#936 David Pietraszewski: Alliances, Intergroup Conflict, Politics, and the Science of Psychology

RECORDED ON DECEMBER 11th 2023.
Dr. David Pietraszewski is Assistant Professor in the Department of Psychological and Brain Sciences at the University of California, Santa Barbara. He is an experimental psychologist who applies evolutionary theorizing to Social, Cognitive, and Developmental questions. He is particularly interested in characterizing the psychology of multi-agent conflict and cooperation (“coalitional psychology”).

This is our second interview. You can watch the first one (Modularity of Mind, Coalitional Psychology, and Politics), here: https://youtu.be/5v1qKir4nKE

In this episode, we start by talking about alliance membership and race categorization. We discuss intergroup conflict, peace, and reconciliation from an evolutionary perspective. We talk about how the human mind represents social groups. We discuss what cognitive processes are. We talk about the free will debate. We discuss David Pinsof and Martie Haselton’s alliance theory of political belief systems. Finally, we talk about difference detecting mechanisms, and the influence they have on the science of psychology.

Time Links:
00:00 Intro
00:58 Alliance membership, and race categorization
21:54 Conflict, peace, and reconciliation from an evolutionary perspective
34:20 How the human mind represents social groups
40:50 What are cognitive processes?
50:53 Why free will exists, and how to understand it
1:07:29 David Pinsof and Martie Haselton’s alliance theory of political belief systems
1:24:30 Difference detecting mechanisms, and the influence they have on the science of psychology
1:43:08 Follow Dr. Pietraszewski’s work!

---

Follow Dr. Pietraszewski’s work:
Faculty page: https://tinyurl.com/ye2am2v7
ResearchGate profile: https://bit.ly/3u4yRyN
Twitter handle: @DavePietrasz

Transcripts are automatically generated and may contain errors

0:00:00 Ricardo Lopes: Hello, everybody. Welcome to a new episode of the, the Center. I'm your host as always Ricardo Lobs. And today I'm joined for a second time by Doctor David Pietrzyk. He is assistant professor in the Department of Psychological and Brain Sciences at the University of California Santa Barbara. He is an experimental psychologist who applies evolutionary theorizing to social cognitive and developmental questions. And as I mentioned, this is a our second interview on the show. So I'm leaving a link to the first one in the description of this one. And today we're talking about social categorization, alliance membership and detection, intergroup conflict. David pins off and Marty Hazleton's paper on their alliance theory of political belief systems and some other topics. So, Doctor Piotr Zeps, welcome back to the show. It's always a pleasure to talk with you.

0:00:54 David Pietraszewski: Thank you, Ricardo. It’s a, it’s a pleasure to be here.

0:00:57 Ricardo Lopes: So, one of the things we talked about in our first conversation was social categorization and alliance detection, the alliance detection system. So how does it apply specifically to race categorization?

0:01:14 David Pietraszewski: Sure. So this is an idea that people noticed that people have been categorizing people by race, since people started measuring categorization processes starting around the eighties. And for a long time, people sort of took for granted what the mechanisms underlying that categorization were. When people started thinking about it, there were a few options. One was, well, maybe there are systems in the mind for categorizing people by race. Another possibility is that it’s a by product of other adaptations. And for a long time, people couldn’t get people to not categorize people by race. And you can look at that implicitly by looking at things like patterns of memory errors. So you show somebody a face in a statement, you see, do they mis attribute what one woman said to another woman, what one man said to another man to the degree they do that they have categorized by, for example, sex. But you can do that for things like age sex, race or any other social category. And what people found was people always categorized by age, sex and race. And the question of why do they categorize by race. And so what I’ve done in some of my work and we talked about this a little bit last time you and I talked. And I’ve got a number of papers now along with other colleagues showing that the system that’s categorizing people by race. What it seems to be is a system for tracking alliances. So the idea is that race as we think about it is very superficial skin deep, adaptations for things like climate exposure, maybe pathogen exposure or, and, but those superficial features are not things that the mind is necessarily designed to attend to. You can kind of think of it as analogous to like, you know, differences in the way people’s hair color look or, or other superficial traits. Like we don’t seem to think of those as racial per se. So the question is why do people have this experience pop out of this constellation of features as race. And so the idea is that there’s not a system looking for those physical features, but instead systems that learn about which physical features predict how people interact and get along. So the idea is that there are adaptations in the mind for categorizing people by their alliances and that that system when it grows up in a culture where certain physical features predict how people interact and get along, like where do people live, who has access to what resources things like that, then that system will pick up on those physical features. And we can test that by showing that when you show that race is not predictive of how people interact and get along. So it’s crossed with alliance membership that you can get categorization by race to go down and to be inhibited in the cognitive system. So basically, the mind is treating race as a socially learned, socially constructed category. That’s about anticipating patterns of affiliation. And that’s basically the, the, the relationship between this alliance system and race.

0:04:20 Ricardo Lopes: Mhm. A and even when it comes to that social categorization of race or in terms of race, I mean, w we have to also to keep in mind the fact that the way we understand the concept of race and the different races we recognize, I mean, it’s very much as some sociologists have been arguing, it’s very much a result of social and cultural inputs because even for example, in the us, the different race categories or how people from different origins have been categorized in terms of their race has changed over time.

0:05:05 David Pietraszewski: That’s right. You can look historically across cultures within cultures. So, you know, turn of the century being Irish was literally perceived as a race in the United States. You know, if you look in places, you know, where there’s been histories, sort of genocides or ethnic conflicts, the physical distinctions between the groups. For example, in Rwanda, would were actually just completely made up, there were in fact no basis in reality. And so, it’s pretty clear, the historical record, the sociological record is fairly clear that if you look at it, the the alliances, the affiliations, those dynamics dictate which physical features become important. You don’t need physical distinctions to have alliance differences. And, and those can be projected or hallucinated on. And there are many physical differences that are not perceived as alliances even developmentally, which is a beautiful test of this idea. Even Children learn labels before they learn who those labels apply to. So if you show who to an adults look like people of different races, Children might not understand that those are significant differences between those people. They won’t notice it, they won’t remark on it. But if you give Children labels that refer to count classes of, of people, so these are the XS, these are the Ys Children immediately understand that it’s relevant and important. They remember it, they recall it. They seem to infer its social significance even before they understand what criteria place somebody into those categories. So they really seem to go label first, alliance first Not which Q happens to go along with that. Yeah. Mhm

0:06:58 Ricardo Lopes: And we, when we are dealing with alliance membership specifically, does the way we consider the categorizations in terms of race change is there and the effect there.

0:07:11 David Pietraszewski: Well, so it depends what you mean. There’s a couple of ways to answer that question. One is that so traditionally, when people had been measuring social categorization by race, they hadn’t been providing additional alliance information. And we’re gonna use alliance and coalition sort of interchangeably. We don’t really know all the details. But essentially, if you don’t give people a specific acute social alliance, like this is a group and here’s another group and this is their relationship, that kind of thing. If you don’t provide that people will default to what are in their culture, the most relevant or important alliance cues. And so for many of the studies conducted over the years in social psychology and these social categorization studies, the most relevant alliance dimension in the stimuli was race. So race was always a feature of categorization in these studies, even when it wasn’t relevant. So even when people are talking about what are they going to do on a summer day or what is it like to be at this university or something like that? So, so in that respect, racial categorization seem pretty immovable. But the crucial thing is that for the alliance tracking system, part of its evolved function, one of the things that has to do to to make its living as an evolved system within our evolved design is to be really good at tracking alliances, storing instances of alliances, but then also figuring out which alliance is relevant for this situation because there are many, many alliances. But in any particular situation, only a subset of those alliances are relevant for predicting how people interact. And that’s the point of the system, it’s to predict behavior, guide behavior. And so, one of its design features needs to be the ability to take all of the alliance cues that it is stored, use some sort of criteria for establishing which of all of those is most relevant in inhibiting the use of the others. And so what we have in our studies is evidence of that process where if you show this previously diagnostic alliance to race, but you show people now a specific context where there’s people of different races, but there’s new better alliance membership information. So here’s one group, here’s another group, here’s the way they’re interacting can be positive, it can be negative, it can be coordinations in any case. What we have very good evidence for is that the system can very easily take the racial information, inhibit the use of it categorization in many cases, goes completely away or it’s reduced by, you know, somewhere between a third to two thirds. And in some cases, it’s not even happening while the new alliance cues are are very quickly picked up in ca base the basis of categorization. So the answer to your question is the relationship between race and alliance really depends on if there’s other better new alliance cues present in the stimuli. And if that’s true, categorization by race will go down. But if there’s no better alliance information, then categorization by race will be a prepotent sort of default alliance cue on the idea that in the world it is a relevant alliance cue and the idea would be that to the extent it’s not relevant out in the world, it would be no longer be quite as prepotent. And in, in principle could be, it could be something else. It wouldn’t have to be those particular cues you wouldn’t even need, something like race per se. And in, you know, in certain places, that’s, arguably it’s, there are other sort of most prepotent dimensions like, you know, the tribe down the river or something like that. It’s not quite like the way we think about race. So, yeah,

0:11:10 Ricardo Lopes: so while you were talking, the, there were some ideas crossing my mind and I’m going to say something and if I’m incorrect, please say so. But I was thinking that is it correct to say that perhaps when it comes to making alliances, how we think about other people in terms of being to protect allies or not? I mean, of course, we look at people and we see how they are different from us in several different traits and some of them might fall into different social categories related to age, gender, race or any other sort of social category. But, and when it comes to looking at people and seeing the differences that exist between us and them, and thinking about alliances, I mean, the fact that we see that someone might have, I don’t know a different gender, different skin color or something like that. And we think about that in terms of it being a good or a bad trait. It is, it has a lot to do with context. It’s context dependent, right? Because it is probably, at least when we are dealing with alliance membership, it has a lot to do with the people having traits that would be beneficial to us or not. That, I mean, does this make sense or?

0:12:41 David Pietraszewski: Yeah, I mean, I think, I think if you think about it from sort of a first principals perspective, and when we’re actually doing some modeling work, so Dan Conroy Beam, who is another faculty member at UC Santa Barbara and the psychology department, We’re doing some modeling work on this. But basically, if you just imagine a world where like a simple world where we just sort of think of traits as just ABC de, you know, all the way, all the letters in the alphabet and you just sort of think, well, what would an organism or an agent? How would they carve up the world? How would a well designed organism, carve up the world again, well designed meaning, you know, naturally selected. So what would be the design of this alliance tracking system? And what I would want to do is it would want to look out in the world and see, well, you know, what likely is the basis of cooper operation that yields benefits to me. So, in a sense, you know, in a world where there is no conflict, there’s no reason to exclude anybody from your pool of cooperators to the extent they’ll give you resources to the extent that they can generate benefits, you know, gains and trade things like that. And so, there’s no intrinsic reason in the design of the system to sort of discriminate along any of those cues and there’s no reason to not have as many people in your cooper effort as possible. Now, you can add additional principles and you can say, well, look as the pool of cooperators gets bigger, it’s harder to track people’s investment in the common good. And so you get, you know, the free write of problem, those kinds of things. And so you can start to see where theories of sort of what’s the, the size of the group or identity that you belong to. So their ideas in social psychology about sort of optimal distinctiveness. And I think that that’s a, a very valid and nice way to think about things. But I think it’s underwritten by this sort of deeper principle of the trade off between sort of being able to ascertain what’s happening and monitor for it and be be benefited by a collective versus the strength of the collective so a very large collective may not care much about me. Give me many benefits, but it’s very powerful and there’s a bit of a trade off there. So then you start to get into a basis of discrimination where you want, the people who are closer to you maybe care more about you to be the people that you’re most in a group with. Or if they’re sort of different groups, like the, you have one closer group and then one larger group, right? And then you add something like conflict where people have conflicts of interest, same people. The same people want the same things, right? So you want something, I want something and we both can’t have it so excludable goods, rival goods or, you know, we something we have a misunderstanding or just a conflict. You know, I take something here as you take something online, you add that in as well. And then all of a sudden it becomes in agent’s interests to have other agents really differentially exclude some in favor of others. And so you, you add all of those dynamics and, and then you can see where you very quickly get certain cues become the things that mark the people that you feel positively about versus the people you feel neutrally about versus the people you feel negatively about. But the important thing is the underlying calculation in principle is based on these costs and benefit resource flows, benefit flows, the superficial cues take up a lot of our attention. So, is it about skin color? Is it about what people dress like? All those things sort of take up our attention? Because those are the things we’re conscious of because they actually distinguish between these alliances. But the, but the idea is that below that superficial variation are these very common deep principles in that from an engineering perspective, from the perspective of describing the psychology and then also the dynamics those those dimensions of costs and benefits are really what’s important for the evolved psychology.

0:17:14 Ricardo Lopes: And just to close the section on race categorization, is there any evidence that it can be lowered by cross categories that have nothing to do with alliances?

0:17:28 David Pietraszewski: Hm Yeah, this is a good question. So I’ve been doing these reducing racial categorization studies with alliances for many, many years now. I mean, it’s been a decade and really beautiful data, lots of different kinds of alliances. So positive alliances like two groups, not in conflict, but each group highly cooper within each group. And then other situations where there’s antagonism between groups and then other situations where each group is just has opinions or are coordinate. All of these things are print in principle alliance cues and all these things produce this expected effect of reducing categorization by race. And I should point out not categorization by other social categories like sex or age, which is important because it shows that it really seems to be something about race and alliance, not just that alliance is just cause you to not attend to any other cross dimension like age, sex or race. It’s really specific to race. That’s actually important just as an aside because most theories of social categorization, at least those that were most prevalent at the end of the 20th century and early 21st century can’t account for that selective pattern because on those theories, age, sex and race are all chronically activated categories. So anyway, that’s, that’s sort of an aside, the question you’re asking is what about non alliance categories? And the reason why you’re asking that is I realized after all these beautiful studies had been done that there was one counter hypothesis that I never tested against and that could falsify the entire account. And so I thought in a, in a very pian way, I, I needed to try to falsify my own theory. And so, so I just mentioned categorization by race would go down when categorization by sex and age, for example, would not. And specifically when it was through alliances. So when we would show alliance stimuli in the in the studies, so there’s one group and another group in their alliances that specifically was causing categorization by race to go down and not sex or age. And we were using that to infer the activation of the alliance system. But what I realized was that it could just be that categorization by race is just more flexible now, that would not be consistent with the past work. But at least it would be consistent with my pattern of results. Selective decrease in was compared to age or sex under the exact same experimental conditions. So what I wanted to do is do the exact same experimental structure where instead of the social categorization dimension being alliances, it was something that there was no alliance information. And so what I used is whether you’re older or younger siblings. So some people were older siblings, some people were younger siblings, they were visually marked. So whether they were older or younger, they had different color t-shirts, the conversation was about being older or younger sibling. I made sure to minimize the alliance cues. So there was no information about the older siblings or the younger siblings sort of teaming up together, roughly speaking. And so this was really important to show that categorization by race would not be selectively reduced by these non alliance stimuli. And fortunately for the theory, you know, and you know, just to editorialize for me personally, it was nice to see but you know, what, what matters is is what the data tell you. That in fact, categorization by race was not affected at all by crossing race with this older or younger sibling dimension. But people categorize very strongly by this older or younger sibling dimension as well. So, categorization by both sex and race were unaffected by it. So it’s a nice example of taking a hypothesis and not only confirming it with a number of studies, but then also once you’ve confirmed it actually starting to try to actively falsify it. And this was a nice example of failing to falsify the hypothesis. So the hypothesis is getting more, more support. So that’s, that’s the the answer for why we ran those studies and why categorization by race is not affected by these other things. Yeah.

0:21:55 Ricardo Lopes: So changing topics, I would like to talk a little bit about conflict and peace from an evolutionary perspective here. So, first of all, from an evolutionary perspective, what drives intergroup conflict?

0:22:12 David Pietraszewski: Yeah, I mean, what drives intergroup conflict is in some sense, it’s very very simple things. So things like safety, access to resources, worrying about your safety, worrying about access to resources, right? So it can be offensive and defensive. And it’s about social relationships. So so people have things that other people want, you worry about other people taking things that you have. And especially as we talked about before there can be conflicts of interest between individuals and once that occurs, those individuals are incentivized from an evolutionary perspective to get allies in support of them in, in comparison to the their, the people they’re competing with. And so that sort of sets up a dynamic where each agent is incentivized and will do better to the extent that they can shore up allies in defense of themselves against the people that they have conflicts of interest with. And so, and so, yeah, so, at every scale of social organization, the idea is that really what’s driving conflict, including multi Asian conflict are pretty simple, fundamental things like concerns about safety concerns about resources. You know, and like, you know, resources, it’s easy to sort of, I, I often worry when we talk about resources, particularly sort of in the evolutionary social sciences, people sort of think about, you know, like luxury watches and things, but it’s, you know, you know, it’s like food and like, you know, things you actually need to. And in, in a world where there’s not police or military and even in a world where there is police and military, there are many people who don’t have access to those crucial resources and access to those resources, the function of social relationships and social power. And so that’s what conflict is about. All the way through it still is today. You know, even even with organized large scale social world institutions. There’s still massive differences in access to resources that are based on power. And that’s true for, from an of all perspective at every scale.

0:24:32 Ricardo Lopes: But OK, so that’s about conflict. But then do we also have a good understanding of perhaps the psychological mechanisms that drive peace and reconciliation between groups?

0:24:47 David Pietraszewski: Yeah, I mean, I would say we don’t have a good understanding of psychological mechanisms generally at a certain level of specificity. But I think in terms of broadly how they would be working or what they’re attentive to then yes, we do. So at a very high level, the answer would be yes. And it’s, it goes back to what we talked about earlier, which is that from a first principal’s perspective, there are always benefits to expanding the pool of cooper operation if people are willing to trade off. So there are things like, you know, variation in resources. So if you get and I enter in a relationship, and we’re both, you know, hunter gatherers and hunting is high variance returns, maybe you’re successful one day, I’m successful another, it has nothing to do with how hard we try. It’s just sometimes the animals near you, sometimes the animal is near me. We can buffer that variability in the world by entering into a mutually beneficial exchange relationship. And so there are many, many situations where those kinds of mutually beneficial exchange relationships, can be incentivized. And so from the perspective of agents who find themselves in a world where there are conflict, related groups where there’s violence or, sub violence but still host in positions, things like, you know, people don’t like each other. People are, trying to sabotage each other, in whatever way they’re still, in principle of all the design for other people looking at that and saying, well, it could be different. Maybe the people who are fighting each other, maybe they don’t have to fight with me and maybe there’s a basis of cooper operation that can transcend the, this, this pattern of violence. And so the idea of imagining counterfactual alliance memberships, imagining a different basis for trying to make it through the world by and not getting hurt and making sure you have enough stuff that is baked into the design of the system. And so, and also not taking groups for granted. So, for example, and Mike mcauliffe and I have a, A B BS commentary on really nice target article by Luke Gachi about the evolution of peace. And and one of the things we talk about is so from, from the perspective of, again, sort of in a hunter gatherer society, non a non wear society, people who are in a group, and some of the people in that group are sort of rating another group. The people who are not doing the rating, who are not getting benefits from the rating are incentivized to not frame the situation as all of us in this group group A are attacking group B, right? It’s actually a subset of us. So that’s an example of where the evolved system is also not taking for granted the group framing. And so it wants to narrow down the scope of the belligerence to the individuals actually doing the belligerent actions. And not the broader category. And that’s part of the, the evolved psychology of peace and reconciliation is to, for those people who are technically in group A to say, look, other group members of group A are doing bad things. It’s imposing a cost in group B. And and they can contact group B, for example, they can talk, contact the belligerents in group A, they can just talk among other people who are in their same position and non belligerent in group A and sort of negotiate about what to do going forward. So those interests of those people are an example of the kinds of designs in the mind that should be there for group group based conflict reduction and reconciliation. So, and again, it just comes down to this issue of conflict is very costly. And it causes retaliation. People don’t like it when you hurt them and when you take their things. And so the the people who differentially benefit from that have an incentive to expand the class of people who believe they’re in a conflict with one another. But there’s always countervailing interest for the people who are not benefiting from that conflict to curtail the scope of who is considered in a conflict with one another. And I think a lot of current and historical conflicts can be understood as this tension of the people who benefit from a conflict want to obscure who’s actually benefiting how widely the conflict in fact goes. And it’s in those people’s interests to sort of obscure and make very broad the the the classes of people who are in conflict. And, and then there’s this countervailing force to sort of narrow it down to who is actually benefiting who’s actually doing the violence as a way to curtail that and to possibly undo some of the dynamics that are leading to that conflict and violence.

0:30:15 Ricardo Lopes: So I want to ask to ask you something that particularly in the context of my show makes a lot of sense to ask because over the years, I’ve had interviews with people that place themselves on either side of this ongoing debate in the anthropology of war and peace between the people who think that war is very ancient in our revolutionary history. That is the deep rooters. I mean, I’ve interviewed people like Richard Wrangham, Zar GT Raymond Hames. And I’m also about to release an interview with Luke Lo Ai. And then on the other side, people who think that war is much is a much more recent phenomenon. I mean, I’ve interviewed people like Brian Ferguson, Douglas Fry and others. So do you think that from your perspective, what you said there about conflict, peace and reconciliation should also inform this debate or not?

0:31:22 David Pietraszewski: Well, it depends on what the people are debating about and why. I think one doesn’t wanna over attribute or psychoanalyze debates too much. But there’s different levels at which a debate can be happening. And I think 11 possible assumption going into a debate like is war ancient or not is the issue of how inevitable is war and conflict. And I think that if you, I think one should be careful about, assuming that that’s a straightforward thing where if war is recent, then it means war is avoidable. But if it’s long standing, it isn’t. I think that’s oversimplified and that’s sort of wrong because actually, to the extent that conflict is a long standing element of human evolutionary environments because conflict is costly, we would expect all kinds of very good design to avoid conflict. We see this in biotic conflicts all the time. Right. So there’s a large literature called the Asymmetric War of attrition where animals, basically avoid the cost of escalating conflict by doing things like, you know, very ritualized, you know, some sort of like deer, sort of walking back and forth in parallel to assess size differences. And if one’s much bigger than another, they, they, they’ll just the, the smaller one will back down, both are doing that because there are selection pressures to avoid conflict because conflict is costly. If you can, if you can basically find the information that you would eventually get at the end of a very costly conflict before the conflict even starts. There’s good selection pressures for to do that on both parties sides. So that’s an example of die a conflict over something like territory is a long standing part of deer psychology. And because of that, there is extra design to avoid the cost of that conflict. So analogously, if we say something like large scale conflict and what counts as large scale, that’s a definition in there are just facts on the matter on the ground of how many people, historically, how many hominids lived historically, what were the scales of conflict? Those are, you know, facts that we can establish either deductively or looking at the arch archaeological record or possibly genetic record. And we’ll just find out what we find out. But in terms of this issue of if large scale conflict, something like war, was pervasive. That really doesn’t mean that war is currently more inevitable because of the design of the mind. For example, if to the degree large scale conflict has been a long standing aspect of human psychology, we would expect more design to avoid the cost of conflict as well. So, oh, yeah, I’ll just, I’ll leave it at that. Yeah.

0:34:19 Ricardo Lopes: Ok. Fair enough. So, but we’ve been talking here today and in our previous conversation about social groups or groups a lot. So, but do we have a good understanding of what the human mind is really representing when it represents a social group?

0:34:41 David Pietraszewski: Yeah. No, the short answer is no, we don’t. And, and really, I mean, if you even, you know, what is the human mind representing when it represents objects? Even that we don’t really know, I mean, we have certain ideas about like early level perceptual things. But once you start thinking about, well, the rest of the mind has to use an object representation. How does that work? It’s still very, very difficult and we still know very, very little. And so we’re in that same boat for something like a social group. I mean, one of the papers that I’ve done a B BS paper and other people are doing things like this is to sort of just start to take this question seriously about what are the representations and this is based on a long history and sort of cognitive science, for example, like people, people like David Marr and Mar before Mar Marvin Minsky, asking really what are the representations? And so my contribution is a small little contribution, but it’s a contribution to also just sort of say we need to start asking these kinds of questions. And so what, what the proposal is is that there’s something called the frame problem in, in artificial intelligence and cognitive science, which is sort of an event happens. And then the question that is faced is given event X, how has the world changed? So what changes when event X happens? And so that’s not a trivial information processing problem for a robot or a human being. And so, addressing that has been a long standing issue and there’s no such thing as a general solution to that because it depends on you know, what domain you’re talking about. So if I knock over your mother’s priceless antique vase, you know, for the physical system, what follows from that is there’s a bunch of glass on the ground, but for something like a social system, you are gonna start feeling differently about me. You might be angry at me, other people need to represent sort of both of those things. So the physical system and the social system you know, both need to be activated for that for that kind of a thing. And so this issue of how do events change the structure of the world? What is it, what is anticipated at time? Two, given an event at time, one is fundamental to group representation on this idea in my paper. And one idea is that what makes a group representation is not just a DTIC interaction where one agent interacts with another. So imposes a cost on, so a imposes a cost on B that is just a diode interaction or a, you know, gives a benefit to B that’s still just a die interaction, it can go back and forth but it’s still just die. And so a long standing issue has been how do you get multi agent psychology to scale up from from die things and there’s lots of species that have die reasoning but don’t have sort of group based reasoning. And so one idea is that really what changes is the involvement of a third agent. So if you and I, if you and I are allies, you know, and somebody comes and hurts me, You may react to that event of somebody hurting me, even though you haven’t been directly affected at all. It’s an indirect social consequence. It’s the term for it, it’s an indirect social consequence. And so the fact that third parties can become involved in a dieted conflict just by virtue of their background relationship with the, the die agents involved that on this model is sort of the crucial difference between die and multi agent representations. And so the paper goes into details about for a conflict, what would a group representation be. And so the sort of idea is that there’s only four ways that a third party can be drawn into a conflict. And we won’t, I won’t go through all the details here. But basically, the idea is that group membership is the assignment of agents to particular roles within those four tragic interaction types. And so, it’s it’s a, it’s an idea that what it means for something to be a group is is an expectation of behaving in a certain way in tria interactions where one agent is coming to the aid or attacking another agent because of some initial cost imposition, the same model also applies to benefit conferral. Although we haven’t published that work yet. And we’ve done some agent based modeling, Dan Conroy Beam. And I on, if you kind of plug these rules about roles and triads into agent, into agents in a, in a Asian based model world, do you get something like groups emerging? And in fact, you do. So, it’s a partial answer to what is the mind representing when it represents a group. It’s not complete But it’s a literal, as opposed to sort of metaphorical or descriptive account because currently, a lot of the things that are talked about, which are accurate and valid, but they’re not literal mechanistic descriptions. So something like, you know, these people belong to the scripts, you sort of put them on a container and it’s sort of left implicit. Well, what, what does it mean to assign agents to that container? And when you start to have to flesh that out mechanistically, as you would, for example, in building a robot that can see and reason about groups and understand groups. You realize that you actually have to plug in these fairly strange low level rules like, you know, if, if you and I are friends and somebody comes and hurts me, you’re gonna come and hurt them if they hurt me. You know, so that kind of their literal contingent behavioral reaction is something that’s probably somewhere in the system. So that’s, that’s the example of the kind of thing that the mind is representing when it represents a group. It’s a behavior based framework.

0:40:51 Ricardo Lopes: Mhm. So you’ve also already mentioned there how having an understanding of how we represent groups would apply to the study of conflict. So, I have another kind of question here then that perhaps from a theoretical perspective is even deeper. So what are cognitive processes? So we’re always talking about cognitive processes, cognitive mechanism. So what are they exactly? And how do you think we should approach their study? Because you have also this very interesting paper where you talk about integrating social ecological and evolutionary approaches to a having a better understanding of what you call the instrumental nature of cognitive processes. So, could you explain that?

0:41:45 David Pietraszewski: Sure. And it’s not something that I just think too. I mean, you know, this is something that is has been worked out on already. I would say, and so, yeah, what do you mean by a cognitive process? And I think the simple answer is, well, what do we already think of when we think about a process? When we think about the body from the neck down? Right? So, and there’s fairly simple things that follow from just using that analogy that quickly become nontrivial in the case of psychology and pretty quickly diverged from what we are currently doing in psychology. So, one of those things is, you know, as a body you experience, well, like, you know, I, I raise my hand because I want to, right? But I want to is not a good explanation of, well, how does it work? Right. And so because, you know, if, if I can’t raise my arm, what do you tell me? Well, try harder, Dave, you know, like that you would not be a good doctor. If that was, if that was your explanation, you know. So, so Dave tries to lift his arm and, you know, one of those words isn’t working, you know, like Dave isn’t working or, you know, the arm isn’t working, that’s fine. That’s an explanation of it. But at some level, but it’s not an adequate mechanistic understanding. It’s certainly not the kind of thing you would be satisfied with if you wanted to go in and fix the fact that I couldn’t lift my arm, for example. So you need an idea that, well, there’s this intuitive experience of sort of unity that there’s an agent who wants to do things who believes things, you know, this is something that people call our intuitive theory of mind or Dan Dennett talks about this is the intentional stance. And he works and I have a paper about modularity where we talk about this is the intentional level of analysis in all cases, it’s just sort of this highest level everyday intuitive way of thinking about people as agents and what they want, what they believe. And it’s fine. It’s correct in the sense of it describes, you know, Dave wants to raise his hand, that’s why he raised his hand. But in the same way, if we want to understand, well, how did I raise my hand again from the neck down? We understand there’s lots of systems and so Dave still exists. But Dave, if you go down a level, we break Dave up into lots of different systems with lots of different functions. And what you wanna do is you want to have an adequate account of how I raise my hand. You don’t worry about. Well, we’re only gonna let there be three systems. Right. We don’t even really worry about where does one system start in one system end? We don’t worry so much about, you know, does it feel hard to raise my hand or not? So, there’s all these sort of like, things kind of in the way of just getting to the simple fact of the matter of, well, there has to be like mechanical stuff inside the body that makes it possible for there to be this thing where Dave wants to raise his hand and is able to raise his hand. And so, well, how do those work? Well, and you look at the history of physiology and the history is very interesting because, not to too, too long, but aggression, but people had access to physical bodies for as long as people have been around and scientifically, you know, for a very long time, if you look at, it’s very interesting to look at the history of physiology. So you look at something like, you know, fifth century physiologists, like Galen. They’re, they’re talking about like mechanisms like moisture darkness. You can kind of see that they have an idea that they need to apply function to the body. But they’re not doing things like nephridial, right? Or like or microtubules or, or things that are really detailed descriptions of functions and sub functions in the body. And the point of saying that is, you can’t just look at the physical system and infer the function you need some idea of what’s the broader context in which this thing is, is is operating. And so when we talk about well, Dave lifts his arm, right? And is able to and wants to there is a physical level description, that’s the, that’s the physical stance. As Annie Wertz and I talk about it, that’s the implementation level of analysis. But there’s also this middle functional level, this is the the functional level where that was what allowed physiology to become a progressing science where people realize, well, the function of the heart is to pump blood. And all of a sudden all the material things they were looking at why they were structured, the way they were made sense and even the way they saw the material things changed, right? They understood this thing was not just incidental, it was a pump and the fact that it was moist was incidental, the fact that it was dark incidental. So the units of function changed. And so just as so for the case of raising my arm, we need a middle functional level description of, you know, various systems that execute certain functions about arm position about planning system. So what it means that I wanted to is there’s some representation and some system about the location of the limb representation of a future state of the limb that corresponds to lifting, right. And then there has to be some system that takes that representation and translates it into a series of electrochemical signals that eventually lead to muscles moving the arm. That’s a, that’s really nontrivial. If you wanted to build a robot that could do that, it would take you a while, but it’s that middle abstract functional level. That’s also really important. And so you have three levels at the end of the day, you’ve got the intuitive intentional level. Dave wants to raise his arm, he’s able to, you can look down at the meet the implementation level. And you can look at this middle functional level and exactly. So that’s all of that is the preamble to the simple answer then which is you do the exact same thing with mental processes. You say, look, you know, people believe things, they wanna do things, they intend to do things. But there’s a level of description that is above the the meat in the mind, you know, the neurons and the electrochemical signals that’s this functional level description and it’s of systems and you don’t worry about how many there are, there could be billions, there could be three. You don’t worry about that. You just try to say, well, what are all the mechanical steps that would be required? And at every point, the point is you try to expand on what is currently known into more and more detail. And so there’s a big long history and psychology of avoiding that middle level because it’s not intuitive. It’s also controversial. And people like to have evidence about things. And it’s the level that has the least evidence because in a lot of cases, you have to speculate about what the abstract functions are before you go looking for them. As Marvin Minsky put it, you can’t look for something till you have the idea of it. And I think a lot of the history of psychology and cognitive science has been hindered because of what G Gert Geens would call defensive theorizing, which is, it’s very easy to avoid making claims at that middle level. It’s easier to just talk about what physically exists. It’s easier to just talk about behavior, it’s easier to talk about the intentional level. And because it’s difficult to operate at that middle level, there’s a lot of uncertainty. It’s also easy to glom onto distinctions that may not matter very much. So, for example, is there one system or two systems, you know, or is a system completely isolated from another system? You know, you would never start off asking how does a body work or how does a car work by, by really glomming onto there have to be two systems or we have to show, you know, 90% degree of isolation for us to count these two things as being separate systems. Like those are sort of backwards ways of doing it. You just don’t worry about that and you say, well, what are the functions? What would they, what would it have to be? And so anyway, that’s I’m going on and on, but basically, the history of psychology is sort of interesting. It’s deeply interesting and it’s all repeating the history of physiology here where that middle abstract level is crucial. It’s very difficult and it’s kind of being it’s emerging towards the end of what has been a very productive century and a half of science, but it’s still not very established despite despite that century and a half. Yeah.

0:50:53 Ricardo Lopes: So please correct me if I’m wrong. But, but a a as you were explaining all of that one thing that came to my mind is if all of that would connect in any way or relate in any way with discussions surrounding a free will beca I mean, I’m asking you that for several reasons, one of them was because when it came to the part where you mentioned, oh, I want to do something, I want to raise my arm or something like that. I mean, that, that sort of level of cognitive A AAA G control that we have. I mean, in a way reminds me of debate surrounding free will. Then the, the second reason is that just earlier this year, two very interesting books came out when by Kevin Mitchell and the other by Robert Sapolsky, arguing from two opposite sides from an evolutionary and biological and neuroscientific perspective about free will. And also, I mean, you told me that you’re working with Christian List, right? The previous guest on the show as well. He has a very interesting book on Free Will on that issue. So, I, I mean, would, would there be a connection here or

0:52:16 David Pietraszewski: not? Yeah, the argument here. Yeah. The, the paper is with Christian List and Thomas Hills and we’re, we’re, we’re quickly trying to get it out. So, I’ll sort of brief allude to it. I mean, the bottom line is that the argument is that free will is real. But it’s mechanical. So, so, so, so so I can unpack that a little bit. So the idea is that when you say something like, you know, I want to raise my arm and I’m free to do so. That is a representation of an evolved organism. And you want to ask the question, what is that evolved organism mean? When it’s saying it’s free to move its arm. And the argument is that what it means that I’m free to do something is essentially that nobody is coercing me. So you didn’t, you know, put a gun to my head, right? So I did, it means that you weren’t forcing me to do it. It also means something like, you know, maybe, you know, my arm gets injured, right? And I can’t control it anymore. So, so that’s another situation where I no longer have free freedom or the use of my arm, right? So from the perspective and evolve organism, the body that it’s composed of and sort of its planning mechanisms, its representational systems, there would be selection to represent its own ability to determine what it’s gonna do in the future. So, and there’s just some things that it has, it has more or less of that ability to do. So if you hold a gun to my head, it’s my planning representational systems matter less because you’ve constrained my options because of the strong cost benefit nature of that. So for example, you know, I can’t, I’m not going nothing I do about sort of ruminating about, well, what should I have done in the future? There’s no way for my information processing systems to intervene in that kind of situation, right? The, the, the, the point of maximum difference in whether or not I do the behavior is to get you and the gun off of my head, right? That’s sort of the point of maximum intervention. So you want, if you want to change the outcome, you change the point, the gun at my head, not, you know, my my planning systems, same thing with a broken arm. So if I can’t use my arm because it’s broken, you can’t coerce me to like use my arm, I can’t will my way to use my broken arm, right? Like it’s just broken, we have to just let it heal. And so the idea is that in our everyday intuitive folk psychology, we have a representation of agency. And that’s specifically because agents can make decisions and we can intervene on agents, whether it’s other agents or ourselves by changing the state of certain planning systems, certain motivational systems, that’s all coherent. But then there are some aspects of the world that those planning representational changes can affect, like holding the gun to my head or the broken arm. And so the idea is that to put it pretty simply the intentional system or the intentional level has a slot for an agency and then a slot for a mechanism and the agency and the mechanism are two different things. And so things that I’m free to do are sort of things that go in the agency slot, you know, like I’m free to, like, decide to wake up early in the morning or not. I’m not free, like, I don’t have free will to sort of, you know, you know, will my arm into being not broken, right. So that goes in the mechanism slot. And so the idea is that where the idea of free will comes from is this idea of distinguishing between those situations where interventions of, of treating an agent like an intentional being, like changing motivations, changing desires, beliefs are effective or not. And when they’re effective, it goes in the agency slot. When they’re not effective, it goes in the mechanism slot. And so that intentional level system views agency and mechanism as c as in as as mutually exclusive, right? What it means that something is an agent or is free means that it’s not mechanical, right? It’s not just billiard ball causation. And the argument with Christian list and Thomas Hills is that, that’s a psychological explanation for where incompatibility comes from. So incompatible is the idea that free will and determinism are not compatible with one another. Of course, it’s complicated because, you know, it’s, this has been a long standing debate for, for as long as there’s been recorded history, the need for free will so it’ll take like a couple more minutes to to unpack it. But basically the other part is we were just talking about this middle functional level, right? The functional level analysis, abstract function. And that unlike what we were just talking about a second ago is probably driven by something like an evolved artifact psychology. So it’s sort of saying, you know, you’re just treating people like robots, right? You’re saying, look, the person has a bunch of software, it’s like abstract if then rules that was sort of turing’s point, right? Like you can just any computational system can be described as like a complicated set of contingent rules that are abstract. And so that artifact system views everything that people do as a bunch of software. So, and in that sense, so Dave decides to pick up his arm and he can, I’m free to use my arm from the perspective of the artifact system, from the perspective of this functional level of analysis. Everything you do is deterministic, right? Or it’s mechanical and we can put aside determinism for a second, but just say it’s mechanical. So from the artifact system mechanism and free will are not incompatible with each other, right? So, in fact, in order to be free, in order for a robot to say, hey, look, I’m free to do this, that’s actually a bunch of mechanical steps. It’s a representation of what we just talked about. Which is, the robot is representing, it has not been coerced, right? That, it can change this outcome if it were to, for example, have a different, waiting of something like, you know, costs in the future, you know, so like I’m free to change when I wake up in the morning. And what that means is I can represent that if I want to get more work done, I’m gonna have to sort of, you know, do something differently. I can, you know, I can make a plan about how to do that, right? And so that’s all mechanical steps when I say that it’s all mechanical steps. So you have one system, the artifact system, the functional level of analysis treating free will as mechanical. So that’s compatible. So there’s this long history of people having a compatible list intuition about free will that free will and determinism or mechanical world are consistent with one another. But then that’s con contrasted with the output of our theory of mind system, the intentional level where free will and mechanism are completely opposed to one another. And so our argument in this paper, it’s sort of an argument is we have the, we, we have the, it’s the meta problem of free will. Why does anybody think there’s a problem of free will to begin with? And this is sort of borrowing from the meta problem of consciousness, right? Why do people think there’s a problem of consciousness to begin with. And the argument is that you have two different onto or two different evolved systems for carving up the world. One is viewing the world in terms of agency and it’s agency that’s inconsistent with mechanism. Then there’s another system that is just viewing the world as everything is a mechanism. And those are two different systems that can be going on inside of our heads at the same time, maybe they’re stronger in 11 system is stronger in some people than in other people, but they give completely conflicting, outputs in terms of the compatibility between free will and, and mechanical slash deterministic systems. And the argument is that, that’s probably the psychology underlying this, this whole free will debate and why it feels like free will is a paradox because I don’t know about you Ricardo. But even me, sometimes I wake up in the morning and I’m like, I feel like more of a compatible and sometimes, you know, late at night, I feel like less of a, less of a compatible list. I mean, you know, even within the same person in, in, there’s nice work, empirical work being done on, you know, examining to what degree people are or aren’t. And it’s important. And, and we’re finding out a lot, but basically, the argument is that this is an example of where we’ve had this long standing philosophical issue. But we can actually describe the psychology of where it’s coming from, you know, where the psychology is coming from is essentially that we have two different evolved systems that are following different rules and they’re output conflicting answers about the same question essentially. And that’s, that’s a perfect recipe for for having something that feels paradoxical or unsolvable. And so the answer actually is that the, the, the way out of the philosophical problem is to understand the psychological mechanisms, understand their logic. Such that one can understand why there feels like a paradox, why it doesn’t feel unanswerable. There’s a few more details and we’ll save some of those for the paper. But that’s, but that’s, that’s the answer to the free will problem.

1:02:30 Ricardo Lopes: No, no. That, that’s really fascinating. And by the way, if you’re open to it, whenever the paper is out next year or so, I would love to have you back on the show also with Doctor Christian List. Because it seems really fascinating and I’ve had my own journey when it comes to free will because up until late 2020 I was a full-blown determinist. But then I read Doctor List’s book and I read him on the show and I was like, OK, so perhaps there are a few things that I have to consider here. So he was one of the first people that actually help, helped me deconvert from the Church of the Truth. And and then now more recently with Kevin Mitchell’s book and a few other people I thought with, I mean, I have at least a more nuanced view of free will. I, I, I’m not sure. I, I’m still not completely sure if, with the degree we have free will, let’s say. But, yeah, I’m more sold on it now.

1:03:38 David Pietraszewski: Yeah, I think it’s, it’s also, I mean, to be clear too, like, I don’t think in this paper, I don’t think either side of the debate is, is is guiltless or, or there’s sort of corrections on both sides because you can say sort of, well, it’s free will and illusion. And there’s a version of which that there is, that’s right and wrong just very briefly. One version is, if you think that the way of the mechanical system works is at some point there’s like a little ghost that’s sort of like doing something non material, then that’s wrong, right? If that’s what you think free will, it means we have free will. But the argument is that that was never a coherent understanding of what free will is to begin with. The idea is that what’s a coherent and representation of what free will? Is it something like what I talked about earlier, which is, it’s a mechanical system whose job is for an evolved work to basically assess what is, what is the what are the kind of events that give me the maximum leverage for future interventions is this a situation where I can change outcomes by changing motivation planning systems inside me inside other people, by convincing them talking to them. Or is this something where that’s not possible? It’s just like a broken bone, right? And so that’s what the free will representation is. If that’s what you understand the free will representation to be, then yes, we have it. But if you understood free will, having free will to mean there’s a ghost that pokes on the mechanical systems at the functional level of analysis, then we don’t have it. So it’s like many of these things where we just have to unpack it just a little bit more. Otherwise, if you just say is free will real or not, you can have an endless debate because what you mean by free will, what the sides mean in that is in that issue is is unclear. There’s also an evolutionary mismatch just very briefly. That, that’s the final part that’s sort of, I think fueling the confusion, which is that we never evolved in a world where we could know all the mechanical steps in our mind. So, in that sense, knowing that science can give us a purely billiard ball, mechanical description of how we do everything that’s not something our mind have ever evolved to deal with. So, in other words, from an evolution perspective the idea that everything can be placed in the mechanism slot in, out of the agency slot, even in theory, right? For the intentional system is an evolutionary novelty. And so I think rightly so people sort of feel worried about that, right? Because if I’m mechanical from an of all being perspective, like you can manipulate me, right? There’s all these examples of like neuroscientists coming in and like doing stuff and that’s true, right? Like we are mechanical. So like everything we do in principle could be manipulated and that’s a reasonable concern for an evolved organism. So the i the concern that free will goes away isn’t crazy from this perspective of this evolved system, from the the perspective of this evolved concern about manipulation. So again, the the point is we can understand conceptually why everybody is taking the position they’re taking, we can understand even the misunderstanding and we can understand the motivations behind it. So the answer is there’s not a single answer, it’s just that there’s a lot of pieces and people were sort of talking at each other from cross purposes. And at the end of the day, we still don’t want people to be manipulated. But, you know, if we learn enough about how the mind works, if stuff’s broken, we can fix it, for example. But these are the issues of like, you know, when we’re done with psychology, what are the implications? And we’re still a long way off from that. So, yeah,

1:07:29 Ricardo Lopes: great. So, let’s talk now about David Pins off and Marty Eelin S paper on their alliance theory of political belief systems. I mean, I’m asking you about this, the reason is that you’ve commented on it. So, and I mean, basically I have two different questions to ask you. So if, if they’re right with their argument, what do you think would be the challenges to current political science and political psychology that would stem from

1:08:05 David Pietraszewski: it? Yeah. Well, and I want to start up by saying, you know, I’m not, I, I have a foot in political psychology, but I’m certainly, it’s not my area of expertise. And so I want to qualify everything I say with that. Their argument just to sort of recapitulate for people who may not be familiar with it is that there’s been a long history in the behavioral sciences and particularly political science, political psychology to look at partisanship from the perspective of ideologies. So, so from a us perspective, for example, you know, Republicans and Democrats, they just hold different ideologies. So certain ideologies like, you know, concern about fairness, concern about you know, the sanctity of life, for example, or the sanctity of you know, not starving to death, like those kinds of things. So like kind of abstract principles. Their argument is that sort of been the currency of theorizing in political science in political psychology. And what they do in this paper is they basically just say, well, look, if you take this ideology principle based perspective and you really look at, well, what are the policies that, people are supporting or not supporting? It? Sort of doesn’t, it doesn’t quite work, right. So, and this is something that a lot of political satire and commentary sort of traffics in which is to say, well, look, you know, if you really think life is very precious, then like, what’s your attitude about war? Right. So, you know, if you’re anti abortion because you think life is precious. Well, what about sending people off to war? More pointedly, people point out Well, what about like, you know, fertility clinics, but that also kill a lot of embryos, like, are you for that as well? Right. Are you for are you for you know, prophylactic? So, birth control contraception that will prevent unwanted death as well. And if you sort of, and in this, the argument, the pins off argument is that this applies broadly. This is not just, you know, anti aortic attitudes that if you really push on the attitude and you say, well, what would, again, we can do a thing of like, if there was a robot that actually was the program was, you know, protect life, what attitudes, opinions, policies would it support? And if you follow that logic versus what do people actually do those things look very different. And so their argument is that, ideologies, even many principles are sort of a readout of a much more, sort of complicated, but in another way, simple, sort of as we talked about, sort of an alliance psychology where people sort of look out in the world, look at what are their interests, what norms are serving their interests. We can unpack that a little bit because that’s one of the issues with their idea. But basically, their idea is that interests of people are what’s driving the content of ideologies. And so what people are usually expressing when they’re talking about ideologies is a content that continues to support the growing rights or protections of the people that people sort of treat as their allies or people that they’re concerned about. And so this is the sort of pins off argument is that ideologies are sort of instrumental. And they call it, you know, propaganda basically from the perspective of an objective outside maybe objective, but we’ll call it objective from, from an objective perspective. These are sort of propagandistic biases that people use to say, look, you know, the other side is so bad, they’re not doing a principled thing, we’re principled. And here’s the principles that we uphold and their argument is that both sides are doing that. And, and they, but they both sides sort of seem to be operating as if they’re principle based, whereas the other side isn’t and their argument is in both cases, they’re not really being principle based. They’re just sort of choosing allies to focus on for some of the reasons we talked about earlier. So that’s all, I’ll, I’ll leave it at that as a summary. Yeah.

1:12:44 Ricardo Lopes: So that’s the summary. And you’ve already pointed there to perhaps one or two aspects of it that you disagree with. But before we get into your disagreement, if they’re right, what would you think would be the implicate, the main implications for political science and psychology?

1:13:05 David Pietraszewski: Oh, yeah. And I just want to say at the outset, I actually don’t, I don’t think I disagree. I actually, I kind of as a human being disagree with it. I don’t like it because I have my own political beliefs and I’d like to think of principle and it, it sort of offends my sensibilities and for that reason, I think they’re probably right. Like, I think to the extent that something makes me feel uncomfortable, I would iii I would prefer to err on the side of assuming it’s probably right. And when you look at history and, or, yeah. So, so, yeah. No, I actually do think they’re right. I think that there’s still things to unpack about it. But I mean, and this, this is to specifically answer your question, what they’re really contributing is this argument that against the existing status quo or dogma within political science and psychology. And again, this is not my area of expertise. To the extent that I’ve looked into it. I, I agree with their assessment that this seems to be the approach, but I, I don’t want to suggest that I know every I don’t know that literature as well as I know many others. But my understanding and what they’re arguing is and, and the answer to your question is, what has currently existed is this idea that what matters is ideology and once an ideology is adopted, so people are sort of empty vessels, they take on an ideology for different reasons. And then that ideology, what follows from that ideology then determines their political attitudes, their their political decisions, et cetera. And so, and so this ideology from this ideology perspective, the enterprise is measuring differences in ideologies and then determining who is sort of most attracted to what ideology, for example, if that’s what you’re interested in. And, and then, and then looking at well, what are the consequences of the ideologies for for things like voting behavior or different kinds of decisions? And then if you want to change people’s beliefs or opinions or voting behavior or who they support. The idea is that you, you change their ideologies, right? And so at every level, the existing theory was ideologies are sort of the central focal point of the wheel of political action. And that’s really the efficient place to measure political attitudes and opinions. It’s the place to change them. And the pins off at all argument is that’s basically backwards where the ideology is sort of this confabulation. And, and you can see that again in all these cases where if this is, in fact the ideology, what follows from it is not what people are actually thinking or doing. And so that’s the main contribution and that’s the thing that would need to change is that it shifts the narrative of what the science of political psychology is from an ideology based approach where I think as they describe it, the the pins off at all, sort of they, I, I looked at a draft of their response to the comments of their paper and I forget where exactly if it’s in that or the target article, but they basically talk about, you know, they characterized the, the, the, the previous enterprises basically looking for the essential elements of ideology, like just treating political science as the science of figuring out ideological essence of people. And their argument is that’s, that is, that’s misleading, that’s, that’s the exact opposite. And really, in fact, it’s understanding these much more prosaic but fundamental things about people’s alliances, people’s interests, people’s decisions about, they look out in the world and they say, well, I think these people are on my side. I think these people aren’t. Now there’s a separate issue of whether it’s accurate, but in any case, that’s their argument is it’s alliances, its interests, it’s cost benefit analysis and the ideology is secondary and is a coordinations device that is painted on top of this alliance structure and this jockeying for things like status access to resources, et cetera. So that’s the change. It, it, it essentially is a transformation of the, of the science to the extent that their characterization holds, which I understand it to be fairly accurate.

1:17:41 Ricardo Lopes: OK, but what do you think perhaps are some of the things bigger or smaller that they might be getting wrong in their argument?

1:17:53 David Pietraszewski: Well, I don’t think it’s wrong. I don’t, I don’t think the argument is wrong but I think it’s incomplete and I think that’s as complimentary as one can get about anybody’s work including one’s own. So I think they’re right, but I think it’s incomplete in the following sense. Number one is, well, what counts as an alliance, right? So they’re saying that an alliance sort of is what underwrites people’s ideas. And so for me, I my alliance might be people who are like poor. Poor people, you know, in my country. And so I don’t want my, my country to sort of, like, you know, bomb other poor people in other countries or, like, you know, take money away from poor people in my country by building bombs. Right. But other people’s allies might be, the, the people doing the bombing because they’re concerned about certain kind of economic systems prevailing over other kinds of economic systems. You know, the, those kinds of alliances, the question is sort of, well, why do people look out in the world and decide certain, certain alliances are sort of part of their interests or not? And, this is one of the issues with the, with the idea of an alliance is that, who people support is sort of their de facto alliance. But then the question is, well, why do they support it? And you say, well, there are benefits, there’s benefits that, see the, see the costs. Well, is that actually true? Right. Like there’s lots of people who, as people point out, you know, support economic policies that, so, you know, poor people, receive propaganda about, you know, you really should keep the rich people who are keeping you down, you know, in power because they’re protecting you from, like this out, this threat from the outside. This is like, you know, demagogues throughout history have done this thing and we know this, you know, and if in any one case, you want to say it doesn’t hold in this particular case, if you look across history, this has happened many, many times. And so, well, what’s the alliance then? Right? Is it a, is it a false representation? And maybe that’s the answer, right? So maybe it’s something like the poor people who are being duped by the demagogue are falsely representing their own interests, right? And so they’re entering into an alliance with the demagogue because they think the demagogue will protect them from like an external threat, that’s not really real. And the demagogue is like exploiting them for, you know, resource reasons. So saying that everything is sort of an alliance, which is not what self at all are saying, but if you just take the idea sort of simply on its own, you know, which is useful to do, you just see that the idea of an alliance sort of needs to be complimented with the idea that alliances are perceptions. And they can be wrong. It’s not an easy question to ask in which way are they wrong and who’s wrong at which time? But that’s an example of how the theory is limited. It’s also difficult to predict how alliances might change. I mean, the alliance theory gives you an idea. So for example, as we talked about before, you know, if I’m a member of group A and there are other people in my group who are doing rating of group B and I’m not benefiting from that, from an alliance perspective to the degree that, that the theory or the systems in the of all brains are sort of representing. There are some people in group A who really don’t benefit from rating. And there are some people in group A who do, you might anticipate that in the future, that alliance structure between the people who benefit and the people who don’t within group A might start to break up, right? And so an alliance based account may give you traction on how alliances change. But that’s still another thing that is sort of not fully unpacked in the pins off paper, but it’s also, you know, that’s a complicated dynamic, right? So one paper can only do so much. So I wouldn’t say that it’s wrong. But something like the fact that alliances shift in trying to have content in one’s model or theory to predict how they shift is also something that’s not in their, their particular paper, but that’s not a limit of their paper. It’s just a description of sort of the scope of it. But I would say that’s another example of the kind of thing that is still missing, not that they’re responsible single-handedly from doing it. But, but so those are two examples, you know, just the dynamic nature of alliances and then also this issue of sort of alliances can be wrong in a sense or at least, you know, misperceptions. But I think those are, those are benefits of the theory in the sense of we, one quickly realizes that these are things that we have to like come to terms with. And I think both things have come and come to terms with in in past work. But I think the alliance model is a nice way to think about it and it connects it more up to these basic questions about this alliance psychology that removes it from these very specific idiosyncratic things. And it’s also very nice in the sense of it removes it from sort of this exotic ideology thing, you know, where you sort of read things. Well, you know, like these people were, you know, are hostile or belligerent because, you know, they have like, and there’s some very esoteric, you know, philosophical ideology, you know, as opposed to, you know, well, they’re like belligerent because, you know, like the generation before people bombed them, you know, and like their Children don’t have enough food. Sometimes as as, as sort of the, the intelligent the, the, the sort of intelligentsia, right? The, the, the class of people who work by thinking making their living, it’s very easy among, among that collective of people to sort of think that like, you know, esoteric ideas are what drive behavior and are what’s really important as opposed to very simple things. Like we started this conversation. Why does, why do groups happen? Why does conflict happen? It’s simple stuff. Like people don’t like getting hurt, they don’t want their family to get hurt, they need stuff to live. And so that’s another benefit of the pints off argument. Is that I think it, it brings the narrative back down to, to ground reality about those kinds of things. Again, that’s not saying that people aren’t being duped, but it actually again, clarifies if they’re being duped. Um

1:24:29 Ricardo Lopes: OK, so I have then one final question, perhaps more of a topic I would like to explore today. So you have this very interesting preprint titled What The Difference The Mind Makes, where you point to the fact that psychologists are also using their Evolved psychology when studying the mind. And basically you talk about the difference, detecting mechanisms and then you argue for possibly the influence that they have on the science of psychology. So co could you explain that the reprint? I mean, what are you arguing there? Exactly. Yeah.

1:25:13 David Pietraszewski: It’s basically the evolutionary psychology of psychology, right? And, and we even talked about that a little bit with the free will thing, right? Where it’s sort of saying look psychologists, philosophers, these are still people doing stuff and if you want to study people doing stuff, you can use an evolutionary psychological approach. So it’s basically the evolutionary psychology of psychology. And so this is this is basically a vehicle to sort of introduce that enterprise. The past modularity paper was about that the free will paper will be about that. And then this paper that you’re talking about is also about that and they’re all under that umbrella. This specific paper is about difference detecting mechanisms as you’ve mentioned. And I’ll just introduce sort of what that is. And so this is an idea coming from David Bust in particular. And then Aaron Lukaszewski is also somebody who has sort of picked up and, and advanced this idea. And the idea is that the human mind really should be designed from an evolutionary perspective to detect important differences in situations. Like is there a predator coming at me or not? Right. Or, you know, am I safe or not? And the idea is that for an evolved organism, what matters are cues that allow the organism to detect important things, not necessarily having a really rich true representation of the causal structure of the world. So for example, a gazelle you know, needs to know, am I going to be predicated upon by a lion? So it might have detectors for like, you know, a lion, it might have detectors in its brain for sort of like does the lion see me is the lion close to me and those are all important things for the gazelle to sort of have about lion nature, right? It needs to be able to detect, am I in a situation where the lion is gonna get me or not? Anything else about how the lion works would not be selected for it, right? Like the gazelle is not gonna know like the cognitive neuroscience of lions and how they work, right? Like that’s not, that’s not worth it for the cost benefit calculation of the gazelle. So the idea is that the gazelle’s brain is just like ours. We are not evolved to understand how our minds work or other minds work. We’re equipped with the analog of the lion detector in all kinds of domains. And so what that means is that our psychology as a default is using systems when we’re doing science that are difference detecting mechanisms, they’re designed for detecting, am I in a predation situation or not not systems that are just sort of set up already to sort of look at objectively the causal structure of the world. And an example of another difference detecting example outside of like gazelle thing would be something like, you know, let’s say you go to a city and you wanna pick out a good restaurant, right? What would you look at? Well, it would be the stuff that varies. So how would you determine what’s a good restaurant or what’s a bad restaurant? Well, it might be something like, what’s the level of upkeep of the decor? How hard is it to get into a, make a reservation? Is there a line in front of the restaurant or not? Those are the kinds of cues that, allow you to distinguish between the quality of the different restaurants and those are the kinds of things that a difference detecting mechanism would attend to. But what wouldn’t it attend to? It wouldn’t attend to things like cooking food, right? Because that’s how constant across all the different restaurants, you know, it wouldn’t attend to things like, you know, serving the food. So all of the fundamental causal elements of actually having a good restaurant, right, which is actually cooking food, preparing it for people getting tables and chairs and lighting and buildings, those are all held constant. So those disappear from the perspective of the difference, detecting mechanisms. And so the idea is that when we use our psychology to reason about the mind, we can be misled into only looking at the little bit of variation as opposed to the large scale causal structure as the example of the restaurant just showed. And so, what Aaron Kraszewski has done is argued and David Buss as well has argued that in personality psychology, what personality psychologists have been doing is focusing on differences at the expense of sort of underlying deep causal structure. So as the analogy with the restaurant would be that they’re f been focusing on sort of the long lines and the upkeep of the decor and not at the making food. You know, making sure there’s a table and a chair and you have waiters serving food and things like that. What I argue in this paper is that our theory of mind system is one of these difference detecting mechanisms. And so it’s sort of instead of just taking for granted this thing I said before, which was that like, you know, Dave wants to raise his arm. You know, that’s the structure of our intuitive theory of mind. So there’s sort of an agent slot, a stance slot and then a content slot. So like Sally wishes it would rain. You know, Sally believes there’s a lion. Sally remembers a lion. Sally is, you know, picturing a lion like this is the structure of our intuitive everyday way of talking about mental states. And this is called the propositional calculus. I mean, you can be jargony about it. But it’s sort of a, a real fundamental bedrock of philosophical and psychological theories of how the mind works. There’s sort of an agent slot, a processor stance slot and a content slot. And what I do in the paper is just basically argue that’s probably not a description and we certainly know it’s not a functional level description now that we’ve talked about that it’s an intentional level description. And why is that the content of the intentional level description? And the argument is it’s an important difference detecting structure. So you can kind of think about the agent slot, the the stance or process slot and the content slot is drop down menus and computing. So, and to just give you a very, very simple intuition is sort of I see a bear versus I’m remembering a bear is very different situation, right? So, so, so for the purposes of you and I are like cold, we’re huddled together like behind a tree. We, you know, and we’ve been, we’re out in the middle of the woods and I tell you, you know, I see a bear versus I remember a bear, that’s a very different situation. If I remember a bear, you don’t have to get scared if I see a bear, you do. And so the idea is that the process or stance slot is essentially doing the work of distinguishing situations, a number of different evolutionarily important situations, like does an agent perceive something like that’s an important, there’s a set of inferences that follow from that? Are they currently perceiving it or not? Are they remembering it? Right? Are they thinking about it? The idea is that the reason why we have these categories of sort of seeing, remembering, thinking, feeling that they don’t necessarily carve up the causal structure of how the mind works in a very deep sense. But they they distinguish between really important kinds of situations for an evolved organism, like seeing a bear versus remembering a bear, right? So they’re sort of, they’re the kind of representations, the rough take representations that an evolved system should have about con specifics in one’s own mental states. And then the content slot is the difference between, you know, I want a bear versus I want a canoe, you need to know like what that content is about. And the reason why there’s an agent slot is it’s the difference between, you know, data Cesar bear versus Ricardo Caesar bear versus, you know, somebody else sees a bear. Christian sees a bear, those are important things. And so the point of the paper, it’s pretty basic, it’s just saying look this propositional calculus, agent process content which forms the basis of propositional calculus, which has sort of been just taken as a given since, you know, philosophy of mind essentially started. You can do the evolutionary psychology of that and sort of understand why it has that structure that it has. And so understanding it and analyzing it does two things. One is it allows us to think more about its features from a scientific perspective. But it also allows us to understand in a more deep sense, the limits of it. Because again, it’s just like the gazelle, if a gazelle wants to do science, it has to realize that the lying detectors that it naturally has like our intentional psychology for reasoning about others in their minds is really gonna mislead us about a lot of the causal structure. So if we think that Dave wants to raise his hand is the causal structure is the mechanisms in the mind, then we’re probably in trouble, you know. And so I, I talked a little bit later on in the paper about how this structure agent process content has colored theorizing in psychology since the outset. And you can look at very early work like Franz Brannan in the 18 hundreds. As far back as, as one can look on writing the idea that sort of content and process are separate and diss sociable is is rampant throughout much of psychological psychology history. So the idea that sort of, you know, I’m thinking about Mr Smith, And I’m thinking about, you know, Saint Peter’s basilica. This is an example by Bertrand Russell in the 19 twenties. And even then by then, it was old in his really good book analysis of mind. And he talks about how, you know, the process of thinking is the same, what’s different is the content. And you take that very old idea because of the propositional calculus. And even up through today, the idea that sort of certain kinds of algorithms are content independent, certainly they can be applied to different contents. But the idea that you only need a certain kind of algorithm like bay and updating or you know, plug it in and plug in the algorithm. The idea that that’s sufficient to solve a problem, any kind of information processing problem is not true. And anybody who looks in detail at solving any information processing problem, like reinforcement learning, early reinforcement learning. One of the first instances of reinforcement learning was a Chucker playing program. Richards. Well, yeah. Well, I’ll skip the 000 I won’t go in that aggression of the history of the checkers playing program. But basically, the point is you know, early on people realized that reinforcement learning, for example, would not be sufficient on its own as a principle to have a checkers playing program. You need checkers, specific kinds of representations attached to that more general algorithm, how you get those could be plugged in or can be learned, but they still have to be there. And and that lesson has kind of always been learned, but it’s always lost. People are always erring on the side like the behaviorists of thinking you only need about like seven rules and like everything else is trivial. And the argument, for example, in the paper and this may be right or it may be wrong, but it’s an argument is that one of the reasons why we keep forgetting this lesson is the propositional calculus that our minds are structured to think. There’s the unitary thinker that uses principles of information processing, like reasoning or thinking or plug in something else like bay and updating that then applies to all content. And so this is a deep prejudice in the mind and to the extent our psychological theories recapitulate this deep prejudice. We’re just sort of rehashing what we know to be false until we look into it again and find out that it’s not sufficient. So, so yeah, that’s, that’s what the papers about.

1:37:24 Ricardo Lopes: So my final, my final question today will be then if the main argument of your paper is correct, how could we deal with it when it comes to how we do science and more specifically psychological

1:37:43 David Pietraszewski: science? Yeah, it’s a great question. I mean, and again, this is not even my idea, this is whenever we’ve looked, it’s to, as I talked about earlier, it’s to move down a level and it is just sort of be skeptical about the intentional level, say, look, it’s fine as a description of what we need to explain. But to then say, look, we have to move down this intuitive, unintuitive level from the intuitive level to an unintuitive level, this functional level where the systems that we talk about are defined by their functions. And we have to speculate about what those functions might be. So we’re, we’ve been talking today about systems in the mind that attend to alliances, that’s not an intuitive part of human psychology. That’s not Dave believes this or Ricardo thinks that. And then, well, then how does that work? Right. And how does the alliance system detector work? Well, it has to attend to certain cues and how did that, what are those cues? And, right, you just keep unpacking that. And so the, the bottom line is we need a psychology, The the way around the the problem is to embrace this functional level of analysis for one. And for two, what that means if you take it seriously is that we have to be able to speculate about abstract functions, we have to be able to try to think about what are the information processing the mind solves and what are mechanical solutions to that and be prepared that there will be many, many answers to that at many different levels of specificity. And I think one of the, one of the things I’m realizing that I think is just it’s very simple and fundamental to the success of psychology is if we go in thinking that the science of the mind can be about as complicated as you know, seven systems, or we’re gonna go in looking for two systems or anything. If we’re gonna go in with a prior expectation of this has to be pretty simple. It has to be a certain kind of thing as opposed to there’s lots of stuff, doing lots of things, lots of places, which is already how we think about the body. You know, it’s simple in that respect. It’s just we should go in the way we think about the body. There’s gonna be lots of stuff doing lots of things and simple things. Like, is it, is it one system or two, is it intuitive or hard or easy or those kinds of distinctions? They’re fine but they’re not how you get there. And so how you get there is you stay at that functional level and you also then have to be able to speculate about functions before you before you start collecting data. And this is the other problem in psychology is there’s a long standing tension between psychology being a science of how the mind works and science being a collection of empirical facts. And if you’re gonna do an engineering enterprise, which is what essentially psychology is. You have to have theories about what you should be looking for. And the theory crisis in psychology is doing a good job of exposing failures in theory. But I think there’s a much deeper issue that hasn’t been addressed, which is everybody’s happy to say there’s a theory crisis. But if you take seriously what’s required to resolve it, we already know what that is because we, we don’t have a theory crisis in early steps of vision, we don’t have a theory crisis in early steps of motor control. And what, why that is is because those are areas where we really have a clear idea about what the information processing problems are and people are taking seriously that there have to be a lot of mechanisms for dealing with each of those problems. But if you look at many area, other areas of psychology, if you try to write a paper about what are the problems inherent in groups? For example, like my, how does the mind see a group paper which is published at B BS? I’ve tried to, I’ve tried to publish that other places. And people said, well, you don’t have data or you don’t have a model, right? And the point is that what I was talking about is the stuff you need to do before you have data before you have a model because you don’t know what data to look for, what, what to model. And this was not a one off thing. I’ve this, I’ve come up against this issue for many other projects at this stage. I’ve talked to many other people and so there is a culture and it’s nobody’s fault. It’s just a structural issue. There’s a culture of not adequately allowing for the enterprise that derives what we should be looking for first before we start conducting our modeling or empirical studies. You know, I’m an experimental psychologist too, but we have to be willing to sort of be both theorists, speculative theorists and then also check our ideas in psychology is just not structured right now to do both things. So sort of half of psychology is missing the, the front half. And that’s, that’s the the real problem. And, and so functional level of analysis in allowing for speculative ideas about problems and their solutions. Those are the things we have to do to solve this problem.

1:43:08 Ricardo Lopes: Great. So as I said, at the beginning, I will be leaving a link also to our first conversation in the description box of this one which I think all everyone should watch or listen to because it provides a lot of background and context to our conversation today. So, but apart from that, would you like to tell people again where they can find you and your work on the internet?

1:43:33 David Pietraszewski: Sure, absolutely. So, so currently we are having this conversation where I am in Berlin, Germany. But by the time I think it’s posted, I will be in sunny Santa Barbara. So if you look at me at the University of California, Santa Barbara, I’m in the Psychology of Psychological and Brain Sciences. I’m also a member of the Center for Evolutionary Psychology. And so if you look me up that way, or just Google scholar that, that’ll be a way in my last name is sort of imposing Pietra Zi. It’s, it’s a difficult name to spell. But if you just start Pieetr A but, but hopefully by the time you get to A or S something will fill in and you can go from there. But but that’s how to find me. Yeah.

1:44:16 Ricardo Lopes: Ok, great. So look, thank you so much again for coming on the show. It’s always a big pleasure to talk with you. And as I said, when we were talking about free will. If you’re open to that, please come back on the show next year or so with doctor Christian Li to talk more about

1:44:35 David Pietraszewski: that. That’d be great. I don’t think we will have a choice about it. So,

1:44:41 Ricardo Lopes: hi guys. Thank you for watching this interview. Until the end. If you liked it, please share it. Leave a like and hit the subscription button. The show is brought to you by N Lights learning and development. Then differently check the website at N lights.com and also please consider supporting the show on Patreon or paypal. I would also like to give a huge thank you to my main patrons and paypal supporters, Perego Larson, Jerry Muller and Frederick Suno, Bernard Seche O of Alex Adam Castle Matthew Whitting B no wolf, Tim Ho Erica LJ Connors Philip Forrest Connelly. Then the Met Robert Wine in NAI Z Mark nevs called Hafid Governor Mikel Stormer Samuel Andre Francis for Agns Ferger, Ken Herz J and Lain Jung Y and the K Hes Mark Smith J Tom Hummel s Friends, David Wilson, the desario, Roman Roach Diego, Jan Punter, Romani Charlotte, Bli Nico Barba, Adam Hunt, Pavlo Stassi, Nale Me, Gary G Alman Sam of Zaypj Barboza, Julian Price Edward Hall, Eden Broner Douglas Fry Franca, Lati Gilon Cortez or Solis Scott Zachary. Ftw Daniel Friedman, William Buckner, Paul Giorgino, Luke Loki, Georgio Theophanous Chris Williams and Peter Wo David Williams Di A Costa, Anton Erickson Charles Murray, Alex Shaw, Marie Martinez, Coralie Chevalier, Bangalore Fist, Larry Dey junior, Old Eon Starry Michael Bailey. Then Spur by Robert Grassy Zorn, Jeff mcmahon, Jake Zul Barnabas Radis Mark Kemple Thomas Dvor Luke Neeson, Chris to Kimberley Johnson, Benjamin Gilbert Jessica. No, Linda Brendan Nicholas Carlson Ismael Bensley Man George Katis, Valentine Steinman Perros, Kate Von Goler, Alexander Albert Liam Dan Biar Masoud Ali Mohammadi Perpendicular J Ner Urla. Good enough Gregory Hastings David Pins of Sean Nelson, Mike Levin and Jos Net. A special thanks to my producers is our web, Jim Frank Luca Toni, Tom Vig and Bernard N Cortes Dixon Bendik Muller Thomas Trumble Catherine and Patrick Tobin John Carl Negro, Nick Ortiz and Nick Golden. And to my executive producers, Matthew Lavender, Si Adrian Bogdan Knits and Rosie. Thank you for all.