agreeable and easy to get along with and all that, that people would be more likely to rate you as a suitable employee, but that isn’t right. That’s the opposite seems to happen. So, you know, and the agreeable guys are less satisfied with their jobs. No, sorry, the agreeable people are less satisfied with their jobs as well. So, okay, so now let’s look at motivation variables. So, the extroverts work more evenings, the agreeable people, the agreeable people are less likely to. So, it’s a funny thing too, because disagreeable extroverts are narcissists, and there’s some evidence that you can derive from this data, you know, because there’s always this talk about disagreeable extroverts or narcissists being more likely to rise up to CEO level, and there is some evidence for that, at least insofar as agreeableness is a negative predictor of doing such. All right, so one of the things we’ve just found is that one of the predictors for ascendancy and proximity to a high status position is low agreeableness. Now, the next thing you might think about is, well, what role does that play in terms of the factors that men and women find attractive in each other? So, you might say, well, what do you want in a mate? If you’re a woman, you might say, well, you know, you want someone who’s kind and loving and forgiving and empathic, and those are all good things, but it isn’t necessarily the case that the empirical studies show that that’s what drives mate selection. So, we could look, this is an interesting study, it’s a few years old, a thousand French Canadian respondents, 433 males and 700 females, and so here are the variables. One is possession of resources, it’s a composite index composed of occupational prestige, income, and education, and then the other variable is acquisition of partners, sexual partners, that is, number of lifetime and preceding years sexual partners, lifetime occurrence of simultaneous partners, yes or no variable, and lifetime frequency of simultaneous partners, one to five, with five being very often. Here’s the assumption, you can, you know, decide for yourself if you think this is a warranted assumption. The number of partners a member of sex A acquires is taken as an index of how often this individual is chosen by sex B, so that’s an indication of reproductive fitness, desirability, at least as assessed by members of the opposite sex, who you would think would be the logical judges for that sort of thing. Male criteria, 166 unattached women ages 25 to 50, correlation between fertility rates and number of partners in previous year equals 0.94, males choose fertility, indicators, beauty, waist to hip ratio, youthful appearance, and neotenous facial features. Neotenous means there’s a tendency among animals as they evolve to increasingly look more in their adult stages like their juvenile forms. So here’s an example, if you look at the skull of a baby chimpanzee, it looks almost exactly like the skull of an adult human. So what’s happened is we’ve, as adults, we’re more like baby chimpanzees than the adult chimpanzees are. We’ve maintained a lot of our juvenile characteristics, playfulness, you know, the ability to continue to learn, plasticity, all those things. There’s a preference in objective beauty analysis, say, of facial features for men to prefer more neotenous female faces. And you can tell that if you look at pictures of models, they generally have relatively small noses, relatively big eyes, the sort of things that are associated with cute. And cute actually is a pretty identifiable category. Most of the things that people find cute have large eyes, and relatively, the rest of their facial features are relatively small. There’s other things that are associated with cute that aren’t necessarily associated with sexual attractiveness, because cute things also have sort of random movements, like baby-like movements. And so the things that relatively short arms, and just think of a teddy bear. So anyways, those are the male criteria. Female criteria, all respondents, correlation between socioeconomic status and frequency of simultaneous partners. For men, it was 0.49. That’s a big correlation. For women, it was 0.04. That’s zero, fundamentally. So the correlation between male socioeconomic status and frequency of simultaneous partners is 0.5. It’s a huge effect. For women, it’s zero. Now that’s a big difference. So one of the things you might ask yourself too, I don’t know if this is a reasonable thing to ask or not, but I’ve always thought that the hard people to explain are the people who are hyperachievers, the ones who are way out on the Pareto distribution. And there aren’t very many of them, and they’re usually men. So if you look at the median number of publications, for example, that a female academic makes, and a male academic, they’re pretty much the same. But the mean for men is higher, and that’s all driven by a small subset of men who are way the hell out on the Pareto distribution. Now you might ask, well, why is that? Well, it’s simple. If you want to be way out on the Pareto distribution, so you’re in the top, say, 1% of performers, or even 1 10th of 1% of performers, what do you have to do? Well, first, having an IQ of like 145 or above, that’s really helpful. Then you should be insanely industrious. So maybe you’ll work 90 hours a week, or 100 hours a week at nothing but your stupid occupation. And maybe you’ll do that for like 50 years. And so if you do that, well, you’re going to be in that top one in 1000 category. Well, the issue isn’t why there are so few women in that category. The issue is why are there any men at all in that category? Right? Because it’s an insane lifestyle in many ways. You know, it means that there isn’t much left there for the rest of your life. You’re hyperconcentrating on a single thing. Now why would men do that? Well, here’s a hypothesis. One of the things that drives the probability that a woman will find a man attractive enough to sleep with is his status. So then you can think, well, that’s an extra motivational boost for men. Now, whether that’s conscious motivational boost or not, is not the point. The question is, is it a boost that might have been instantiated as an evolutionary element of general motivation? Well, men are more competitive, they’re more disagreeable. And if they’re more disagreeable and more competitive, they seem to do better, especially in high status jobs. And then women are more likely to pick men like that for sexual partners. So at some point it becomes difficult to avoid the conclusion, especially since we know that the men on the way out on the tail end of the Pareto distribution are disproportionately likely to sleep with many women. So Warren Beatty, I think, 15,000? That was his estimate. 15,000 women. And Kareem Abdul-Jabbar, the basketball player, he figured 10,000. So for every one guy who’s got those numbers racked up, that’s 10,000 guys who don’t. So the male success distribution is extreme. And it’s associated with outlier success in some important dimension of socioeconomic success. And then female reproductive opportunity is not. So is that relevant? Well, it depends on how you think. It’s hard for me to see that it can’t be relevant. So questions? You had a question? So what that means is the higher the job, the higher their status, the more partners they have? Yep. Yep. Absolutely. And this is a single study, but this data is common. Now there’s some interesting correlates of this. So for example, you might say, why are males much more likely to engage in homicide? And Dalian Wilson figured this out a long while back. And what they showed was a very, very cool study. They took the level of income inequality in each county in the US and in Canada, and then they correlated that with the homicide rate. And what they found was that the higher the rate of inequality, so the more people at the bottom and the fewer people at the top, the higher the male homicide rate. And the correlation was .8, something like that, .85. It was like, it’s the whole reason. So what happens is that when the competition between men gets intense because hardly anyone can win, then males start to kill each other. Now, Martin and Daly also tried to figure out if that was a good idea, so they studied Chicago. And what they found was, well, most male and male homicides are young men, and a lot of them, they’re within race, and they’re often in dominance disputes. It’s like, you know, maybe it’s a gang thing. And it’s like, well, you shoot me or I’ll shoot you. It’s like, who’s victim and who’s perpetrator is not that obvious. So what happens in Chicago? Well, typically, the charges plea bargain down to self-defense. Then no one’s willing to testify against the guy because, well, after all, he just shot someone. And then he’s in prison for two years because that’s the sentence. And then he’s out in 18 months if he behaves reasonably well. So then he’s killed someone and he’s back on the street. And guess what? His status has increased substantially. So you can think about it as socioeconomic slash sexual decision. And we also know that as economic disparity grows, the social stability decreases. And the reason for that seems to be, well, mostly seems to be male on male aggression. And so what I think probably happens is that if there’s a fair bit of chance for men to climb the ladder, then only the most disagreeable of them become violent. But as you crank down on the opportunities so that the competition gets more and more intense, the level of disagreeableness that it takes in order to catalyze aggression starts, the level of disagreeableness that it takes to catalyze aggression starts to decrease until if it gets so intense that there’s no possibility for status, then the whole bloody society is going to flip over because the guys at the bottom who are like all the guys are going to decide that this is a stupid game and they’d rather flip the board over and, you know, set the markers to zero and see how they come up in the new society. So it’s possible that one of the things that motivates revolutions in human society is excessive inequality. And the reason that that’s a problem is because it’s so tightly linked to male reproductive opportunity. So look at age 30 to 39. Correlation between socioeconomic status and frequency of simultaneous partners. For men the correlation is 0.92. It’s a very small n. For women it was 0.11. It’s the difference between perfect and zero fundamentally. Female criteria. It might reflect the tendency on the part of females to choose high status partners during their own peak reproductive years. This might account for the especially strong relationship found in men aged 30 to 39. Women who typically prefer men three to eight years older than they are would then be in their mid to late 20s, a time when fecundity and thus sensitivity to male resources is at its peak. Evolutionary related hypotheses about gender differences and mate selection preferences were derived from Trivers parental investment model, which contends that women are more likely than men to seek a mate who possesses non-physical characteristics that maximize the survival or reproductive prospects of their offspring. As predicted, women accorded more weight than men to socioeconomic status, ambitiousness, character, and intelligence. And the larger gender differences were observed for cues to resource acquisition, status, and ambitiousness. As predicted, gender differences were not found in preferences for characteristics unrelated to progeny survival, sense of humor or personality. When valid comparisons could be made, the findings were generally invariant across generations, cultures, and research paradigms. So, any questions? What do you think about that? You know, it’s harsh stuff as far as I can tell. Well, we did a study that unfortunately I never published. We showed women. I thought, no, women aren’t after resources. That’s wrong. They’re after the factors that predict resources. So, what we did was we showed women a bunch of pictures. It was several pictures of the same guy. So, there were several guys. And each guy was in four conditions. The photo of each guy was in four conditions. Same photo, right? So, one guy was poor and stupid, poor and useless. One guy was rich and useless. One guy was rich and useful. And one guy was poor and useful. So, the rich useless guy had won a lottery and it paid out like $4,000 a month for the rest of his life. So, no matter how useless he was, he couldn’t squander all of it. He had resources, right? The poor useless guy, well, we don’t even have to define that. Everybody understands that. The rich useful guy, well, that’s easy to understand. The poor useful guy worked for a non-governmental agency. It was a charity, you know, in a charitable organization. And so, he was, he had high occupational status but low income and that wasn’t likely to change. So, then we asked women to rate these men on a bunch of different attributes, including personality, but also on datability, basically, and also on probability of considering a long-term relationship with someone like this. And wealth had no main effect. The only effect was usefulness. So, the poor useful guy and the rich useful guy were attractive, more attractive to the women than the two useless guys, regardless of their socioeconomic status. And the women judged the useful guys as higher in openness and conscientiousness, which was pretty smart because openness is basically intelligence and conscientiousness is hard work. And those are the best predictor of socioeconomic success. So, yay women, good work. Any other questions? No, it’s not exactly appearance. It’s more like signs of health. And that’s associated, and it isn’t even that, the things that have made women attractive to men across time are the things that are associated with health and therefore reproductive capacity. So, for example, waist to hip ratio is a marker. 0.68 seems to be about ideal. Why? Well, it turns out that women who have a lower, higher, higher, can’t figure it out at the moment, more waist than hips are more likely to die, for example, of cardiovascular disease. It’s an unhealthy distribution of body fat for women. And so, that seems to be a negative fitness marker. And then symmetry, for example, I mean, women like symmetrical men too, don’t get me wrong, but men are more attracted, say, to physical health than women. One of the markers for physical health is also symmetry, and that’s associated with beauty, clear skin, all these things are associated with health and youthfulness. Yeah, and that’s pretty much that. That’s a good question. I had actually had a paper that I was looking at that this morning, but I didn’t have enough time to include it. So, I don’t know exactly what the predictors there are. What do you think they’d be? You’d think so. Yeah, that’s right. Well, that would be a very useful thing to look up. Attractiveness? Hard to say, but a quick glance at popular culture would suggest so. Okay, we’ll see you next Tuesday.