Sex and Gender with Dan Conroy-Beam
2026-04-28 12:00:00.000 +00:00

This is Evolutionary Psychology at the podcast. Actual Evolutionary Psychology by Actual Evolutionary Psychologists. Today we have a repeat customer, Dan Conroy Beam. Our first repeat customer. I believe the first guest that we've had on for a second time. Yeah. Is that right? Or no, I guess well, I guess we had uh Christina on as a co-host. Yeah, we had Christina on as a co-host. Yeah. And we will have the Zach Garfield. Uh Zach Garfield episode hasn't been the Ray Hames one isn't out yet. Or or it might have just come out. But this is our first repeat guest in the guest. Substantive, yes, exactly. Yeah. And no one better than Dan Conroy Beam because uh Dan is deriving from scratch, he is uh god among um uh

silicone based agents, recreating the evolution of sex, sex roles, and gender. Gender. Yeah. So this is all about gender. Not all about, but half about. Or the complex interrelations between sex, sex roles, and gender. And I found it just really um, I don't know. Like I I knew a lot of the stuff, you know, reading Dawkins and uh the evolutionary bios stuff, but it was great to just hear Dan synthesize it and review it and show the elegance of these ideas and how they can be cashed out mathematically, uh, and review just the linkages between you know the evolutionary biology of sex and sex roles and how that manifests in gender in a complex uh cultural species such as Homo sapiens. I I found it to be really uh enlightening and enriching.

Um, yeah. Yeah. And I think like, you know, pinning biology versus culture, we always know that that's wrong. But even going through this in as an exercise in working out the details as carefully and frankly in an unbiased way as one can, as Dan did, it was even for me sort of a check on my own. Uh um, you know, I was still sort of pitting biology against culture. And I don't think uh, you know, there's obviously lots of heat and temperature cultural issue wise around gender, especially now. I don't think any I think this trans um uh this is uh orthogonal to all of that, or at least it it it um it rises above all of that because it really talks about how there is learning and there are socials and it is dynamic, but it's

also very much grounded in um uh sex. but what sex is is more abstract, even if it's biological and it's it's it's more weird. It's making the familiar strange. And even both sides in the culture war, well I think what Dan is describing is a far stranger, weirder world than either s either sides um would would imagine. I think John Tuby once said that scientists sort of think that, like, you know, reality is like one side or the other for any at any moment in time in a scientific debate. But reality is far weirder and far far far way further out than even what either side is arguing typically, um, if you look across the history of science. For sure. And because this is such a hot button topic, I feel like Dan has done such a good job of

just providing a level head and a North Star through which to navigate these complicated and messy ideas. And I've I just thought it was a really fun and interesting conversation. I'm sure you'll enjoy it. Yeah, I I loved it. And there's so much to I found it inspiring to just be motivated to think about new things to think about. And it made my life as a kid growing up and just who I thought was a role model just make more sense too. Yeah, totally. Enjoy. All right. Enjoy. Uh Dan, we've already had you on. We're not gonna do any intros. You're well, okay. You're doing you do Asian-based modeling and empirical work based on that modeling. You're doing a new area, uh, you're you're investigating a new thing.

I don't know, maybe you disagree that it's new. Uh what are you doing? Um can you either talk about what you're doing now and why you got into it and or what agent-based modeling is for people who might not know? Um choosing adventure. Uh well and thanks for having me back. Hey guys, good to see you. Welcome. Um really been enjoying a pod. I think it's only two episodes that I are still on my to watch list. Wow, that's good. Nice. We got a super fam. That's cool. Oh yeah, that's great. We gotta give you we gotta give you a sticker. Uh David, you actually this is we're we'll cut this out, but we need to order stickers for HPus. Oh yeah. Yes you do. Yes you do. I'm actually I'm excited to see yeah just from hungry people at conferences and like visits

and stuff like you have you have a listener base and I'm excited to see like at HPES to talk to people see what see what they're thinking about it. Yeah but this is a horrible can't believe I can't believe Dan said that thing he said. But anyway yes enough enough of our um uh sticker procurement um side sidebar. Apologies, Dan. So yeah, either agent based modeling and or what are you doing now? Um I mean, I guess we uh we could start with agent based modeling, I guess. I mean, like the bigger picture thing. So I mean, you say it like I'm doing a new thing. I don't really feel like I'm doing a new thing. Uh I'm doing a thing that maybe people haven't done enough, but I'm definitely like standing on the shoulders of giants.

Um I mean, agent based modeling is just one technique for formally modeling psychological hypotheses. It's a computer simulation technique where what you do is you simulate populations of autonomous entities, which we call agents. Autonomous just meaning they execute behaviors in some fashion. They do so according to decision rules that are inside their own head, uh and they execute those based on information that they capture from their own environment. So you just basically simulate a little kind of proto person with a little simple brain, put them in some kind of environment and set them free to behave however they are going to behave. And this is really good for modeling all kinds of systems.

It's especially good for modeling social systems. So situations where you think you have either some understanding of some bit of psychology and you want to know, you know, if you put a hundred people together that have this kind of mind, what's the sort of social phenomenon that's going to come out of that? Or a situation where you think you have a good understanding of the social phenomenon, but you're trying to reverse engineer what individual level psychologies are sufficient to produce that. So you can, you know, create a little silicone labor laboratory, right? Um And test out a bunch of different individual-level psychologies and see which of them are sufficient to produce that social level phenomenon you're interested in.

I'm not the first person to do this. I mean, I like I have certainly been inspired. Like, I do this in the context of largely mating decisions, although, as you know, Dave, have been branching out into other kinds of social, social partner decisions, things like group formation, friend choice. Uh, but you know, there have been lots of people in the evolution and behavior space that who have used similar kinds of models. I mean, you know, I was super inspired as an undergrad, as we all probably were, by like, you know, Axle Rod's prisoners dilemma tournaments. I was really inspired by um, you know the blind watchmaker. It's blind watchmaker, right? That was one of my favorite Dawkins books. And that's where he has the you know simulations where he's like evolving the different

shapes to look like different animals. You know, the work of people like Jeffrey Miller and Peter Todd back in the day when they were, they were doing agent-based models of great choice. And when I came across that I was like, oh, this is this is how we should have been studying mating all along. Uh normally has done this similar work to, you know, of course like Paul Smaldino, who I know you've had on the pod. Um so there's lots of people over time who have done this. It it just like hasn't maybe caught on to quite the degree that I think it should. And AJ based modeling it's really good for math uh is hard to represent lots of interactions and lots of factors. My understanding is uh in that with an agent-based model,

to some degree you can sort of look at the complex emergent things that you wouldn't be able to tell by looking at, you know, you you you stipulate the rules in each individual agent, but the properties that what ends up happening is not trivial or even predictable ahead of time, which is part of why you do it, is would be my understanding in a way. Yeah. I mean, I I would say a couple of things. One, agent-based modeling is a really good technique if you are interested in building formal models, but like me are not particularly skilled at math. So, you know, you get to instead of having to like, you know, set up and solve like a series of partial differential equations or something like that,

you get to just sort of like, you just have to specialize specify the decision rules inside the agent's head. And then the computer handles the rest, right? Because it's it's all actual interactions between agents. And so a lot of the stuff that like a lot of the techniques that you would need for like uh analytic models um you just sort of like offload to the computer solve it by simulation right um so that's part of why i got into it originally because i was like i want to do these kinds of things but like i never took you know beyond high school calculus so um At the time when I was a grad student, this was like the easiest way in. But yeah, I do think there's virtue in that. Yeah, when you when you're trying to model uh systems that you think are the result of

partially random, uh large numbers of partially random interactions between lots of parts, You can solve those things with math, but you have to make a lot of simplifying assumptions that uh sort of push you further and further away from what you're actually trying to model. And I think in some senses, you know, depending on your question, an agent-based model can be a more realistic representation of the system you're trying to study. That's not always a virtue, but but sometimes I think it is. Yeah. And it forces you to be explicit about things. Like I imagine as you're having a student, for example, come on, they um I imagine it's a sort of thing of like, oh, I realize I have to specify this, I have to specify this.

Things that I didn't even really think about. I have to have an idea about how it works. Yeah, I think that's a major virtue of formal modeling, period. And I think agent-based modeling is like, you know no different from other modeling techniques in that it really draws these things out of you. And you know, David, I know you've done uh also a lot of like formal we were just talking about your hand paper, like a lot of formal modeling as well. I think this is a a huge virtue of trying to formalize your hypotheses, your theories, period. I mean I always tell students, yeah, like I've never sat down to write an agent-based model and not realized an assumption I didn't realize I was making. Um uh this is like just a universal process.

And I think, you know, it depends, it depends on how far you want to get into this, but like so much of our theorizing in psychology is purely verbal. Um, and that's useful. Um It's, you know, honestly, like we're humans, is how we communicate at some level. Our our hypothesizing needs to be verbal. And you can get pretty quickly, you can get much farther into a complicated idea when you're working with it purely verbally than if you try to formalize it at every step. But part of the trade-off with that is like, yeah, you're sort of relying on a lot of your intuitive machinery in ways that you don't necessarily realize you're doing. Uh, and so it's easy to bake in assumptions that you're not realizing you're

making, uh, or uh be overly vague and broad uh in ways that give you lots of degrees of freedom for your hypothesis to appear as though it's explaining a lot of things. Um and when you sit down and you have to say exactly what is the information that is available, exactly how is that processed into exactly what outputs, uh, there aren't as many shortcuts. Um and so you sort of realize, I mean, uh I might have mentioned this last time. Did I talk about this last time I was on the podcast? I don't remember. Um I don't know, I don't listen to the podcast. No, I don't remember what time you were the first guest in, I think. Is that right? It was Jamie, Jamie or you? I forget. Sorry, Jamie, or you better not have Jamie.

I think I think it was Dan. I think it was Dan. Yeah, I think we're and we're getting up on a year. So this is maybe maybe maybe you should be our one year um episode. We'll see how it works out. I don't know. I'll take that. I mean part part of my like whole entryway into doing all this stuff was you know, I was I was at the University of Texas, I was interested in dating psychology. Um, there was some work coming out around the time, including from uh Paul Eastwick, who was also at the University of Texas at the time. He was like the floor below us. Uh, so we were having these kind of like occasional interactions in the building. Um, but you know, there were these papers coming out that were uh we talked about the speed

dating stuff last time, um, these speed dating papers that were like, can we predict what mate choices people will make in a live in-person interaction? And the answer was kind of no. And so there's all this like hullabaloo, like, maybe our understanding of mating psychology is like totally wrong. We need to upend everything we know. Um and I was sort of like skeptical that that was true. And I was like, there's maybe simpler explanations for why we struggle to predict people's choices in live interactions. And I was like, maybe, you know, one explanation is just that I don't even remember where I started. But you know, I was like, maybe one explanation is just that like people want lots of different things, right?

Have multiple different preferences, they've got to balance against them. And so maybe the person you met at the speed dating event who best met your kindness preferences was really far off your intelligence preferences, right? Um, and so if we only look at like how much did the partner you choose you chose match this one preference, we're missing those like trade-offs that people are making across dimensions. Um maybe that's what's going on. Um that feels intuitively true to me, but like, is that really sufficient to produce the kind of data that we're seeing in these speed dating studies? I just finished taking this like Monte Carlo analysis class that I had to take because it was like total serendipity.

I like forgot to register for classes one semester. So the required stats class I needed to take filled up, but they wouldn't make an extra spot for me. And so I ended up taking this other stats class and that like totally changed my life. Um but I had just taken this like Monte Carlo analysis class and I was like, all right, I bet I like I could do one of the models that we do for that class. Like I could just like you know make up some data that looks kind of like make choice data. I could like simulate a process that looks kind of like what I'm thinking, like you know, people making choices across multiple dimensions. I could see what kind of data that makes across a large number of samples, and that would

tell me, you know, is this sort of process actually capable of producing this kind of data? I didn't even know, I had never heard the word agent-based modeling at that time. Like I didn't know that this was a thing. I thought I was just kind of like hacking this together. Um So I started to try it out and I was like, okay, so like I'll simulate some data that looks kind of like made choice data. I sort of know what that looks like from the lab. Um and then you know, I need to have these agents. I didn't I didn't know they were called agents at the time. I was like, I need to have these like synthetic people, you know, sort of meet with each other and they need to like evaluate how attracted they are to one another, I guess, in

order to like make a decision that they want to see each other again. So, oh shoot, like, okay, I can give them preferences, I can give them traits, but like, how do I make them decide from that how attractive they are to one another? Actually, like crap, there's a lot of different ways you could do that. Like maybe I could say the preferences are like a threshold, and if you're above a threshold, like then you're attractive. That's how Peter Todd said it works. But maybe I can make it like a linear model where the preferences are like weights and they sort of like like a regression. They like, you know, multiply the trait by the preference and then add it all up. That's sort of like how David is, David Buss has talked about it before.

Oh man, like there's lots of different ways this could happen. I'd never thought about like this is an important part of the process that we haven't actually described how it works. Uh, and it wasn't until like I had to sit down and like make it happen in a computer that I realized, like, oh, this is a linchpin assumption in how mate choice works, that the entire field is just like mouth studying. So I think that's the that's a key virtue of like forcing yourself to do these things. Uh as you know, Dave, I now have in my lab, sort of like above the main meeting table. Um, I have a there's a famous quote, uh Richard Feynman from his chalkboard at the time of his death, What I cannot create, I do not understand.

And that's sort of like a core lab ethos. Like if you understand how some system works, you should be able to build a copy of it in a computer. David, you were gonna say something. No, I mean it's just um very, very cool stuff. So I'm curious Now that we have this background of uh agent-based modeling and we know your methodology, uh what kind of new stuff do you have cooking in the kitchen with this methodology? And it doesn't matter if it's if it's uncooked, because uh we can come in and we can hope. Well, yeah. Um I feel like I'm being pushed in a certain direction. Exactly. Um well, I mean, so you know, Dave knows. Um uh I've got a big project that I've been working on for the last uh year or so that's that's sort of putting together a collection of agent-based models

on the evolution of sex and gender. Um, this is something that's been on the back of my mind for like several years now. Um just being very unsatisfied with not really knowing how to think about these concepts and not really feeling like there was clear guidance, at least that I'm aware of, in the literature. And so it's mostly just an exercise of like me ironing out my own thoughts on these concepts. And then hopefully other people also find it useful and interesting. We'll see. Um but yeah, I mean I'm on I'm on sabbatical this year. So this was one of my like big sabbatical goals was to finally push out like a big paper that's sort of like me just sketching out my thoughts on like what are sex and gender.

I mean I use it as podcasts, but go ahead. Do do you view them as um so are do you are you analyzing them together or as separate things or as interacting things? So how what are your what are your thoughts there? I was gonna ask you guys what your thoughts were. Yeah, I love this. This is I remember there was like a point, I don't know how it was months ago, where you were Dan was going around and just like very wonderfully serving he's like, what do you think gender is? Like in a like a like you know like a clipboard and a pencil in hand, which I thought was lovely scientific method of just like how would we define gender or is it different than sex? Yeah, what is it? Yeah. Well now that now that we're like local colleagues, Dave knows this, but this is my gimmick

for basically every project. Like you can always predict what I'm working on like six months in advance because at the bar I'll just start asking people, What do you think about X? That's a fantastic way to do it. I love that. Did you when did you start doing that? I don't know. I think it's just uh I don't know. I mean I maybe it was the fundamental attribution error, which is also something I need to write. Yes, exactly. Your favorite. I remember Dan Dan in the hallway being like, what does the fundamental attribution error predict? Does it predict that um internal versus external uh uh will be higher um then like ordinally? Like what's the relationship between those two? Or is it just that um that your internal is going to be lower for another person than it is

for yourself? But it doesn't matter what the absolute magnitude is between external and internal. Great questions, but anyway. I mean, uh, let me throw out a wrong idea. You can tell me w why it's wrong. I mean, I feel like part of the answer might be that like so sex, we already have a bunch of evolutionary models on the evolution of sex. Uh, you know, it's about, you know, uh differences in um parental investment, operational sex ratio, potential reproductive rate, and these things lead to you know the different sex roles we see throughout the animal kingdom. Gender is an interesting thing because um it's a cultural thing. Um I would as a first pass on like part of what's going on, it might be that you have sex,

which represents these, you know, a parental investment reproductive strategies. And then maybe gender is like how you signal this, because it's this thing isn't on the surface for other people to know, like uh necessarily. um it could be how you signal your sex um it so it could be kind of signaling equilibrium um it could be a kind of yeah maybe let me let me start with start with that idea and see I like that idea I like that and you would go straight to signaling that would be I would uh I like that idea and that I think that's actually pretty dang clear. I think you just got like two thirds of the paper. Um I mean l let me ask you, you guys are the host, but let me ask you that. I mean, I've asked David this before, so it's not fair.

So maybe I'm asking just David. Like, what do you think is the average if you ask the average psychologist what sex is? Because you just said like, yeah, we have all these like evolutionary biological models, but like, do you think psychologists are aware of those? Probably most psychologists are. I mean, it depends. I don't know. It depends on how familiar they are with evolutionary psychology in general. And I'll at the end of the day. I think most psychologists would probably go like genitalia. Some would probably go genes, and a much smaller percentage would do gamete investment. Yeah, yeah. And I I think the the the you know first day Trump executive order, I think that's kind of shifted things a little bit.

I was thinking about this project before that executive order and that executive order really blew a hole in in a lot of my thinking. Um and that was the executive order that defined gender by j it didn't define gender at all. Okay. It doesn't quite say this, but it tries to erase gender as a concept entirely. That's my read of it at least. Uh but it defines as being about it's complicated when you get more into this, but it defines sex as being something about gamete size. So I think a lot of gamete size be got onto a lot of people's minds much more than it had been before that executive work. Yeah. But yeah, I mean, my my perspective on this, like, you know, I'm a human mating researcher. Obviously, sex and gender are like really important concepts when you're thinking about you

know human relationships. And so, and I've you know published on sex differences a lot like these are things that I've been forced to think about it at least in some degree and you know coming up as like a psychologist you're sort of aware there are these concepts sex and gender they're supposed to be different things you're supposed to talk about them differently um but when you try to like look at like what is the guidance that you get from psychologists about like what is sex itself and what is gender and why are those different things I just I personally have never deeply understood what I was supposed to be talking about. Um, and when I try to read the literature for like clear guidance on like what are the

definitions or what are our models of these things, I find them really unsatisfying. Uh so if you look at like definitions of sex in the literature, if you look at like the APA's guidance on like what sex is, I don't have it in front of me, but it's something like, you know, sex is a person's biological status. Uh, that's literally that's a near direct quote, and I have no idea what what that means. Um, and it's you know indicated by a number of traits, and they list off a number of traits that's like genitalia, gonads, you know, sex chromosomes, hormones. they list off a number of indicators of sex, which I think is the correct. Those things are indicators. They're not sex in and of itself. A lot of other psychologists will define sex as being like what chromosomes you have, which

I think is crazy. Or like, you know, they'll define sex as being like what sex you're assigned based on your genitalia, which I think is also like, I get what they're going for, but that's not quite right either. Um, but yeah, like the APA's guidance, it doesn't say like what biological status of what. Like I have lots of statuses as a biological organism. Like, what what are we talking about when we talk about sex? Why are those things the indicators of sex? And they don't mention gamete size at all. The APA's definition of sex does not reference gametes at all. Uh and I think that's I think that's a problem. Uh one, because a lot of the things that psychologists focus on as being indicators of or definitions of sex are things that only apply to humans, right?

Like sex chromosomes, they focus a lot on do you have X or Y chromosomes? Do you have certain like you know chromosomal disorders that lead to uh you know different sex development disorders? Uh you know, it's fair enough for psychologists to be overly focused on humans, but like XY chromosomes, that's not definitive of sex across the diversity of life on this planet, right? Like that's how mammals do sex determination, but like birds don't have XY chromosomes, they have ZW chromosomes. Like lots of reptile reptiles don't have sex chromosomes at all. Like they do sex determination just by temperature, right? So like there's a whole wider world of sex out there uh that is not determined by X, Y

chromosomes. You wanted to define it in terms of genitalia. There's a whole diversity of genitalia out there, right? The way that we do genitalia is not the only way to do genitalia. You want to define it in terms of hormones? Like plants have sexes, but don't have, you know, at least hormones like we have. You could argue they have their own kind of hormones, but like they don't have testosterone, they don't have estradyl. so the definitions of sex that like psychologists use are really, really anthropocentric and don't capture like the concept of sex more broadly, which I think is a problem. Psychologists might think it's fine, but I I think that's really unsatisfying. And then you think about gender, um, like what are definitions of gender?

They're also just really vague. Like, you know, they'll say things, people will say things like sex is biological and gender is like cultural. Um, like quite, I mean, that's not uh, I hope that's not a too unfair gloss of uh a lot of ideas, but like one, that irritates me because culture is biological. Uh, but two, it also irritates me because, and you know, people will recognize this, like it's not clear where that boundary actually is. Like there are many things that uh there are ways in which things you might want to call cultural intersect with things you might want to call biological. And so, where should you like delineate the boundary between what sex and gender becomes really unclear?

And it's also just like not clear what gender is supposed to be. Like people will offer definitions of gender that are literally like, you know, aspects of femininity. And it's like, what does that mean? How do I, you know, we're just talking about like what I can't create, I don't understand. Like, how do I build a robot that has aspects of femininity? And also, like, if femininity is just how you refer to like womanness, then like it's totally circular. Like, there's no information in that definition as to what gender actually is. Uh, so I've been trying to sketch out sort of models, agent-based models, uh, but also you know, agent-based models as representations of like theoretical ideas for like

how should we as evolutionary psychologists think about these concepts? Are these different things? And if so, what are they? Um and hopefully that's also useful to like other psychologists more broadly. So I'm I've been writing this like law on paper. It's way too long right now. I I just finished like the very rough first draft on Tuesday. These ideas are still like very actively cooking. cool. Like you know, part one is sort of like this this rant I just gave, like what what are sex and gender in the first place? Part two is kind of like sketching how I think we should think about sex, and part three is kind of sketching how I think we should think about gender. Yeah. And Dan, you gave a a talk uh at the Center for Evolutionary Psychology, and you re- Um,

in addition to reinventing agent-based modeling when you were um back in Texas, Uh in your model, I if I followed you um sort of re-derived the evolution of sex in your model, which was kind of cool. You started very much from like first principles. Yeah. So I'm I'm being a little self-indulgent, uh, to be honest, in in this project, uh, which is partly contributing to its excessive length. And I know reviewers are probably gonna hammer me for, but I just really want to do it. Um so yeah, what I'm trying to do, which I never done before, is, you know, usually when you do agent-based models, you know, you have a system that you want to model, you kind of like write the code for that, you create your little agents, you know, hit run, you let it

run for a while, you know, at some point it stabilizes, or you're just like sick of waiting for it to run. So you pause it, you have your little population, you analyze them, you report some figures and statistics in the paper, and you just kind of like freeze those agents in time forever, put them on a shelf and like never really worry about them ever again. Um in this project, I'm trying to uh sort of work with a set of populations of agents throughout uh a variety of different models. Uh and again, yeah, starting from first principles. So like the first question is like why do we even have sex? So a build an agent-based model that is a uh sort of recapitulation of John Tuby's model of the evolution of sex.

Like this is, you know, this turns out to be a major question that we still don't have like a super satisfying answer to. Like, why do we have sex in the first place? Why are we not all asexually reproducing organisms? Asexual reproduction just seems like the better way to reproduce. Like you don't have to pay all the costs of searching for a mate. you certainly don't have to pay all the costs of like establishing and maintaining a relationship like we do. Um, and you know, if you're a gene inside an asexually reproducing organism's body, like you know, suppose your organism produces two offspring per generation, you know you're getting two copies of yourself per generation, right? But if you're a gene inside a sexually reproducing a uh organism's body, right, every one

of your offspring only have a 50% chance of inheriting you. So like if the reproduction rate is fixed, like your your replication rate as a gene is halved, right? Like, why would you do that? Um, so there are some hypotheses out there. I think this might have been John Tiwee's first paper. Uh I actually I keep saying that, but I need to verify that it's true. But like he offered uh a model that at the time was novel. I think he got kind of unfortunate. He was a little bit unlucky in that W.D. Hamilton proposed a very similar model at the same time. So I think Hamilton's version is the one that like most people know, but I'm gonna credit it to John because he independently came up with it. Um John's idea was that it has to do with pathogens, right?

Like so we're all in this long standing coevolutionary arms race with pathogens. They want to invade our bodies, you know, consume our bodies to produce copies of themselves. That's good for their reproduction. It's bad for our reproduction. So we evolve defenses to try to keep them out. They evolve counter defenses to try to get in. You know, that goes on forever. The problem for us is, you know, as large-bodied, long-lived, you know, slow reproducing organisms, pathogens reproduce much more quickly than we do, right? Like the pathogens that are in us right now can have, you know, orders of magnitude more generations to adapt to our defenses than we have to evolve, you know, new defenses. Um and so John's idea was that the function of sex is, you know, if you're an asexual

reproducing organism, you're invaded by some pathogen, it gets many generations to evolve unique counteroffenses to your defenses. So it's really good at exploiting your physiology. Then you produce offspring, which are genetic clones of yourself, but to the extent that your defenses are heritable, the pathogens that are well adapted to you are immediately well adapted to infecting your offspring. And so you're never going to shake them off. Uh, but if you reproduce sexually, what you're doing is you're taking half of your genes, randomly scrambling them with half of David's genes. Uh, you're producing an offspring that is genetically similar to both of you, uh, but is genetically unique, right?

So the pathogens infecting, you know, that are well adapted to either of you are no longer perfectly adapted to invading your offspring. Um, and so that, John argued, could be a sort of constant selection pressure that would favor the evolution of sexual reproduction over and above its sort of baseline costs. And you can build an agent-based model, a simple agent-based model, where you know simulate agents that are infected by a pathogen agent, they can reproduce asexually or sexually, and you can show if you turn up the dial on like how bad it is to be infected by a pathogen. There is a point eventually where selection will favor sexual reproduction over asexual reproduction. That doesn't have much to do with the rest of the rest of the project, but I just

wanted to build an agent-based model of that idea because I love that idea. Yeah, it's one of the great insights of evolutionary biology that like and one of the great m mysteries of biology is why we sexually reproduce. And it's a particularly, I think particularly satisfying solution to that problem. Uh and it's cool that you know you can cash it out mathematically uh in a model. Um it's also a good example of like asking a question you wouldn't ordinarily think. Like it's I think it's a great example of like even seeing the mystery there at all and not just assuming, well, of course there's sex and just taking it for granted. Yeah. Um yeah. And I also I like it's like a nice example of like the germ like pathogens um

seem like such a window dressing thing from like a modern everyday perspective. Like that's what Lysol's for. Like the idea that like we're built around the fact that pathogens exist, I think to a very modern uh intuition. It feels like, oh, pathogens, whatever. That's like the stuff in the carpet. I do, you know, I build rockets. You know, it's like, no, you are designed around and to defeat pathogens, and you have to take um the uh the world is much more powerful, even the microscopic world is much more powerful than you might think. Yeah. Yeah. So anyway, were you going to continue with the with the journey? So we we got the evolution of sex comes from viruses. What's the next step here? Yeah, yeah, yeah.

So I mean that gives us sexual reproduction. Right. So then what I do is I take those agents that have now evolved sexual reproduction. And we ask, well, where does where do sexes come from? Right? Why, why would there be sexes? You have another sexual reproduction. And so here I'm just like, again, the whole like first part of this paper is like recapitulating models by people smarter than me, but trying to do them in a sort of unified and simplified way. Um so here uh I'm building off of models by people like, you know, Coco and Parker and you know, again, Tube and Cosmics. Honestly, Lita uh was angry at me for not including her model in the in the draft I presented to the CEP, but it'll make it into the paper.

Um in the supplement, probably, but it'll be there. Uh so you know the question is why do we have sexes? And earlier, David, I mean you gave a definition of sexes in terms of like differences in parental investment and operational sex ratio and stuff like that. I don't think that stuff is actually sex either. I think those things are things that importantly follow from sex, uh, but they're not like definitive of what sex is. Um and I think like biologists have roughly known this answer for a long time, but psychologists just don't know about it. So this I see this part of the paper as like just communicating the established biology to psychologists. Uh where does where does sex come from? Well, if you reproduce sexually, the way you do it is you produce gametes, sex cells,

right? Those contain some kind of genetic payload that's like these are the genes that I want my offspring to be constructed from, as well as some just like general cellular machinery and like energy and nutrient stores that like allow the cell to function. Those gametes fuse together and you know if that fusion is successful, if there's enough energy and nutrient stores, if there's like the correct kind of machinery, then that fused zygote can develop into an offspring. So there's some probably like minimal investment that each uh sort of parent organism needs to make. Like if you each put in, there's some minimum investment a zygote needs to be successful. If each parent puts in 50% of that into the zygotes, when those zygotes fuse, you know,

they have some probability of success, right? Um, the way the model works is sort of built off of that kind of assumption. We're assuming there's some kind of like minimum sort of energy threshold that a zygote needs uh to be able to develop successfully into an organism when two gametes fuse, right? Um that said, uh as long as both parents put in that sort of 50% minimum, there's you know some probability of zygote success, but there's inevitably going to be variation in how much parent organisms invest in their gametes, right? And that variation is going to have consequences, right? There's going to be an inevitable sort of quality-quantity trade-off in gamete production. Uh, so I can put more uh energy and nutrients into each of my gametes

that is probably going to increase the likely success of those gametes in developing into successful offspring. But if each of those gametes is more expensive, that means I can't afford to produce as many of them given a fixed energy budget, right? So I can make bigger, more nourishing gametes, but the consequence of that is I'm not going to have as many reproductive opportunities. On the other hand, I can sort of jeep out, I can make smaller, uh, less nourishing gametes. Uh, they're gonna be less likely to develop into successful offspring when they fuse with others, but you know, the upside is by investing less into each gamete, I can afford to make more of them with a fixed budget, right? Uh this is again like very inspired by Dawkins, right?

The the description of the evolution of sex, I think, in the selfish gene kind of follows exactly this logic. He was also sort of basing this off of like Parker's ideas. So, you know, you could think you know, one potentially optimal strategy is everybody just invests 50%. But another potentially optimal strategy is you can make lots of really small gametes, which provide very little investment into the resulting offspring, but give you lots and lots of opportunities to produce offspring. And you can make those gametes such that they specifically target relatively large partners. And that sort of balances the cost and benefits of both extremes of that investment trade-off, right? If you make lots of small gametes that specifically target large gametes, then your

offspring are relatively likely to be successful because they're targeting gametes that provide the energy and nutrients that you're not providing, but you get lots of reproductive opportunities. So you can set up, you know, we take those agents that have involved sexual reproduction. Now they make gametes. They have a gene that regulates do you make big or small ones? They have a fixed energy budget. And they have another gene that regulates like what size of gamete do your gametes target. And you can show, you can start off as population where there's like uniform, whatever initial distribution of gamete production you want, but you can start it off where there's like a uniform distribution of gamete size, kick that off.

And very rapidly, what you see is the evolution of a bimodal distribution in gamete size, where there's some individuals that specialize in producing as small of gametes as they can that target partners that are as large as they can. And then there's other individuals that specialize in producing relatively large gametes, and there's not very much selection on their preferences. And so that that is the evolution of sexes, right? And I think that's how we need to think about sexes. Um, what sex is, is there is this inevitable quality-quantity trade-off in gamete production. This is, by the way, a part where I'm still trying to work out the language of the argument. Uh, but there is this inevitable quality-quantity trade-off in gamete production.

There is, given the details of a particular species like life history and ecology, there are going to be some optima in terms of where you fall on that trade off distribution. So there are isogamous species out there. There are species where the optimum is everybody contributes the same amount to their gametes. All gametes are equally sized. There are there are no sexes, or maybe I would like to say there's just one sex. Um, but there are a large number of species where the optima are just the two extreme ends of that trade-off distribution, right? Like one optimum is produce gametes basically as small as you can, the other optimum is produce gametes basically as large as you can. And those optima is what I want to argue is what we should think of as the sexes for that

species. Dan, can I ask from an intuitive perspective, like already existing as a like a huge human, it feels like the investment in gametes is trivial, even when they're um large. Yeah. Is the idea that phylogenetically, i.e., over evolutionary time, for even small critters, that was that really was a big investment or a big investment difference. And then there's a bunch of concomitant differences that follow from that change in investment, um, such that, you know, the generally speaking, there's obligate investment on the larger gamete producer. Is the is the idea that it's sort of a um a timing thing in the sense of the gamete difference came first and then the the big macroscopic sex differences or dimorphisms came later because from the again from like the the

you know like the the it would be trivial for me as an adult male if I could physiologically to produce an egg like that's nothing right but we're talking about it as if it's this enormous difference but I think that might be a bias perspective. Or like I think you know what I mean, but I mean that's that that's sort of the the question. I think people might be wondering when they're hearing that. And I'm I'm wondering it too. Yeah. Yeah. And I think I think a lot of psychologists are going to wonder that. Like I I want to communicate this to like a broader psychological audience than just evolutionary psychologists. So I do think that's like um that will be an issue. Um and I mean we're also like immediately falling into assuming

uh I always forget the term for this, but like uh gonakora species where it's like either as an individual you specialize in one. Yeah, and that's right. Yeah, no, totally totally that need not be the case. There's lots of things. Yeah, yeah. Yeah, exactly. Yeah, yeah, yeah. Lots of fish where yeah, you you can do both at the same time or either a different point of time. My bias, my bias is showing. My species. Okay. But yeah, I mean, the argument is that we're talking about sort of very uh early evolution of sex. Um at a time when gamete production would be like a more sizable chunk of your overall energy budget. Um and I think what happens is that the reason why this bimodality in uh gamete production matters is because that does set

up other selection pressures for the kinds of things that you're talking about, David, which themselves shape other aspects of phenotype that we should care about, including as psychologists. So those are kind of the subsequent models. So we start with okay, we had evolution of sexual reproduction, we move on to evolution of sex. We do produce these like gonochoric uh agents that I hope that's the right word, uh, that are um uh producing either really large or really small gametes, um, then I take those same populations and start introducing new selection pressures. Uh so now, you know, the the first one I do uh I think is um, you know, we can introduce a gene that regulates like investment in sort of costly mate competition,

you can think of it as. Uh so if you're high on this particular gene, then what happens is like you're more likely to die before you get to produce gametes. But if you do produce gametes, like your gametes are basically at the front of the line when it comes for fertilization. So you get earlier reproductive opportunity, you're less likely to reproduce, but if you do get to reproduce, you get to reproduce first, right? That's distributed uniformly throughout the population, independent of gamete size. But we can ask like, well, what would selection do with that in a world where there are some individuals that are specializing in producing large numbers of small gametes and other individuals that are specializing in producing small numbers of large gametes.

Well, if you're producing small numbers of large gametes, you know, the sex ratio is in your favor. There's lots of small gametes out there. You're virtually guaranteed that like all of your gametes are going to get fertilized, uh, assuming you know an equal ratio of those two morphs. And so, you know, dying before you get to reproduce, just to make sure that you get to reproduce first, like that doesn't have any benefit to you. And so selection should really strongly select against that kind of like risky, costly make competition among those individuals. But if you're a small gamete producer, right, you're there's lots and lots of small gametes out there. They're all competing over a smaller number of large gametes.

So making sure that your gametes get to the large gametes first, like that's a huge bottleneck in terms of your reproduction. And so it's worth trading off a little bit of your survivability to make sure that if you do get to produce gametes, your gametes actually get to develop into an offspring. And so selection should favor those kinds of costly competition genes in the small gamete producers. And that's exactly what happens in the models, right? You can show, you can start off these populations where these competition genes are uniformly distributed independent of uh gamete production strategy, and they very quickly evolve into these, like this sort of sex differentiated mating strategy, where small gamete

producers are investing in costly mate competition, large gamete producers are not. You can do the reverse. So at the same time, introduce what I'm calling like a parenting gene. This parenting gene, if you're high on it, uh, it costs you some of the energy that you use to produce gametes. So you can't make as many gametes if you're high on this parenting gene. Uh, but it increases the probability that if you produce a gamete, that gamete actually survives, right? And so we can again distribute that uniformly throughout the population and we can ask, well, what would selection do with that? Well, you know, if you reason it through, if you're a large gamete producer, right, you have relatively few reproductive opportunities.

The population can only support so many offspring. We stipulate that as an assumption. So if one of your offspring, you know, uh dies because you're you know over carrying capacity or something, like that's a huge chunk of your reproductive success. It's probably worth producing slightly fewer offspring in order to make sure that all of your offspring uh actually survive to the next generation. Uh, whereas if you're a small gamete producer, like you don't even know if you're going to get to produce an offspring at all, right? Uh, because you're the the sex ratio is not in your favor. And if you do, you're probably going to produce a lot of them. So any one offspring is probably a relatively small chunk of your overall reproductive

success. So it probably makes sense to just focus on buying as many lottery tickets as you can and just hoping they all work out rather than you know limiting your reproductive opportunities and keeping those alive. And that's exactly what happens in the model, right? At the same time that selection is favoring mate competition in the small gamete producers, it favors the evolution of these like costly parenting adaptations in the large gamete producers. So these are things that I think have been argued like kind of intuitively. And they've been formally modeled before too, but like we're showing in sort of one model system now, we're starting with the evolution of sex, getting to sexes, and now showing

like sex differences in these other like mating and parenting phenotypes that follow directly from this bimodality in gamete production. Yeah, it's amazing. It's a good example too of people always, you know, accuse, well, not always, that's hyperbole, but often accuse just so stories, right? Where it's sort of like, well, it could have gone either way, but then there's just a post hoc story. And it's a lovely example of you literally are specifying the world as equal, and it's showing that you know, given there's there's deductive logic to differential reproductive success. So yeah. Yeah, and like these things are, you know, I think part of what these models illustrate is these things are really contingent on the assumptions that you make.

So I don't wanna um I don't wanna like uh give the impression that like these sex differences are like the inevitable consequence of having sexes, like these kinds of things are like the essence of maleness or the essence of femaleness. Uh like there are species out there that don't show these patterns, right? Like phalaropes are the the classic example. There are these birds where like it's the females that are big and colorful and compete over access to the males. Um but you can explain those kinds of outcomes within the same sort of model system. Right. So you know, one of the key variables that regulates uh the evolution of these traits that we've already been talking about is the sex ratio.

So the way I'd set up these models is that there's an equal sex ratio, right? There's an equal number of large gamete producers and small gamete producers. But if you change that assumption, if you change the environment, so like I have a version of the model where, like, you know, you can assume this does happen. Like, let's assume for whatever reason the large gamete producers are like really broadly dispersed throughout the environment. Like resources are really patchy, let's say, uh, they're really sparse and patchy. And so the large gamete producers are like spread out over like a really large uh, you know, physical space. So if you're a small gamete producer and you want to reproduce, like you've got to travel

really long distances between these patches, that's going to slow down your reproductive rate, right? Even though even if there's an equal number of large and small gamete producers, if it just takes a long time to travel between patches of potential mates, that's going to necessarily slow down your reproductive rate. If you model that kind of system where, yeah, if you're a small gamete producer, you can produce a lot of gametes, but you can't necessarily use all of them just because of travel time, then you can actually see the reversal of that like competition difference. You can see evolution favoring high competition in the large gamete producers, higher competition in the large gamete producers and the small gamete producers.

So it's not like these things, that's why earlier David, I was saying like I don't think I don't think these things like differences in parental investment, they aren't sex in and of themselves. They're things that follow from sex. So like what sex is doing is it's acting as kind of like a moderator variable that is shaping like how does your ecology lead to particular phenotypes? Cool. Interesting. I mean, it's it's uh the phalarope example is a is a is a lovely example, uh, you know, with you know an exception that proves the rule. Um But it's it's interesting because you see as what what's called a sex role reversal. Um do you think w is it does it make sense to think of phalaropes uh their different parental investment strategies, are is that sex?

Like given that the the males are investing a lot and the females are competing over the males, should we flip it around and say that the males are females in some sense? It's a good question. Uh and this is um this is like kind of where I lost the Center for Evolutionary Psychology. It's like, you know, w what are the implications for all of this in terms of how we should think about like what sex is? I mean, I should say there's another model, it doesn't super matter for this conversation, but there's another model after this where I'm modeling like sex differences in mating strategies. So we add like another gene that's kind of like a mating parenting trade-off gene that regulates like, do you pursue lots of mating opportunities and do not very much parenting?

Or do you do like more, do you spend more of your time on parenting and less of your time pursuing mating opportunities and you know the the sex difference you expect does evolve in this kind of context, assuming you know relatively equal sex ratio. But then if you crank up the the strength of selection on a parental investment, if you're a species like us that produces like highly altruitial offspring that really benefit from having biparental care, then what you see is selection favors sort of lots of parenting and not very much mating in both the large and small gamete producers. And so that's how you can get like the evolution of like pair bonding and biparental care. So yeah, the point is like sex shapes these phenotypes, but it doesn't inevitably lead to

particular phenotypes, right? Um, so yeah, what does that mean for how we should think about sex? Yeah, should we call like the big colorful phalaropes the males? Uh or should we stick to like gametes as defining what the sexes are? I think we should stick to gametes because I think if you want to try to define sex in terms of these manifest phenotypes, like there's all kinds of phenotypes out there. It's like there's not going to be like what is universal across all these species is gamete production, right? These particular phenotypes are really particular to this species ecology and phylogenetic history and life history. And so if we want to be making like bespoke definitions of sex based on phenotypes, like

we're going to need lots of really complicated definitions. Whereas, you know, I think gametizes is like a clean, universal um uh universal, universally applicable definition. Uh but I think if we want to do that, we need to like, there are some things we're gonna have to give up, uh, which is I I think if we want to be serious about uh gamete production as being core to sex, I I think the argument I want to make is that like what sex is, is it's not really a variable that properly exists at like approximate level of analysis. It's not something that's in the organism itself. Uh, what sex is, it's it's a set of reproductive strategies, right? It's these optimal gamete production strategies that selection is always pushing organisms

toward. Uh, but any individual organism is always going to be some kind of noisy approximation of those optima. And those optima are going to have lots of consequences for the kinds of phenotypes that evolve, but there's no one element in that phenotype that is the organism sex. So I think like we find it very useful to perceive the world in a way that like there are males and there are females, and that organism is a male, and that organism is a female. Because sex does exist, it does have these effects on phenotypes. It does structure the covariance matrix of phenotypes such that certain bundles of adaptations tend to go with some organisms and other bundles of adaptations tend to go with others.

So it's like a useful hallucination to see sex in the real world, but like I think it's a hallucination, right? I think if we really want a proper scientific understanding of what sex is, we need to understand that like it's these abstract conceptual optima. It's it's the designs that selection is pushing towards. It's not anything that's in the physical organism's body. Interesting. That's where like people are like no, it's great. Yeah. I mean I think it's all I mean be I mean it sounds like it's something similar. I was thinking while you're talking like other analogies, excuse me. And things like risk-taking or being not risk-taking are also sort of these abstractions. But then again, of course, like risk-taking can vary by situation.

And something like sex, there are the chemaphroditic species where it's sort of, you know, both, but it's not as facultative or situational specific, it seems. I mean, you know, so I think it's it's also a trick of our evolved psychology probably having strong biases about um having entries for sex, basically. Um, I think is probably part of why we're having trouble as scientists thinking about it the right way, as this abstraction or in this in design space, or sort of a set of dynamics in design space. Yeah. And I think there's really good reasons why we intuitively feel that way, which is part of like the gender stuff. Um, but yeah, I think we need to like sort of fight our intuitions. Uh I and I think it's fine to have those intuitions.

I'm not saying like you in your everyday life need to like stop thinking about people as having sexes, but I'm just saying, like, as scientists, if we want to have like an accurate model of these things, we need to understand that like our intuitions are sort of leading us in the wrong direction, and that like really what's going on, you know, I think we should think about sex more in the same way as we think about like something like species. Like it's useful to categorize organisms into species. Like there are real evolutionary histories that cause you know humans to be different from chimpanzees. It's a consequence of like, you know, certain facts about the evolutionary history of those organisms.

And it's a really useful concept. Like, if I say that's a chimp, that tells you a lot about the kinds of things you can expect to find in that organism, versus if I tell you that it's a human, but there's nothing in a chimp that like is its chimpness, right? There's no like one thing you could put your finger on that's like all chimps have this, and everything that's not a chimp doesn't have this, right? Like it's just a useful shorthand for saying like this is the evolutionary history of this particular individual is X. Right. Uh you know, sex is Can take a stab at defining sex? Yeah, yeah. So please uh is this right? It's the set of adaptations uh that tend to evolve as a function of gamete

size. I wouldn't say it's the adaptations. No. Okay. Yeah. I I because then then I think you do get into like, well, phalarope sex is different than like human sex. That's why I said tend to. Not necessarily I mean there are c there are exceptions like phalarop, but you instead of actually patients that tend to evolve as a function gamete size. I really want to avoid tend to oh okay all right. How I Indicated by indicated by the case. Well, I mean, too in most cases evolve, but you know, with some exceptions. No, David, you good job taking a stab at this. This is good. We just we just I want to hear where you're going with that, but that that's part of why I want to push towards like what sex is is these optimal trade off

points in the quality quantity trade off in gamete production. Right? Like no matter what kind of sexually reproducing species you are, you have to make some sort of decision about how much you invest in each of your gametes. That's going to have some costs and benefits in terms of sort of the quality versus quantity of gametes you can produce. There's going to be some set of optima, given your you know the details of your species. But it's often going to be one or two optima. That just seems to be what happens. Um So what sexes are, like the sexes are those optimal trade-off points. So got it. Okay. So the sex is the get is the gametes. And what you're either bigger or small gametes, basically.

Can I can I say can I poke? I think it's what if Dan, it's that it's a sex is a set of selection pressures given uh certain stipulations about the world in this particular case, largely being gamete trade-off. So it's it's the it's sort of it's a set of selection pressures uh driving phenotypic driving phenotypes but it's it's in it's those set selection pressures specifically predicated upon gamete trade-offs I'm a little I I think that's I think that's a lot closer. I'm and that might be right. I think I want to say it's more it's the design that selection is pushing toward, which might be equivalent. But it it's it's the the optima toward which selection is pushing. Those are the sexes. That's what I'm saying.

Yeah. Yeah, yeah, yeah. Yeah. Yeah, that is very uh uh it's like physics where you it's not intuitive. We don't have an entry uh intuitive entry for an object in that kind of model, right? Like that's great. Yeah. Sex is the set of adaptations that specialize in either large or small gametes. I'm a little bit nervous. I mean, maybe I'm just a little bit nervous to define it in terms of the adaptations because, like, how do you then define what those adaptations are, especially across species that do it in lots of different ways. Um the uh the specifically the gamete uh is the is the adaptation. So the argument. Yeah. Or would it be the argument that there's evidence of special design that those phenotypic

features are the results of the gamete associated selection pressures, which is maybe just a way of redescribing what you're saying, Dan? Um it could be. And then I think it gets really what is sex becomes super contentious. Um it becomes like super contingent on like states of evidence for like special design for certain things. Um are you thinking of it just the simplest possible way as just big or small gamutes? Um kind of. I mean, I I I do want to be open to the possibility that it's not just big or small. Like I do think it's it seems to be very often those are the kind of like most stable optima. And certainly there are isogamous species where it's like all one kind. But in principle, like we can imagine we go to like an alien planet, and for whatever

reason, you know, the optima are like three there's three gamete sizes that kind of work out, or four gamete sizes that kind of work out. I don't know. Um or maybe cheap, cheap or expensive gametes. I I you know, I think that's how it seems to often work. Um but that's why I'm trying to like push it out to like it's not a particular sex is not a particular manifestation, right? Like what sex is is just like whatever are the optima for this particular species, right? Like and very often the optima are yeah, cheap or expensive or big or small. Um, but in principle it could be something else. So it's the equilibrium for selection on gamete size. Exactly. Exactly. So if it was a yeah, so yeah, that's great.

But so if there was a species that was produced big uh gametes or small gametes facultatively or every other day of the week, it would be one sex. Because that's the universal um you're talking about like a sequentially hermaphroditic species. I think I would say there's two sexes there. It's just you like individuals alternate between the sexes. But that's like that's maybe really getting into semantics. Like I don't know. I should have I we added out my last comment because David had a nice wrap to wrap that up nicely. David, can you repeat that? What was that again? Sure, sure. I believe um how I framed it's the it's the equilibrium uh of the selection pressures on gamete size. Yeah. Yeah.

Yeah. It's there's that is specific to the equilibrium. Yeah. It is the speci yes that is specific to a species. Great. Put that on a hallmark card. Okay. So now we've figured out sex. Okay, hold on. I I think there's one more because what I was thinking, so there's also sex roles, which is I think different from gender. So what I was thinking of in terms of you know potential reproductive rate, uh, you know, uh parental investment, stuff like that, that tends to follow from uh gamet size, but doesn't always, as in the case of the fowler opes. Um and so you have sex, which is the equilibrium selection pressure on gamete size, and then you have the sex roles, which tend to follow from that equilibrium, but do not always

follow. And and sex roles would be something like parental investment, intersexual competition, stuff like that. I think that's right. Yeah. Okay. And I I think they do follow. It's just um it's not a one-to-one mapping of like gamete size to a particular role. It's just gamete size dictates it's very likely that there will be some kind of sex role. And gamete size dictates is one force that dictates what the sex roles will will look like. Got it. Okay. So we got sex. We got sex role. Now let's move on to gender. Yeah. So why what is gender and why do we have it? This is like so that whole part, the whole first part is just like recapitulating what evolutionary biologists have already said for the most part.

Maybe that equilibrium thing is like that's my like tinfoil hat addition. But gender, I think, is something that has been way under theorized. In part because it turns out it's like a really new concept. I was not really aware of this. Like people, the like the first reference to gender as a concept separate from sex was from a like a psychiatry paper in like 1955 or something like that. Like people have not really been thinking about these, at least in technical ways, people have not been thinking about these as separate concepts for all that long. And, you know, like I said, the definitions are really vague and I think unhelpful. So the definition I've been trying to sketch out for gender is like what gender is, is

a sort of set of folk theories about how sex manifests in your particular ecology. So why do we need these folk theories? Why can't we just know, like, if sex leads to sex roles, why can't we just be like born knowing, like, okay, these are human sex roles? This is my sex, I'm in an activist role. Um, I think it has something to do with like our species' unique distribution and foraging niche, right? Like, so we are really widely dispersed hyper generalist foragers, right? Like humans live in all you know inhabitable ecologies on this planet plus Canada. And New Jersey. And New Jersey. And New Jersey. And we, you know, wherever we are born, wherever we migrate, we just like figure out how to extract food from that ecology.

And we tend to specialize in the things that are hardest to extract, right? We tend to specialize in things like, you know, big game, which is very, you know, calorie and nutrient dense, but really doesn't want to provide those calories and nutrients. And so it's uh very difficult to extract. Or we, you know, rely on these like difficult to extract gathered goods, like you know, like tubers and honey and stuff like that, that requires like great skill and technology and time to extract, and also potentially great skill, technology, and time to process into something that we can eat. But in exchange, you know, we get these like huge energy packages that you know both require but also feed this like big expensive brain that we have, right?

So this is all like you know, known. I'm like standing on the shoulders of like anthropological giants here. Like this is all very much inspired by like you know, Kaplan and Lancaster and Gillener Toto and stuff like that. Um so I think we have this unique foraging niche where you know if you're a gene inside of a human infant body, right? Like your organism opens its eyes for the first time. It doesn't know where on this planet it's been born, It doesn't know like what kind of like, you know, how do my people make their way in this world, right? Like what kind of food are we primarily fishers? Are we primarily large game hunters? Are we, you know, industrialized Americans? Are we primarily eat like processed fast food?

Like, where does my food come from? I think is a major question. And I think that's a major question because it that shapes all kinds of other aspects of life, right? Like how you subsist in the ecology where you've been born, like that shapes how you spend your time, right? It shapes what technologies you invest time learning into. Um, it invests like what skills do you uh try to build up over a lifetime? It shapes, you know, how do you live? Like, are you sedentary? Are you very mobile? What kinds of relationships do you have with others? Do you have like you know, risk pool sharing, or do you have like uh just like you know individual reciprocity? Like all of these things are shaped by the ecology that you have been born into, heavily by

like the ways that that ecology you know affords you getting your food. Um and so humans have to be able to learn how to do all those things. They have to be able to learn like what do I eat in this ecology and how does that shape uh all the rest of it? Uh from the Martin Daly interview you guys did, uh He was talking about like one of his first uh postdocs was in a lab that was that was like a socioecology lab. Um which I don't think I'd actually heard the name of that discipline before, but I looked it up and it's like you know the study of how like ecology shapes you know social structures. And I thought like that's perfect. Like I think like because of our unique foraging need niche, Like humans basically have to

be like intuitive socioecologists, right? You need to be able to like look around in the place you've been born and ask yourself like, what's my ecology and how does that shape how people live their lives here? Um, I think that's just a question that humans need to ask. Um go ahead. So just uh sorry, it's an interesting thought you just had. Um so socioecology tends to study the relationships between ecology and social structure that emerge over evolutionary time. But what you're saying is that um those relationships between ecology and social structure um in humans uh emerge over the span of uh you know a few generations rather than evolutionary time because humans learn how to adjust their social structure to their

ecology. Um so it it's a similar discipline in that it's still mapping ecology to social structure, but the mapping arises over ontogenetic time as opposed to evolutionary time. Yeah, yeah, that's exactly right. Yeah. And I, you know, I think like this is what, you know, forager studies is all about. This is so much of what like cultural evolution is all about. Uh, I think like these things are well established and it's like good literature to draw in there that like, yeah, like the ways in which people tend to live their lives are heavily structured by in large part the ways that they get their food, which is structured by the like the apologies they happen to live in. And there are, you know, I think the adaptations, the psychological adaptations underlying

these things, this is where evolutionary psychologists could do better, like mapping out how those adaptations actually work. But there are clearly adaptations for like, you know, learning how to produce uh you know technologies that are going to be useful in your particular ecology, or you know, learning how to flexibly deploy uh you know different kinds of foraging strategies or different kinds of cooperation strategies depending on you know the structure of the environment you happen to live in. So I think, you know, the variations in our ecologies generate this variation. So now there's an adaptive problem. If you're a human, you need to figure out like, what is my, you know, what does life look

like in my in my population? This is something you can maybe help me like workshop this, because I've been trying to figure out a way to describe this. Because I feel like as I describe this to people, what they take away is like, oh, so you're saying gender is about learning how to forage. That's not really the argument I'm trying to make. What I'm trying the argument I'm trying to make is variability in foraging generates variability in all these other aspects of life. That's what you need to learn. Like you need to learn how to live your life given you know the way that uh foraging generates variation. So I'm kind of been trying to think of a name for like what's that set of things that like

vary potentially as a function of like your ecology or foraging. I'm currently thinking like life way as sort of like an homage to like Kelly's foraging spectrum, the life way of hunter gatherers. Like as a baby human, like a pressing problem for you is like you need to learn like what is the life way for people like me? Like what foods do I eat? What technologies do I need to master? What skills do I need to master? How do I need to relate to other people so that I can you know successfully capture energy from this environment and turn it into offspring, basically? Dan, can I ask uh this is uh uh a question that's also though maybe uh helping maybe in that articulation. Um is one way of putting it, because David, I always think about David's notion of status

games, is one way of thinking about it is that the status games that are in a community are not arbitrary, but they're sort of tethered to um resources and therefore the ecology. So that so that's that that's another way I think of putting the same thing, right? Like that's sort of there's this tethering between the status games that can exist, what counts as being good or bad, or what the games are. There can be some proliferation, some elaboration, but ultimately there's some feedback from the ecology and the resources because ultimately that's really where the rubber hits the road. I think that's exactly right. Yeah. And I think importantly, you know, we could be the kind of species, we could be multiple

different species, right? And I think you know that's what racists think, right? Like we could do like there's an African morph and there's an Asian morph, and like, you know, uh, it could be that like you're just bored knowing, like, oh, I'm Asian, therefore, I eat these kinds of things, therefore, these are the sort of status games, but that's just not how human genetics works, right? Like, there's huge gene flow across human populations. And so there is this variation that's tethered to ecology, but it's also importantly, because there's constant interaction between different human populations, there's constant gene flow across different populations, the that variation is unpredictable from the genes

I view. It's again, I'm going back to Dawkins. It's like it's the Abra Andromeda problem that he talks about in Sylvis Gene, right? Like the genes need to build one psychology that can work in any kind of ecology. It doesn't know what that ecology is going to look like in advance. And so it needs to build a learning program that can like look around in its ecology and say, well, this is how my people seem to live. So I'm gonna do the same. Yeah, that's amazing. Yeah, and that really suggests like a very rich um, this is very much about the develop development then. And it in and it's um and it really suggests like a very rich, unexplored psychology of resources, uh, ways of making a living, status.

And these are not the right words because we don't have the the words in our natural language, but it it suggests that um, and I think some developmental psychologists have studied these kinds of things. But yeah, but I think um this is amazing because yeah, it sounds like it's really opening up this idea of yeah, it's intuitive. I I hear what you're saying, that it you're sort of looking at like what's the way of making a living, broadly speaking. Um, but fleshing that out in a way that you could implement it in a little critter, uh software critter, is uh that's cool. That yeah, that's that's a hard, yeah. But a good it's fruitful. Like you I can I can start imagining ways of doing that. Yeah.

Yeah, I wouldn't say it's unexplored. Like I feel like developmental psychologists have been studying this and anthropologists have been studying this for a long time, but maybe just not organized as like this is the sort of unified, this is the elephant that we're all trying to describe, you know. Right. Um so I mean, where does gender come from in this though? Like I'm just talking about like, you know, learning life wave. I think um because of sex, because sex does shape phenotypes in ways that we've already talked about, uh, those phenotypes that follow from sex are going to have implications for that life way, right? And we know this from lots of anthropological research, right? Like one major aspect of this is like some individuals have high levels of obligate

childcare in the form of internal gestation and uh and lactation. That's going to make certain forms of foraging more difficult in some ecologies, right? It's like difficult and dangerous to large game hunting in some ecologies if you have to breastfeed an infant on demand. And, you know, Kaplan and Lancaster and colleagues, they have like these really nice models that show like because large game hunting is so skill intensive, like and because of the amount of time that like, you know, a sort of like natural fertility uh woman would spend uh, you know, pregnant or breastfeeding, um, like it's just not worth the investment to try to build up the skills to do large game hunting. Like it's it's a much better use of your time to focus on other resources uh than it is to

try to uh invest in those skills. And so they argue like this is why you get sexual divisions of labor that you see across, you know, many foraging societies. Like it's it's just a more efficient arrangement for everybody if you males focus on these activities that are not very compatible with childcare, because they have lots of time to build up the skills to make that a worthwhile investment. And females invest more in other activities that are more compatible with child care demands. sex is going to shape optimal lifeways, but it's going to do so in unpredictable ways across ecologies, right? In some ecologies, maybe sex won't matter that much, right? Like if you're primarily a fishing population, like anybody can fish a lot of the time.

You know, it depends on the details of your ecology, is the point. So it's very likely if you're, you know, uh an infant human, it's very likely that you live in a place where there's sex differences in your this kind of like life way. uh, but the nature of those sex differences is going to be opaque to you. Uh, so part of learning this life way is my argument is like probably a more efficient way of learning this life way is if you do have this kind of conceptual slot already. If you're born with an expectation that there are probably different lifeways based on your gamete production strategy, um and you can learn different lifeways from different models, then you can deploy a life way that like works best for you.

And so I think what gender is is like your models of like what is the life way in my particular ecology separated by you know sex. Awesome. And we can model this. Yeah, yeah. Interesting. Um think presumably there's also um a a psychological dimension to the downstream consequences of of gamete size. You you mentioned that, you know, among males, they might accept a a lower survival probability in exchange for greater reproductive uh uh potential. Um and so that might manifest at the level of psychology with greater appetite for risk-taking and dangerous activities among men, that they're willing to trade off some survival risk for a large reproductive payoff potentially in status or whatever.

And so that could be another variable that affects these life ways. And so it's not merely that men are, you know, don't have the physical gestational costs that allow them to hunt big game. It might also be that they just have more of a tolerance for taking dangerous risks. And women just are not as enticed by the opportunity of hunting a dangerous animal as a man would be. And so there are psychological Predispositions that cause certain cultures to develop lifeways as well as morphological. I think that's right. So certainly I don't take me to be arguing like selection stops the neck, right? Like this isn't uh this is honestly I've sort of backed into social role theory in the process of doing all this, but like this is totally different from social role theory.

But I do think sex does structure psychological adaptations. And those psychological adaptations, in addition to the physiological adaptations, are going to be part of what shapes sex differences in these kinds of lifeways. So I'm focusing on like foraging and like gestation just because I think those are like really clear. Like we definitely know those are uncontroversially exist and uncontroversially shape these aspects of lifeways that have important consequences for other things. But yeah, like I'm confident that there are psychological adaptations that follow from sex that also play into these kinds of processes as well. Right. So in the Oh, sorry, Dan, go ahead. Go ahead. I was just gonna say, so like from a learning perspective, I for me it's helpful to think

about the ways to get a robot wrong. So it'd be something where, like, if you built um a learning robot that was like, you know, I'm a baby, look around at the other babies, and um figure out the baby that like gets the most attention. And that's like what I want to grow up to be. Like that's a bad design, right? So you have to like build in that there's like life or like there's there's uh developmental morphs. And so part of you know, maybe I should compare myself to other babies too, for the purposes of currently being a baby, but you can't design me to sort of aspire to be the best baby 40 years from now that I can possibly be, right? Like that's that's a that's the wrong sort of social learning target to be.

So you have to build somewhere de facto into the system like age morphs too. And the argument is just like age morphs, then it's also sex morphs to the extent that there are that there are differences in the ecology or in your psychology. So that if you built a robot that didn't see sex or gender, you might sort of mash things up and you in and it wouldn't work right. Maybe I think that's exactly right. So I mean it might be helpful to talk a little bit about like what the models look like right now. So you know the way this the way I've been doing this, you know, we it was take that population of agents where we evolved like uh pair bonding and biparental care. So now like these agents are forming into monogamous pairs.

And what I then expose the agents to is basically like they have like a household that they need to manage. Like they need to gather a certain amount of resources and they reproduce together proportional to how successful they are at gathering those resources. The resources they need to gather, I'm just focusing for simplicity on like three things. They need to gather, like, there's two nutrients that, like, I'm calling them like nutrient P and nutrient M for like plant and meat. And then they need to do a certain amount of parenting. And they have a fixed time budget that they can use to like gather those resources. At the end of that time budget, they reproduce based on how successful they've been.

They can get those. It's like the model's complicated. I'm not going to be able to describe it today. You can read the paper when it comes out. Uh, but it's a Markov decision process where like there's a bunch of states the agents can be in, there's a bunch of actions they can choose. Those actions probabilistically generate those resources. The important thing is like there's 11 actions they can take. One of them is like parent your offspring. The other 10 are just different food resources that the agents can gather. Those food resources generate a certain amount of those of those nutrients that they need. And so they need to learn like what's the right balance of parenting and gathering these

other food resources to like maximize my household's sort of store of resources. And so we can create different environments for the agents by manipulating properties of those food resources. So we can make some of them like in this ecology, you know, meat number one is super productive, meat number two is not very productive. But in that ecology, it's meat number two that's super productive, and it's meat number one that's not very productive. We can manipulate like, are these dangerous? Like is there some probability of injuring yourself if you pursue this resource versus that resource? Um we can manipulate like what's the success rate? Like is this a really risky resource like large game hunting or is it a really stable

resource like you know like lots of gathered resources? And then the critical one is we can manipulate is foraging this resource compatible with childcare or not? So in some ecologies, maybe all the resources are pretty compatible with childcare. Like anybody can forage anything relatively easy. And so you know the sex differences in life ways are going to be relatively small. But in other ecologies, like maybe one of the most important resources you need is really not compatible with childcare. That would be like a sort of large game hunting kind of situation. And in that case, like the optimal life way is for small gamete producers to focus on foraging that resource and large gamete resource producers to specialize in producing other

resources, there should be a sexual division of labor. So we like throw these agents into ecologies that vary unpredictably on these different dimensions. And then we give them different ways of evolving, sort of like I'm calling them foraging strategies. Maybe I should call them like life ways. But basically there's like three kinds of agents in the first model. There's ones that just have like a genetically encoded foraging strategy. Like they're just born like, I like meat one, I don't like meat two. And that doesn't change throughout their lifetime. There's other agents that like have a psychology that does reinforcement learning. They pay attention to how successful have I been at pursuing this resource in the past

and that adjusts how they behave in the future. And what you show is what you can show what I find in the models is if the ecologies are really predictable, right? So like there's no gene flow, there's like different environments, but like offspring are always born into the exact same environments as the parents. The environments never change over time. So this is sort of like the racist model of the world where there's just like, you know, stably different kinds of humans. You can know what sort of environment you're going to be born into. And In that kind of environment, selection favors a genetically encoded foraging strategy. Like it just makes sense to be born knowing like my people eat fish, I'm going to get ready

to forage for fish. But if environments are unpredictable, if there's some amount of gene flow, like there's some probability you might be born in a different environment than the ones that your parents were born into, then selection no longer favors that genetically encoded strategy. It favors a psychology of reinforcement learning. It favors a psychology that tries to figure out um uh you know how to make a way how to make a a living in the ecology where you happen to have been born. And it specifically favors uh a psychology that engages in what's called what's called model-based reinforcement learning. Um so these agents, they not only pay attention to like, you know, how much reward did I

get when I tried that action last time, but they also slowly build up a model of how the environment works. So they can make guesses like if I do this action, this is what's going to happen next. And so they can use that to like make even more accurate predictions about how to uh how to behave. And I think it's that model, uh, that model of how does my ecology work, that I think is where gender really exists. So that like that first model is just showing like you know, variability in foraging will generate selection favoring this kind of learning psychology. The next model is okay, we have if we set up environments such that like they're random, but there's some probability that a sexual division of labor

is going to be optimal, like there's some important resource that is not compatible with childcare. We have a population of agents who all do reinforcement learning, but now we add in a new mutation that regulates whether they do social learning, right? So the original agents are just learning based on their own experience. But I mean, exactly as you were saying, Dave, now we have agents who, when they're born, they can look at the adults in their ecology, they can observe their behavior. And instead of just learning based on their own behavior, what they can do is they start to form a model of the environment based on observing adult behavior. So they get to, they're sort of like they're getting a head start on learning their life

way because they're using the information that already exists from the adults in their ecology. And you can do that in one of two ways. Like you can do that ignorant to sex. So you can just say, I'm going to look at any adult, any adult behavior is useful. I'm going to use that to train my model, the environment, and then I'm going to use that model to guide my own behavior. Or you can do it in a sort of sex-aware way where you can train different models for adults of different sexes. And then you can use whichever model is applicable to your body. Right. And what you see is as long as uh at least some ecologies have a sexual division of labor, at least in some ecologies, there are some resources that are not

compatible with childcare. Uh selection favors the learning agents that learn from adults in a sex differentiated way. Like they're born with an expectation that there's some kind of sexual division of labor and it's my job to figure out like what is my specialty in this, in this economy, right? And so I think that is how we should think about the psychology of gender, right? Like what gender is is your model of how sex manifests in the particular ecology where you happen to have been born. Cool. And it also suggests that there's a representation of what you are. That that that is uh almost by definition not the adult version yet. Right. Yeah, I mean, so it implies a lot, it implies there's a representation of what you are,

which I think is what people often talk about as gender identity. To your point earlier, David, there has to be a representation of what other people are, which will involve probably some amount of signaling. And I think that's what people often talk about as gender expression. In the model, those are always perfectly aligned, but of course, in the real world, they're not going to be. There are going to be people who, you know, identify with a sex that is different from what their physiology is. And so I think that gives you a model of like how can you start to think about like transgender identities, for example. So I like, I think you get a lot of things that we think intuitively are involved in the

concepts of sex and gender out of this, as well as, you know, a model of like what is it that kids are actually trying to learn and how are they trying to learn it? Right. And I, you know, it's like to your earlier point, like these ideas do exist out there, right? Like there's lots of work in developmental psychology on like gender essentialism. Like kids, you know, essentialize gender as this category and they like attach all kinds of ideas to it. There's like developmental intergroup theory, right? Which suggests that like kids learn from adult behaviors about like what categories there are in the world. So because adults treat gender as a category that matters, kids learn to do the same. And I think like it's it's similar to those in that like it's saying, yes, that it's right,

but it's not just like an arbitrary categorization that kids are like picking out of the world. Like, no, like kids for very good reason are born with an expectation that gender is probably a thing that shapes lifeways. And their job as kids is to figure out like in my particular ecology, how does how does sex manifest? Like what is the right way to be male in this place I happen to have been born? So I think it like aligns with those things, but it's sort of like it connects them with what we know about our species' evolutionary history and it gives us a computational model of how that psychology is actually working. Yeah, and and like for my own expertise in thinking about alliances or even race, right?

People always talked about like, well, there's categories that are inferentially rich, like age, sex, race, slash, alliance. But people always acknowledge that learning about alliances and or or race or whatever arbitrary sets, as Jim Sidanius called it, was different than gender. Um, but in but nobody there was no theory of that. It was just sort of like we all have that intuition, that's what the data shows. And I like that this gives teeth to that, which is it's the difference between there's a difference between alliances and the which which morphs or morphs or ways of making a living, to what degree are those different? Those are different features of the world, right? Like the particular alliances, um, and who's on whose side is different than this is

the way you kind of make a living given the the nature of the ecology or the environment, which again is different than age. They all interact, right? Um to some degree. But I I really like that, yeah, that it it it gives it computational teeth and you could start to think about like how a robot would be built. Um it it creates um a task, you could do a task analysis of of this. Yeah. Yeah, yeah, yeah. Um good. Uh I I want to explore um how intrasexual competition fits into this. Because this is a I I feel like this is probably a really important uh process uh by which sex and gender manifests. Um so you know, this the sexes choose among the other sex whom whom to mate with. Uh within sexes, you know, uh people choose their friends and and allies and they compete

with other members of their sex for status, for access to to mates. Um and I feel like this is probably playing a big role in in in gender expression, gender signaling. Um it seems to me that if if there's a particular sexual division of labor such that say men invest in big game hunting and women invest in, say, gathered resources, um it seems to me that women would have an interest in mating with the best big game hunters, and men would have an interest in mating with the best gatherers. And and there may even be some invariances as well, insofar as the small gamete sex is going to uh invest more in dangerous risk taking and have a a lower um or have a higher appetite for dangerous risks. Or they might be bigger and stronger and more they might be more geared toward fighting

with other members of their sex. You know, in cases where there are extreme sex roles, the females will prefer the biggest and strongest males because they're going to give them the most successful offspring. or in cases where there's more intersexual selection, they might prefer the the males with the most colorful tailer tail feathers or the the most eye-catching song and dance routine. Um males similarly might prefer the females who uh m uh uh have cues of higher health or fertility or going to provide more investment for their offspring. So it seems like there there is selection within each sex to signal a kind of prototypicality. Where if I'm a male, I want to signal that I am a pi I'm the biggest and strongest, most

competitive male, or the male with the the most colorful tail feathers. If I'm a female, I w I might want to signal that uh I'm going to provide the best investment and nourishment for uh offspring. I'm the healthiest and the most fertile or or whatever. That might be occurring at an abstract level across sex roles, across species, but you know, it it'll obviously manifest differently uh in in a human setting with you know sexual divisions of labor. But it seems like sexual uh or gender prototypicality is is playing a big role here and signaling that you know I'm a particularly good big game hunter, I'm a particularly good at the at the tasks that that my sex uh pursues in this life way. I think that's right.

Um so I mean, uh, there's lots of things to pick up on there. Um I I think one is um, so I've been talking about this in terms of like what the gender concept is, is like you learning the best sort of life way for you. It's like a very selfish, kind of individualistic way of thinking about it. But yeah, like if we think that's gender, you could ask, well, why would we ever care about other people's gender concepts? Which we very clearly do, right? Like so much of like sort of contemporary sort of social issues and so much of moralizing is like moralizing how other people think about or enact gender. And so, like, why would we do that if what gender is is just like you figuring out how to live your life in your ecology.

Um, so I do have like the last model in the paper is sort of like starting to get into those kinds of questions. Like, why would we ever care or reason about other people's gender concepts? And I think there's lots of reasons why we should, if that's what gender is, and you've touched on some of them. Like there are lots of cases where how other people think about gender is relevant to your interests. I think one is, yeah, that you hit on already is uh if we are, you know, we're sexually reproducing species and we're also these like, you know, biparental, uh, you know, relatively lifelong monogamous uh maters, like it really matters if there is like a sexual division of labor, or just even if they're even if there's not, it just like

really matters. If I'm going to be attaching my reproductive interests to one other person, uh, it really matters that like we have compatible understandings of like what our roles are. Um, so that we both are like, you know, investing in the right like skills and uh um activities that are going to be like, you know, maximizing our our shared reproductive interests. And so I'm gonna have yeah, strong opinions that like whatever my concept of gender is, I'm gonna wanna make sure that my mate has a concept of gender that like plays along with that one and that like they're enacting the role that I think they should be enacting in a relatively effective way so that we can maximize our shared reproductive interests.

So I think that's one case. Like in the mating market, like you're gonna wanna make sure like whoever you mate with, they're on the same page as you in terms of like how do you make a living in the psychology that we that we that we live in. Uh, I think there's another case where like you should really care about your offspring's gender concepts, right? Because if your gender concept is shaping your life way, like you as an adult invested a lot of time and effort and energy learning a good life way. Uh, like you want to make sure that like your kid does the same, because if you know there's no value in you successfully reproducing if like your kid then learns like the optimal thing to do is like eat rocks and hit my head hit my head with a hammer, right?

Like so you want to make sure that like they're learning a good lifeway as well. And so it makes sense that parents would have strong feelings about like what is their their children's models of their environment and therefore like what is their children's concepts of gender. And does that align with you know what you think is a correct life way for the ecology that you live in. and that I have a model of that, right? So we can go into this population uh that has evolved this like sex differentiated learning strategy. We can manipulate like are environments relatively stable over time so that like you know, the kinds of environments where parents tend to live in are similar to the kinds of environments where offspring tend to live in, or do they change relatively quickly?

Like, you know, is the optimal foraging strategy pretty different from generation to generation? Um, and we introduced, I can introduce this gene that's basically like if you're high, I call it like a gender policing gene, for lack of a better word. We're basically like agents that are high on this gene, what they do is parent agents who have already learned a good policy, what they can do is they can sort of intervene in their offspring's behavior. They can prevent the offspring from enacting their own policy and they can just force their offspring to do what the parent would have done, which I think intuitively we feel like parents often do to their children. So you, if you're high on this gene, you sort of prevent your offspring from learning their

own policy. They just implement your policy. The advantage of that, if the environment is relatively similar, you've already learned a good policy. So your offspring are gonna, they're not gonna waste a lot of time sort of learning and exploring and trying out things that don't work. They get the benefit of your knowledge. They're going to implement a good life way right from the beginning. They're going to outcompete those other kids that are left to like try things out on their own. Um, that works if the environment is stable. If the environment is unstable, though, that's a terrible strategy, right? Because the environment your kid has inherited is different from the one where you trained your policy.

And so letting them learn their own policy is probably going to be the more optimal thing to do. And yeah, you see, depending on the structure of the environment, when environments are stable, selection favors this kind of like policing strategy of like impose your gender concept onto your offspring. But if environments are unstable, then selection favors parents that kind of back off and let their children learn their own their own strategy. So I think like these are the kinds of situations where agents would have strong feelings about and would try to intervene on the ways that other agents think about and like express their gender concepts. And if you think about like if you take that for granted, right, okay, parents have an

interest in making sure their offspring develop good gender policies. Well, then if you know that the way that your offspring are developing these gender policies is by looking at adults and building a model based on adult behavior, well then that gives you reason to care about what other adults are doing, right? Because you want to make sure your kids have good targets around. So that's why you would care about like the gender policy, even of like unrelated adult strangers, right? Like, I don't want you projecting a what I think is a bad life way, because my kid is going to look at you and that's going to become part of their model. And then they're going to start behaving in suboptimal ways.

And so all of this just creates grounds for like all kinds of conflict between individuals about like what is the proper way to think about, you know, gender, think about life ways in neurocology and to enact those things. And I think that potentially explains like so much of like culture war about sex and gender today. Yeah, it's interesting. I mean there's also um, you know, parent offspring conflict I bet plays a role. So there's research showing that uh parents and children diverge in their uh preferences for what for for mates for their children. So uh child children tend to prefer uh tend to care more about physical attractiveness, say, than their parents do uh when choosing mates for their for their children.

Um and parents care more about cultural similarity than the children do. Yeah. And you could uh imagine that you know cultural similarity is uh is going to have better inclusive fitness benefits for the parent uh because that's going to allow for uh more um for less investment, number one, less investment from them that's going to be required because they're gonna be better at at investing and and pooling labor. And number two, perhaps, uh uh better cooperative opportunities with with their kin. Um and so parents want more cooperation among kin than the kin wants. You know, the kin want twice as much stuff as their sorry, uh kids want twice as much stuff as their sibling, but parents want to uh uh distribute investment equally.

So, all else equal, the parent is going to want um the kids to choose mates that are more culturally similar than the kid will want. And so that could play a role in here as well, where you know uh uh parents are going to want um adherence to whatever the life way is uh among their kids more than their kids will. I think that's exactly right. Um yeah, I think even in cases where you think the environment is the same, yeah, parents and offering don't have overlapping interests. And so there's room for conflict over like how you think about and manifest gender, even in that scenario. There's also like, I mean, so my former postdoc, Elizabeth Aggie, has like done a lot of that work. You should have her on the pod, she'd be a great guess.

Um uh, but I, you know, I I think that's exactly right. I think it's a really good point. Um, it's also like I just for simplicity, I've been modeling this as if like environments are either stable or they're not stable. But the reality is like environments vary in how stable there are. And that's something you need to make an inference about, right? Like when you're born, you need to ask yourself, like, is my ecology going to be the same as the adults in my ecology? And different people can make different inferences based on the based on the kinds of information they're getting and also just like the kind of body they particularly have, right? Like it might be the environment is the same, but like what worked for my dad just won't

work for me because my dad and me are different people. And so my dad might be getting information, like, my policy is still going to be really good for my son. He needs to behave this way. And I'm getting information, like, nope, I gotta chart a different path. Like, what worked for him is not gonna work out for me. And so that can generate sort of intergenerational conflict, even in the same ecology, just because like there's no uh there's no way to know who's right, right? Both agents are having to make inferences based on uncertain information, and they might just make different inferences about the best way to go, and that's gonna cause conflict. And there will be adaptations, there will be selection for adaptations that sort of tip

that conflict in either direction. That reminds me, Zach, we had Zach Garfield on, and he talked about how the son of a really good hunter didn't want to do that. I think it was a Hunter. Um, and and he just and Zach characterized it as the kid seemed to know soon and early that he didn't want to do that. Wasn't his life way. And this is not even per se a gendered one. Uh I mean, it might, it might be also, but but it was more just sort of that, you know, this was not, I'm not built. Um, it it I'm not built to do the trajectory that I see my successful dad pursuing. Um, and I love that about this too, is it hooks up to this sort of niche construction that the whether it's gendered or not, there's also this

thing of looking out of the world and saying, How am I gonna make a living and succeed. We've talked about that a lot in a lot of different ways with a lot of different people for a lot of different reasons, like parenting, status schemes, resources, um, and thinking about like what what am I particularly good at, you know, and figuring that out. And I that's really cool that it hooks up to that too. Yeah, I think that's exactly right. Um yeah, and that's there's a yeah, there's an individual difference angle that I'm also sort of like under exploring. But yeah, there's just like there's a you can think about like what is the kind of optimal set of life ways in your ecology, but like those are not those are almost guaranteed to not

be optimal for you, right? Like at the end of the day, you're still gonna have to tailor your life way to you in particular, and that's gonna differ from how everyone else sees this just because of you know reactive inheritability, right? Like the body you happen to inherit is gonna shape what is the the right way to do things for you. It's almost like oh sorry David, I was just gonna say um it's almost like gender is almost like gamete based personality where it's sort of like you know it's like yeah yeah I like that way of dis of of describing it and thinking about it. And and of course, your personality is is affected by life way and and you know the different socio-ecological niches that can be filled.

Um that's interesting. I I feel like there's there's something missing here in that gender is not merely a dichotomy but a continuum, in that I can think of men who are more or less masculine, I can think of women who are more or less feminine. Um and to some extent that is a pref a mate preference. You know, men might prefer more feminine women, and women might prefer more masculine men. And so how do you how are you thinking about the uh the continuum of masculinity or femininity and how that might play into mate preferences, status games, uh uh, etc.? I so it's a great, it's a great question. And I think one sort of upshot of thinking about gender in this way that I haven't been super clear about either is, and this comes out of the models as well.

Like uh gender, yes, like I think we should think about gender as being these kinds of folk theories about how sex manifests. And yes, sex tends to be, you know, binary if we think about it as like there are these what sex is or these optima. And in our species, there tends to be, there seems to be two. Uh, but that doesn't mean that gender roles themselves need to be binary. Um and you can have, depending on the structure of the ecology, like there can be multiple good strategies for a large gamete producer, or there can be multiple good strategies for a small gamete producer. Um, and people can specialize in like different ways of being a man or different ways of being a woman. Um, so you don't need to think about gender as being binary.

it thinking about gender as being folk theories of sex does not imply that gender itself has to be binary. And even if it were the case that, like in a particular ecology, you know, there's just one sort of relatively good strategy for being a male, like because of how all this works, like every individual is going to learn a slightly different, you know, perspective on what that is, right? Because every individual, they, you know, they're forming their model based on, you know, somewhat random observations of adults in their environments. So all of their models are going to be slightly different variations on that same thing. Um, you know, part of doing reinforcement learning is like there is randomness inherent in

the learning because you have to try out different things. And you're going to, there's going to be randomness in terms of the reward those actions give you. So some agents are just going to, you know, there's going to be some variation in how rewarding they find different actions. And so there can be multiple optimal strategies. And for any given optimal strategy, any agent's sort of like assessment of what that strategy looks like is going to vary to some degree. And that variation is also going to feed forward into like the role models that those agents provide for the next generation. So I think thinking about gender in this way like naturally gives you all of this variation in gender concepts.

uh, you don't have to think about these things and they don't emerge as being sort of like strict binary, like this is what a man is and this is what a woman is. And then, of course, these things are all ecologically specific. So, like, what is a man in this environment is going to differ from what is a man in that environment, because it's all about learning how to sex manifest in this particular ecology. Uh, so there's, you know, it also handles and also gives you this like cross-cultural variation in like what is masculinity, right? Yeah, and I think the non-binary part also would explain the continuum, right? Because if, for example, let's just say risk taking, if there's uh uh just because of

the the gamete selection pressures, uh there are uh versions of stable equilibria of male, we'll just call them male and female for exposition, male uh type stable equilibria of ways of making a living. on average, some of those are going to be really high risk taking, but maybe not everyone. Maybe there's a lower risk-taking male stable equilibrium. But so therefore, you're going to get a continuum of risk taking among men, and same for women. It might just be that there's a really low risk-taking stable equilibrium for women, but there's another stable equilibrium for women that's not as averse to risk taking. And so consequently, because it's not binary, but there's actually multiple stable

equilibria within each uh class, that's exactly what what you would perceive as a continuum, right? Yeah, yeah, yeah, that's exactly right. And yeah, you we can we can model this if you want. Like you I've set up models where um, you know, let's say there's two resources that are compatible with childcare, but they both are really skill intensive, like they both require a lot of time invested to get good at foraging them. And so you can't specialize in both of them. You set that up for these agents, they naturally like pick one or the other to specialize. So you get sort of two kinds of life ways that emerge within a given uh a given sex. And you can imagine building more complicated versions of that exactly along those lines.

Like, you know, one of the dimensions is like how dangerous is this food to forage? Maybe how dangerous it is depends on some aspect of your phenotype that varies relatively continuously. So yeah, maybe some individuals are willing or able to take on those risks. And so that becomes like a viable life way for them. Others are not willing or able to take on that risk. So that becomes a viable life way. And yeah, so there's like a, you know, maybe spectrum of how much people are investing in those uh uh those particular lifeways. Or maybe there's like a balance that you can do. Um, there's a range of different lifeways based on you know how you mix that activity into your sort of portfolio of overall activities.

Um so yeah, these things don't need to be like rigid categories. Uh they can emerge as as uh uh more variable, the more variable structures that they actually are. Yeah. And it suggests, like, I mean, you mentioned you alluded this to the mating thing, but I think a lot of partner choice, whether it's like choosing a mentor or a friend, like I think we've all had, I've had the experience of you just meet somebody, and there's something about the way they're constituent, they're they're they're made up of that just really it's very like compelling. And you're just like, oh, I want to really learn from this person, or like this is makes a really good friend. And I and sometimes it's maybe that you're the same.

You know, like you detect that there's something weird about me and I don't know how to make this weird thing in me work as to make a living. And then you find somebody who has that same thing in you. And then you you can sort of learn from them. You know, like I think that that's like a lot of the tropes of mentorship or like I found, I found my tribe, you know, is sort of the the the the that trope. And then there's also just like the complementary, complementariness too. You know, somebody else is good at this thing that I'm bad in. Um, and this suggests that both for like finding mentors and friends and mates, um, it suggests there's this very rich psychological like landscape of these folk

representations that we might not there might be literature on that maybe not be framed this way, but it really suggests that we probably like a lot of friendship choice, like if the friendship literature, a lot of it is sort of like similarity writ large. But I suspect it's actually a lot of like what we're talking about, but what you're talking about here. It's about like ways of making a living status game kind of things. Yeah. Yeah. Yeah. I think you're right. The paper's already like 15,000 words. So this is and this is definitely a BBS paper. It would be a sh a sin if this doesn't get into BBS for people who don't know because it gives people it gives people a uh a venue many many areas to comment on because i as a

development partial developmental psychologist I find this like it connects so much area so many areas of developmental psychology that have been unconnected with so many other areas that are not developmental psychology like mating or parenting or friendship yeah good i'm glad I'm glad to hear that Ten out of ten. Well yeah. Looking forward to the BBS paper. Um from your lips to Paul Bloom's ears. Yes, if you're listening, Paul Bloom, you gotta you gotta accept this one. Um Pins off twenty twenty six. All right. Well, I mean, we're we're uh about two hours now. Yeah. Um is there anything we we should have asked you that we that we missed, Dan? Oh shoot. I I knew that question was coming and I still failed the preparedness.

We don't maybe we don't have to ask it for repeat. I don't know. Should we have asked you that? Should we Oh should should we have asked you that? Maybe a better question. Maybe a better question. Is there anything we asked you that we shouldn't have? That should be that should be for repeat, repeat guess we should what shouldn't we have asked you about? I mean I since we're cut we know we got our one year anniversary, like maybe we should consider whether we should continue that tradition or not. Yes, exactly. Oh you know, I do like the tradition, but I do feel like it's got a low hit rate. Like, how many guests are you gonna be able to do that's true completely something? I mean, I like the idea of you have to pick something we shouldn't have asked you.

That's I like this. This is like the signal detection thing. We shouldn't just err on the side of the things we missed, the hits there or the things we should have Or what's what's the dumbest thing we said? What's the thing you least like about this? Who's the dumbest guest we've had on, Dan? And and you can who's the dumbest person on this podcast that we've and it can include the hosts. That's the Uh I'll tell you my answer offline. I think it's me. I think it's me. Yeah. Well this is yeah. Uh do you have an answer? No, and I I think this has been a it's been a really helpful conversation for me just to try to like air all these things out as I'm trying to write this paper. So uh I've got things to think about.

Awesome. Yeah. Yeah. Every time and Dan, every time I hear this, I think there's so much developmental uh stuff to do on this too. So it's just I I this is super exciting. I just this is awesome. Yeah. I think I hope I hope people see it as like yeah, I I I'm worried it will come off as like we've been saying this all along. Like, why are you just coming around to this? Uh I need to figure out how to pitch it as like, yeah, like I we're trying to sort of join your stream rather than kind of like steal your thunder. Yeah. Well, I think connecting it up from like a your I mean, and what I like, I don't I don't think it's self-indulgent. Uh like my will editorialize for a second. You said earlier it's self-indulgent to start at like the evolution of sex, but I actually

think it's crucial in the sense of you're doing this form like really methodically, logically. And if there's one thing evolutionary behavioral sciences is about is sort of you're just like, I'm checking my assumptions. Does this follow? Without sort of grasping at some phenomena. Like obviously the phenomena is a thing to be explained, but it's not like all of a sudden you're like, well, we gotta, we gotta like kind of, you know, round up and round down to accommodate this known phenomena. You're like deriving it from first principles. And of course, the fact that you're recovering it, that's a virtue, not a vice. Right. So I think I don't think it's self-intentional. And I do think it's a contribution to provide a causal framework attached to all these

different pieces. Yeah. It's a synthesis. It's a helpful. I think that might be one way to frame it. Yeah. Um you're you're bringing all this stuff together in one place when it hasn't been together in one place before. And I think that is is doing a service to the field. Yeah, I like that. A synthesis of gender. That's actually a lovely way of putting what you're doing. Yeah. Uh yeah it sounds like uh the name of a book I might find in a humanities section, which is fine by me. Or or a robot factory that's gonna build a robots with gender. Yeah. Uh no, I that's helpful. I need to I need to remember to frame it that way. Maybe that'll be the sub subtitle of the paper. Synthesis. There you go.

Awesome. Uh well, um on that note. A sexy synthesis. I will. All right, Dan, awesome. Good to see you. Um and uh good luck. We look forward to hearing more about this. Thank you. Thank you guys for having me back. David, um stickers. Oh yeah, we gotta print out stickers for HBES. Do you you think stickers is the way to go in terms of the the stuff to hand out? Yeah, because you can put in stickers on anything like a a mug or um That's true. Uh uh a fanny pack. Yeah. Not an animal. You wouldn't want to put it a lapel pin can s is sort of limited to clothing, I guess, which is the the downside of a pin. Were you thinking of lapel pin? Uh uh just throwing out possible options. Um That's pretty highbrow.

Yeah, it is a little highbrow. Yeah. Although you could use a um what are those things that are like the stuffed animals that all the kids are hanging off their backpacks? It's little boo-boos or something like that. Oh yeah. Kids are into the should we do evolutionary psychology like boo-boos? I don't even know what that would be like. I don't even know how what that would look like. I'm imagining a care bear. Do you remember the care bears where they had like symbols on their bellies? I'm imagining a a care bear with like a brain and the logo on its belly, but I don't know. Maybe it would be a brain. I don't know. Yeah. Not the best. I feel like that would take a lot of funding to uh to figure out.

A lot of time and effort. We're we're we're worried about the ability to get stickers done. So yeah, that might be yeah. Yeah. Okay. Uh anyway, on that note. Uh yeah. Support us. If you want those laboooboos, you gotta pony up on Patreon. I think we'll I think this will get posted before HBest. So we look forward to seeing everybody at HBest. And um yeah, maybe we'll um maybe I'll bring a lavalier mic and we can do a few like short recordings at HBus. Oh yeah, that'd be fun. Oh yeah. So sort of like man on the street kind of thing. Yeah, exactly. Yeah, exactly. Assault people with a microphone and ask them. Wasn't there a shtick like people would ask people like what's the capital of the United

States? And people would like, I don't know, Albany, you know, like like wasn't that a thing it was like Conan or J Leno or somebody? Oh yeah, classic bit. Yeah. Yeah. Yeah. Should we should we ask gotcha questions to everyone at HBS? That seems like a really um mean thing to do. So probably yes. No, I don't know. I was I was joking. No, I don't I don't know. Yeah, but we can um it might be a nice way to like feature students or just like the uh diversity of people attending HBus. I don't know. Yeah. No, it'd be fun. Yeah. We'll see. Good idea. Yeah. All right. We'll probably edit this out because we won't well, we might do it. Yeah. All right. Uh thank you everybody for listening. Um, and uh we'll see you next week.

Yep, see ya. Pontum Pont, Pantum Pum two, Pant, Pontum Pan P, Pontem T, Canton.