Factually! with Adam Conover - Where Beliefs Come From with Celeste Kidd

Episode Date: February 12, 2020

Adam sits down with Celeste Kidd, Professor of Psychology and head of the Kidd Lab at UC Berkeley to discuss how humans form beliefs and knowledge, and how easily that process can be corrupted by outside forces. Learn more about your ad choices. Visit megaphone.fm/adchoices See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

You know, I got to confess, I have always been a sucker for Japanese treats. I love going down a little Tokyo, heading to a convenience store, and grabbing all those brightly colored, fun-packaged boxes off of the shelf. But you know what? I don't get the chance to go down there as often as I would like to. And that is why I am so thrilled that Bokksu, a Japanese snack subscription box, chose to sponsor this episode. What's gotten me so excited about Bokksu is that these aren't just your run-of-the-mill grocery store finds. Each box comes packed with 20 unique snacks that you can only find in Japan itself.

Plus, they throw in a handy guide filled with info about each snack and about Japanese culture. And let me tell you something, you are going to need that guide because this box comes with a lot of snacks. I just got this one today, direct from Bokksu, and look at all of these things. We got some sort of seaweed snack here. We've got a buttercream cookie. We've got a dolce. I don't, I'm going to have to read the guide to figure out what this one is. It looks like some sort of sponge cake. Oh my gosh. This one is, I think it's some kind of maybe fried banana chip. Let's try it out and see. Is that what it is? Nope, it's not banana. Maybe it's a cassava potato chip. I should have read the guide. Ah, here they are. Iburigako smoky chips. Potato

chips made with rice flour, providing a lighter texture and satisfying crunch. Oh my gosh, this is so much fun. You got to get one of these for themselves and get this for the month of March. Bokksu has a limited edition cherry blossom box and 12 month subscribers get a free kimono style robe and get this while you're wearing your new duds, learning fascinating things about your tasty snacks. You can also rest assured that you have helped to support small family run businesses in Japan because Bokksu works with 200 plus small makers to get their snacks delivered straight to your door.

So if all of that sounds good, if you want a big box of delicious snacks like this for yourself, use the code factually for $15 off your first order at Bokksu.com. That's code factually for $15 off your first order on Bokksu.com. I don't know the way I don't know what to think I don't know what to say Yeah, but that's alright Yeah, that's okay I don't know Hello everybody, welcome to Factually, I'm Adam Conover

And look, we think of ourselves as rational beings, right? We believe that when we make decisions, we weigh the evidence and we sort through it. We throw out the bullshit and we base everything we believe on pure, rational reasoning. But that's not actually how we form beliefs or opinions. Human knowledge formation doesn't take place in a Senate of toga-wearing neurons arriving at final correct conclusions. First of all, neurons are, they can't wear clothes. They don't make clothes that small. But more importantly, humans aren't built to respond to the world like perfect scientists or even mediocre scientists. Our knowledge and decisions are based on the opaque, somewhat random processes of our messy,

messy brains. Let me just give you one example. Even just the order in which we're presented information influences what we believe. The information we receive earlier gets more weight than the information we receive later. Our guest today did a study on this, which asked people to make a relatively neutral internet search for activated charcoal. Now, you might have heard of activated charcoal. It's pretty similar to the stuff you use on the barbecue. It's actually pretty useful if you overdose on pills and end up at the ER. It's good for that purpose, and it can save your life in that case. You'd think that would be enough for people, but no, there's also a pseudoscientific

fad going around where people are trying to use it to cut down on their body odor, clear their acne, and in a triumph of counterintuition, people are using this coal product to brighten their teeth. It does none of these things. This is just a pseudoscientific wellness fad. But in the study, when people started searching for activated charcoal, they were undecided about its benefits at first. But after clicking on just a few search results, people went from undecided to 50-50, and then finally to unabashed believers in the health benefits of coal. The search results presented a positive spin on activated charcoal that was out of line with actual science. It was irrational, yet people followed along because that was the information they saw first.

If the search results would have presented a more accurate spin on activated charcoal, they would have formed different opinions. The point is, we're not really as discerning and rational as we want. We're more like piggies at the informational trough. You know, we form our beliefs and opinions out of whatever crap is fed to us. What we see first, we believe first. Here's another way to imagine this dynamic. Picture Twitter, right? Imagine there's a spat between two presidential candidates. Difficult, I know. The candidates are arguing over the contents of a private conversation they had a couple years ago. Now, it's impossible for us to have direct knowledge of what's happening.

So on Twitter, your understanding of what happened, your emotional feeling about it, your belief about what went down is likely to be based on which candidate's framing of the dispute you saw first. And it could be right, could be wrong, but your certainty about your rightness is no guarantee that you are actually right. You are profoundly influenced by which one came to you initially. And this is just one of 10 bajillion potential instances in which your good faith efforts to seek the truth might lead you into error because your brain works far more messily and far more rationally than you'd like to believe. The truth is that human knowledge

isn't perfect or permanent. It's an ongoing attempt to approximate what is true in the world, and it is shaped by an endless stream of random ass factors. And to help us understand how human beings learn, how we form knowledge and beliefs, our guest today is Celeste Kidd. She's a professor of psychology who heads up the Kidd Lab at UC Berkeley. Please welcome Celeste Kidd. Celeste, thank you so much for being here.

So I want to tell people how we met. You came to see, I was doing my show Mind Parasites Live, which is a show, a comedy show I was touring in 2019. It was an hour all about these weird biological parasites that control their host's minds and the cultural parasites that control our minds and how we're not really, we don't have free will in the way we believe we do. And you came to see that show. I did. I did. And I came to see that show because of your Twitter account. Oh, just because of my, oh, my excellent Twitter account. Your excellent Twitter account.

That I pretty much only use to tweet about shows. No, no. You also tweeted some very nuanced understanding of some, I would say, anti-feminist comments that a certain science communicator had recently relayed. Oh, okay. I appreciate it. I didn't realize that. Um, uh, yeah, that's, uh, well, every so often I like to get out on a limb and say, I can, I can say a little something that's in the cultural consciousness, um, to, uh, to push back and, and shape things the way that I think they should be shaped. But I do that very judiciously. But thank you for saying so. But you came to the show, and you found that the material in the show intersected with your work to a certain extent, right? That's correct.

And as a result of that, we're now sort of informally working together a little bit. Like you asked me to help you out on some of your, on some of your research in a way that I'm still understanding exactly how you want me, a comedian to contribute, but we've had a couple of phone calls and you've bounced ideas off of me. Yeah. Yeah. You've been, you've been invaluable. I was like, that's how science works. So like you're, you're doing it right now. Really? Yes. Yeah. You've been doing it for, for, for years and just didn't realize. I've been doing science the whole time. You've been doing the science the whole time. You had novel hypotheses that were things that were in the domain that we were currently working on. And we were very impressed. Like that was science. That's step one. That's the hard part of science actually, is generating novel ideas for things that you might be able to test. So I would say you are full on in the lab. You are a full on research scientist.

That's what we do. Oh, my gosh. You're blowing me away right now. Box checked. I hope my parents are listening to this because folks might know about me. My entire family is scientists except for me. I'm the only dropout with a bachelor's instead of a PhD. So maybe they'll finally get a little bit of credibility come Thanksgiving time.

For sure. so maybe they'll uh uh i'll finally get a little bit credibility come thanksgiving time for sure i think i think the further along you go in academia the like less respect people have for the actual degrees so uh that's like inconsequential nobody cares um what what degree you have it's it's the ideas that you have and your ability to execute them and your ability to uh critically evaluate evidence That's what we do, and that's what you've been doing for many years now. I do my best. I don't have – I'm a dilettante about it. I don't have time to fully – I get the evidence as best I can, and I try to bring it to the people through comedy. But you're the person generating the evidence and generating the study.

So you have a lab at Berkeley, correct? I do. And you study knowledge formation among other things in humans. Correct. That is what we do. So tell me what you've found. Well, what's your, what's your big picture insight that you feel that you have gotten? Because we think of, as I was saying in the intro, we think of ourselves as being rational beings who we, we weigh the evidence and consider, and then we use that to come to a conclusion, and that is the knowledge or beliefs that we have. You study that. Is that true?

That is not as true as I think it feels like it is. If I've learned anything about how people If I've learned anything about how people operate, I've learned that there's not the knowledge in the way that I used to envision it. I used to think about what infants were doing as figuring out, given their observations and experiences in the world, what's true. They come to knowledge, and once you have that knowledge, you get to keep it for life. what's true. They come to knowledge. And once you have that knowledge, you like get to keep it for life. And I now appreciate that much more than is the way that people think of it. Everything is beliefs. All of the knowledge that you think that you possess is your best guess, given what you've experienced in the world. And you're never done. Even for concrete things, I know what the word cup means. I use the word cup. And you might think that I have acquired the concept of cup.

It's done. That knowledge is set. I'm going to use that for life. But that's not really how it works. Every time I encounter a cup, I am updating my conceptual representation of what counts as a cup. And if I move to a place that has wider cups, every cup that I encounter, I will gradually morph my concept in order to accommodate this change in the environment. So there is not really any such thing as knowledge proper.

Instead, everything is a belief, and you're doing your best to approximate what's true in the world. Instead, everything is a belief, and you're doing your best to approximate what's true in the world. And it's an ongoing process that starts in infancy but continues, hopefully, if things are going well, until you die. There's no such thing as knowledge. Meaning that you're trying to approximate what is true in the world, but it's always your best guess. Yeah. It's really, I think of it as probabilistic expectations. Wow. Even for things like language that we share.

It turns out that when you probe it, when you dig in a little bit deeper, it is not as shared as you might expect. So when I use a word, you use a word, you might expect that we have the same concept in mind. But when you probe it, it looks like that's not really true. We both have slightly different concepts because we're inferring our concept based on our unique set of experiences, the particular sets of data that we've seen, and what I've experienced and what you've experienced are two different sets of things. Are there cases where when people are coming to these beliefs that are, that would seem non-rational to us, like the reason that I end up with a particular belief is not because, well, for instance, you talked about infants when they're like discovering the world. And I almost imagined that, yeah, our conception of them is that they're little enlightenment

philosophers. They're like, oh, solidity. I see. Like the, like this object can't move through this. So things must be extended in space. All right. They're rationally figuring it out. They're not doing that. What are they doing instead? And what are some of the weirder ways that also want to say it's possible that they're doing something like that, but really we don't know. We could get into why we don't know, but the short answer is everything that we know about infants, we are inferring based on our best guesses as to why they are showing the particular visual interest that they are. If they look longer at one thing versus another, we try to guess why. So you can't, you can't pull them on what they believe. You can't, you can't say, what do you think that is?

You can't pull them on what, what they, you can't, they can't talk. Yeah. You could pull them, but you wouldn't get, you'd get, you'd get nonsense. You'd get nonsense back. You could give them like classic psych task. You could give them like a, you know, a two alternative force choice You could give them a classic psych task. You could give them a two-alternative forced choice task, give them a button box, but they'll mash it. You're not going to get anything useful that way.

So what we do instead is we show them different types of events. The example that you gave about why we think infants know something about solidity, infants know something about solidity. The classic tasks are things like showing infants displays where a drawbridge moves and then stops where an object, they've just seen an object placed, versus seeing a drawbridge appear to go through the object. And when you show infants those types of displays, they often look longer at what we would call the impossible event. So based on them just looking longer, we say, oh, they're surprised.

They must be surprised because they understand that objects are solid. But really, we don't understand a lot of why infants are doing the things that they do. Infants could be looking longer because that's what they expected. For the drawbridge case, they could look longer because the drawbridge is moving more in one case versus the other. It's a bit of an inference. It's a really big leap between an infant is looking longer at A versus B. Thus, they understand that objects persist in time and space and anything deeper than

that. Well, so where do they acquire, where do we get our beliefs from? We get our beliefs from a process that I think is reasonable to call a sort of more evidence in the world that we could use to form them than we possibly have time to navigate. I didn't appreciate that at the beginning. I think especially those of us, you and me, that consider ourselves skeptics and people that like to think critically about things in the world, we have this sort of illusion that we're going to learn everything. I was like, we're going to learn everything that's possible. And all of the knowledge that we have, we came to through reason. Once you point it out, it becomes obvious,

but it wasn't to me like when I was in high school. There's no way. There's too much richness. You can't possibly, it's not an issue of like motivation. You cannot acquire all of the information in the world. Every moment that you are sitting in a place, you are making all of these micro decisions that come with a huge opportunity cost. So if I'm looking at you right now, I'm not looking out the window. And so I am learning things about your facial expressions. And I'm able to tell what you think maybe about what I'm saying. But I'm missing out on everything else that I'm not fixating. So if a unicorn passes behind me, I'll miss it.

I won't know about that. There's way more that you don't know that you can't know that you can't integrate into your understanding of the world, then there is things that you can integrate. So you have to sample. Reality has so much detail. Right. Reality has so much detail and the world isn't static. It's perpetually changing. So you can't know all of the things perfectly. We have a pretty good system in place to do a decent job of approximating. There's our systems that we have in place good system in place to do a decent job of approximating. There's our systems that we have in place that govern our systems like attention, help us pick and choose, help us figure out when we should continue watching versus when we should like, we know enough, cut and run and go look at something else,

look for another window or whatever the situation is. You're making me, you're reminding me of, on my last interview with Scott Soames, he's a philosopher, so I've been thinking about my own, like, times of studying philosophy. And part of my fantasy at that time was that I would, like, prove everything that I believed, that you could start from first principles, right, of you could do, like, you know, Descartes, I think, therefore I am, and, like, expand from there expand from there and then like prove your entire edifice of knowledge. I love that idea. Yeah.

In fact, and some philosophers claim to have done it, you know, that's what Kant's system was. Ayn Rand famously like claimed that she derived an entire system of thought from A equals A, which is. Ayn Rand claimed a lot of things. Ayn Rand was a nut bar, but the, but you know, that's like sort of part of the project. That's like the philosopher's fantasy. And it's not possible to do it.

It's not possible to have literally every single thing that you believe or feel that you know be based on rigorous logical analysis. logical analysis. In fact, the, one of the best arguments against, you know, the Descartes evil demon hypothesis, which is the, you know, the, the fear that what if everything I experience is just an illusion given to me by an evil magician or an evil demon or the matrix, right? That how do you disprove that? Right. Well, one of the ways you disprove it is like, you just don't behave that way. Like you, like it's, you're, you're living your life, you know, like there's, there's too much, there's too many other things to do. Like it would, you can't spend the years it would take you to finally like disprove that perfectly and then prove everything else you need to, you still need to eat breakfast. And you still, you're still going to necessarily

have all of these other beliefs that are not based on your conclusions about that. So to a certain extent, it's almost immaterial because you actually do not work in this hyper-rational way in your life. That's right. But I would add as a caveat onto that, the systems that you have in place do a pretty good job given the limitations of how much information you can process moment to moment. Like the fact that you only have one set of eyeballs, the fact that there's not infinite time and you don't have infinite bandwidth. Our systems are pretty good at giving us a pretty good idea most of the time of generally how relevant information is in the world, but it's predictably imperfect. I think it goes wrong in certain circumstances, and understanding that is important for figuring out how to get people smarter, I think.

how to get people smarter, I think. One thing I've thought is that we're heuristic like animals in that we have these models that are, I think I'm using the word heuristic correctly, that it's like a scheme to figure out a pretty good answer most of the time, like when presented with a bunch of data, right? Yeah. And we're pretty good at that. Like one thing we talked about in our writer's room

was pattern seeking, that like humans are pattern seeking creatures because that helps us like define cause and effect. Like we're always looking for cause and effect. Okay, some random shit happened. Why did it happen? Okay, every time the random shit happens, that happens too. So this probably caused that. And that's like pretty good when you're like trying to figure out, you know, why your friends keep dying in the jungle. You know what I mean? Or figuring out which light switch to use. It was

like, I touched that and then the light turned on. Even if that doesn't look like a light switch, I contemplate whether or not I caused that because it's such a huge coincidence that those two things would co-occur. Yeah. But you're not actually acting like a scientist and proving that whether a cause B every single time you're just infer it, and that works 90% of the time. But sometimes it does not work, right? And that's when we end up – that leads us into error a lot of the time. It's not actually a rational way to work to say, oh, every time I see these two things happen next to each other, the pattern means there's causation. Right.

That leads to some predictable bias, which leads people to believe that one thing causes another. And I was like, that particular example of logic on a rye has become kind of a mantra in like stats 101 classes. It's like correlation is not causation. Even as a scientist, you have to fight the feeling that something causes something else sometimes because it is very, the allure of that, that's how we came to be. It's a very functional system for figuring out how things in the world work most of the time. But sometimes it, yes, goes wrong.

But when you're talking about knowledge and belief, it seems that you could say you understand the statistics very well. And so you understand correlation versus causation. But yet you find yourself believing in that something is causation despite yourself, right? Like something about the way our brains work, like almost override our deeper knowledge of how knowledge formation actually should work. Do you find that's the case? That's absolutely the case. So we have some research led by my graduate student who, you know, a Lewis Marti. That is about where your sense of certainty comes from. Um, that is about, uh, where your sense of certainty comes from.

Uh, and when we started that research program, we were expecting for how certain you feel to have something to do with the ground truth of how certain you should feel. Uh, but maybe it was like exaggerated. You got certain a little bit too quickly. Uh, instead, what we found is that, uh, your subjective sense of certainty, how confident you feel that you are, is best predicted by feedback, meaning whether or not you're getting stuff right. So that is not the same thing as how you should rationally assign certainty to things.

If we were to build a model that was going to do the best job that it could coming up with the right answer, but that's not how we would design that system. We find since we have published that paper, I didn't realize how often my decision making was based on my subjective sense of certainty when I now realize that that's not a predictive signal. It happens all the time that somebody in the lab says like, hey, have you seen that robot

that we were going to do some causal learning studies with? And I say, I think that it's in the cubicle downstairs in this room. I'm pretty sure. And I'm like, actually, how sure I am has nothing to do with it. I have no idea. I now back up and say, I have no idea. I can't, my sense of certainty has nothing to do with how. I have no idea. I now back up and say, I have no idea. I can't, my, my, my sense of certainty has nothing to do with how certain I should be. So we're all at a loss. And I think we've all done that. But your, but your sense of certainty is still, I mean, that is part

of your brain or part of your reasoning process coming to a conclusion. And despite your meta awareness of how your reasoning process works, because you are the person who fucking studies it, you still have that sense of certainty and you sort of can't, you can't divest yourself of it. It still arises and it still affects your decision-making no matter what you do. That's correct. It still drives my behavior. It still drives my decisions when I go to the grocery store. It's a really important part of how the system is to have a mechanism that says, you know what, I'm sure enough,

I'm going to disengage and go look at something else. If you didn't have mechanisms in place for triggering you to say good enough and then move away, you'd spend like forever looking at, I don't know, the cheddar cheese in the grocery store. And you'd like never find anything else. You'd learn a lot about all of the cheddar cheese labels and all of the different variations

in colors and textures and prices. But you would never leave the grocery store and that would not be good. So most of the time when we're not in science mode, it's important to be employing these heuristics as opposed to doing what we all think we're not in science mode, it's important to be employing these heuristics as opposed to doing what we all think we're doing, which is like, I'm very sure about everything that I attend to in the world. You're not, and you can't be. And being a little bit more humble about that,

yes, might be a good thing in the world. And it's, well, we have this weird push and pull between our desire to be rational, our desire to base everything off of evidence and to – but then there's also part of us that is built to not do that. That like – again, you're – I'm trying to express this right. You have this meta-awareness of how your mind works and that it's not that rational. And you know if you work very slowly and carefully, you can figure out what your own irrational decision-making is. And you can say, oh, I'm falling into this trap, this fallacy. I'm exhibiting this bias, et cetera. But then at the end of the day, you still have all them in you, in the way your brain is working. And you can't get away from that.

That is correct. We're like two minds at once. Right, right. That thing that you raised, that I have the intuition that we do, but we don't know scientifically whether people do. I want to go back to that for just a second. back to that for just a second, having a meta-awareness, I think, helps me moderate the tendency to make decisions on the basis of it in certain contexts if I slow down. But that is one of the interesting hypotheses you raised that I think is untested in your Mind Parasites show. We don't know the degree to which awareness of the heuristics that we

employ in everyday reasoning and decision making, we don't know whether awareness of them might change people's behavior. Can you overcome in a particular context, the tendency to conclude something that you shouldn't if you're aware of how your system works. Right. That's an open question and I think a really interesting one. Yeah, I make that assertion at the end of the show that we are not, we don't have free will the way we think that we do. Right. We don't just make decisions because that's what's best for us.

We can come under the control of advertising, of, you know, algorithms are designed to shape our behavior, of addiction. Those are the three A's I talk about in the show, three examples. But I say at the end, and I'm really just trying to give a positive spin on it. And I'm taking my best guess here that, hey, if you can be aware of that fact that you are vulnerable to being infected with a mental parasite, you can at the very least avoid situations in which you might be infected. You can have a meta analysis of your own behavior and say, okay, well, uh, I'm going to act knowing that,

uh, my actions can be controlled and maybe avoid them. Um, but yeah, it's true. There's no, there's no evidence for that. Particularly. I'm just like being very hopeful that that's the case. Yet we're going to, we're going to look into it and we're going to find some or we'll find that that doesn't work and we're all doomed. We'll figure something out. It's totally – that was the idea that made me want to talk to you after the show. That was the idea. That was the million-dollar idea. Oh, I would love to have a million dollars from that idea.

We need a couple million probably for like figuring that out. So if any of the grant foundations are listening. Incredible. Yeah, that would be. Well, what are some other, tell me about some of your other results in terms of how we form knowledge and belief. Let's see. That's a wide open. Is that too wide?

That's a, well, yeah, I like it. I like it. So we were talking about heuristics and the role of heuristics in guiding you towards information in the world. Maybe talking about some of the work on curiosity would be helpful at this point. Yeah. Curiosity is a system that is designed to help us in this task and work with my other graduate student, Charlene Wade. We've done some work to try to look at where curiosity comes from,

what induces it, and what we found is that you're not curious when you determine that you know everything about a particular topic. That makes sense because if you already know everything, spending more time on it would just be a waste of time. That opportunity cost. I was like, if I know everything about this particular TV show, I should change the channel because I'm missing out on the other channels. On the opposite end of the spectrum, we also know that if you encounter stuff that is too novel, that's also not a good strategy. When we first started working on this topic and we talked about like what should a good system do, since we don't know how humans work, let's figure out what are the possibilities on the table. I would have thought that the best strategy

is you should seek out as much new information as possible. So just go for the thing that is least overlapping with what you already understand in the world. But when you actually cash that out, the most new information that you could obtain may be too far to be useful. So if I'm picking a TV show and I pick a movie on a topic I don't know in a language I don't understand, it's going to be very difficult for me to actually get anything from that TV show. If I knew the language, I could learn something about the new

topic. If I knew a lot about the topic, maybe I can figure out some of the words I can learn some of the language, uh, but learning too much new stuff at once is intractable. It's not a good strategy because if there's no overlap, you're missing all of the base concepts and levels of representation that you need in order to start integrating it into what you already believe about the world. So curiosity is this cool system that governs the degree to which we are interested in things in the world. And starting even in infancy tends to pull us towards things that are a little bit different from what we currently understand, but not so different that we can't make any progress. Yeah.

Wow. That brings together so many threads for me. That's so cool. It also can make you feel better. You're like, that's why I wasn't attending in that class. It clearly wasn't optimized to where I was mentally. It was like either overly redundant or too novel for me to be able to keep my attention there.

Yeah. too novel for me to be able to keep my attention there. Yeah. I mean, I remember being in classes where I was being told something and I, I couldn't, it was like, I couldn't do anything with the information. I couldn't take it in. It was just going by and I was like, I don't know what to do with this. And then, you know, number of years later, I taught, um, sketch comedy writing, uh, for writing for a couple of years. And I remember telling students something like, here's something you need to know about comedy writing, about like how the sketch should work, you know, about the structure of it. And I'd be feeling I was explaining it very clearly and it would be whizzing by them. I could see that they were not taking it

in. Right. And then simultaneously I was taking improv classes, which I was bad at improv. And I only did it for about three or four years. Cause it turned out to not be my form of comedy. But I remember my teacher in that class trying to tell me, here's what you should be doing when you're improvising. And, and I was like, I don't, I don't know how I want to do what you're telling me to do, but I don't follow, like, I don't get it. And I started to think of it as being like this, like structure that you're like bolting new, this is such a bad metaphor, but it's like, um, like, like almost like a metal sculpture that you're like bolting new pieces onto. And if you don't have like a piece that the new piece goes onto that fits, right. There's no way to attach it. You need to sort of build your way out to it. Um, it needs to be in that sweet spot or you can't, you can't do

anything with it. That is a brilliant metaphor. No, it's not. It makes no sense. A metal sculpture? Yes. What are you talking about? It's very similar. You're being too nice.

I'm not. I'm mean. I'm actually mean. I'm a very mean. I'm not known for my warmth. So that is very similar to a metaphor that Piaget used. He wasn't envisioning a metal sculpture, but he was envisioning a wooden box

and kids trying to, they encounter blocks in the world, which are like the new concepts and things that they encounter. And the box is like a shape sorter that you'd give to a toddler. And they have some number of holes. Those represent the mental concepts,

their way of understanding the world. He envisioned integrating information number of holes, those represent the mental concepts, their way of understanding the world. He envisioned integrating information as being a process of finding a new shape block in the world. And then you have two choices. You can either try to shove it through one of the holes that you have. And that's not quite right, but it works a lot of the time. Or if you encounter something that's so different that it won't go into one of those existing whole shapes, you're forced to like create a new whole, which is like a new concept. So he envisioned a similar kind of process for building up knowledge in the world. It's not that dissimilar from your metal sculpture. I think that's a very useful way to think about it. When you encounter something that doesn't fit, another way of dealing with it is just like leaving that block on the ground.

It's like leaving that piece of that metal hunk not on your sculpture until you have the hardware that you need to attach it. It also makes me think of, for instance, there's so many activities that this makes me think of how curiosity makes things interesting, right? That I think about traveling and the difference between when you travel to a place that's so foreign to you that it's overwhelming versus the experience of traveling to a place where you're fascinated by it and where you want to, oh, wait, I want to know more about this. I want to go to a restaurant. I want to see, you know, I want to go to a restaurant. I want to go to a place of worship. I want to, you know, understand everything about this place where, you know, but I've always felt the more, the more I travel internationally, the more I feel like I'm like leveling up and like learning more about the world. And, you know, like I've been, okay, I've been to, I've been to Japan now. I think I could probably handle China,

right. Right. Cause Japan's a little bit more familiarity because of there's more Japanese culture that I grew up with. So now when I go, I have a lot of curiosity. Wait, I know I understand this, but now I want to understand how that part comes into it, you know? And then once I understand that part of the world a little bit better, maybe if I go to another country I'm less familiar with, like I've thought about, I've thought about it in that way. And I think about how in works of fiction, like a fantasy novel or something along those lines, so much of the time what's pulling you in is your curiosity about some unrevealed part of it.

And sometimes I've tried to read a fantasy novel where it's like, I don't know any of these words. There's so many proper nouns I don't know. I'm overwhelmed. And you either push through or you bounce off and you say, I don't know any of these words. There's so many proper nouns I don't know. I'm overwhelmed. And you either push through or you bounce off and you say, I'm not getting it. But then once you get about 100 pages in, you understand the world and you're like, wait, but what's this character? What's this area?

There's this whole nation that hasn't been described yet. And that's sort of what pulls you further in. That's what keeps you going is wanting to know what that next step is. That's exactly right. And the idea that knowledge builds is an important one for understanding a lot of the most interesting aspects of human behavior. Knowing something about a topic tends to make you want to know more about that topic. So if you're trying to figure out how you could encourage people to be curious about a wide variety of things, they need a wide variety of exposures.

And yes, knowing something tends to make you want to know more. But that's such an optimistic thing to say about humanity, right? To say that there's this deep curiosity system that's just part of how we interact with things. And it's based on once people know things, it makes them more curious. That causes them to try to know. Am I understanding that right?

That's exactly right. And I don't think it's overly optimistic. I think that with a word like curiosity, when people are using that word to refer to a trait, I see it used in a lot of ways that is sort of judgmental. You say like this person is very curious. This person is not very curious. I, you already know, I don't have a lot of respect for words because I know that when we're using them, we don't all mean the same things. But if I'm trying to approximate that usage of the word curiosity,

it's often used to cast judgment on people that have what some people might call more narrow interests. I would call them peakier interests. If you're only interested in... Peakier. Peakier, meaning we don't know that the amount, the drive, the amount of curiosity that you have is different between somebody that's interested in a lot of different topics and somebody that like only wants to know about like basketball and model trains. Maybe they're only interested in those two topics, but chances are they know a lot about those two topics.

So to say that that person is less curious, especially in a judgmental way where you're like, we're going to come in and try to change that, doesn't make sense to me. It absolutely could be the case that different people have different degrees of drive for novel information. But we don't really have good metrics. We don't have any data that says that we should conclude that yet. Also, when people accomplish great things in the world, when people are writing Nobel Prize winning novels, it's often because they had to specialize. So you do stand-up comedy, and I know how much work it takes to get really good at that. If you were devoting a lot of effort in a lot of other places, again, because of the limitations of space and time, you'd be worse at stand-up comedy. So to come in and say, like, this person has less curiosity, like they should be doing something different. It's not obvious to me that we know that.

Yeah. And you can get curious about one, about one topic very deeply. Right. Like there's, cause any, almost any topic has a lot that you can pursue in it. My, my girlfriend Lisa right now is for the last couple of years has been getting back into horse life. Horse life. That's amazing.

She has been taking horseback riding lessons and she now has a horse that she is like raising and training. And it's such a fulfilling – it seems like such a fulfilling hobby because every day there's something new to do. Today we're working on this. And then, oh, wait, I found a new book on how to, on how to sit just right. And if I, oh, if I move my hips forward, it causes this whole change, et cetera, et cetera. And she does that for like two hours, you know, five times a week. And I've been thinking, I don't have anything like that. Like I do stand up comedy that much, but it's not, it's not a not it's not a hobby. It's my career. But apart from that, I read and I play video games and I exercise and I do all these.

I have like manifold different things that I'm maybe interacting with in a more shallow way. But I don't know how to think about which of these is better, which of these approaches, whether they go super deeply or or super wide. Yeah, I wouldn't say you don't have any of these like pockets of extreme expertise. Uh, the, the, the process that you've described for figuring out how to communicate what is funny to people, even speaking fluently, you as a communicator are a very good speaker. Uh, I have tried through our interactions to learn from you. And I just, I don't talk as much. I don't have as much experience talking. Yeah, everybody has pockets of expertise, but I think due to biases that maybe come from school

about like, what is knowledge? What counts? People don't always appreciate them. Attorneys, I have long admired. I'd seen attorneys and the work that they do and admired them when they talk for how fluent they are and how well they are able to quickly integrate information and then make an argument. That's something that I would really like to do. And I mistakenly for years thought that I couldn't be an attorney because I couldn't do that. I've been good at computers.

I started using computers very early. I'm a good programmer. But I wasn't a good speaker. It wasn't for many years before I realized that because I thought I wasn't a good speaker, I didn't do it. And so I never developed it. Lawyers are made. They become better when they start law school. They don't start like that. The training is doing the thing. And by nature of doing the thing, you get better at the thing you specialize. So my speech is not as bad as it used to be. It's much improved,

as shocking as that seems. And it's better now because I've gotten the opportunity to do more of it in the past five years. So I bet you do have, I know that you do have pockets of expertise. Oh, certainly. It's just, it's, curiosity works in so many different ways, I suppose.

That whether it takes you wide or whether it takes you deep. But it does seem to interact with knowledge formation in this strange way because it is certainly good that we're curious. Yet, again, the knowledge that we're forming so often is not true, right? Or subject to biases or – None of it. All of it. All so often is not true, right? Right. We're subject to biases or – None of it. Yeah. All of it.

All of it is not quite right. Actually, wait. Let's return to that point right after the break. I want to get into more about how we should feel about so much of our knowledge being irrational and incorrect. We'll be right back with more Celeste Kidd. I don't know anything. I don't know anything Okay, we're back with Celeste Kidd.

So, Celeste, we've been talking about how we have this natural drive of curiosity that has a sweet spot where it compels us to learn more about the world. spot where it compels us to learn more about the world. And we need to use our attention in order to filter the vast detail of reality in order to understand it. And we also have these heuristics that we apply to reality that help us to come to conclusions more quickly that influence our beliefs and our knowledge. And influence our beliefs and our knowledge. And also our beliefs and our knowledge are constantly changing, right? Correct. Or our concepts are constantly being slightly revised.

That's right. So that's a lot of weird human processes to be using to filter the world, right? So our goal, I've always felt my goal is to understand the world as it is. Like I want to understand, it's going back to philosophy, Kant, the thing in itself was a big concept for Kant. I want to understand what the world is outside of my brain. If the world is a way, I want to know what it is

so I can act accordingly. And because that's like one of the driving values of my fucking life is to, is I want to understand the world, right? That's, that's one of the things that is, I think is makes life worth living, but all these filters that are coming in between, right? Curiosity only works if it's, if it's right in the sweet spot for the knowledge I have already attention, I'm, I'm constantly making subconscious or unconscious decisions of what to attend to, and I'm missing other things. And I'm constantly revising the concepts that I

do have, and not to mention the other hundreds of biases I probably have in my way of understanding the world. So my question is, is it something that I should despair about that? That my my it seems that concepts of knowledge in my head are always naturally going to be so far away from how the world really is because of these layers and layers of filters. And I'm doomed to be walking around the world with incomplete, incorrect, malformed knowledge that does not correspond to what's actually outside of my brain. How should I feel about that? How do you feel about that? You, uh, uh, you should not despair, but that's absolutely true. It's absolutely true. And also you are, you are a liar.

I was like that your life is about, um, seeking truth and trying to figure out how close you can get to, uh, uh, objective truth in the world. You are definitely a scientist and not, you're not like any other you're not like any other comedian that I have ever met. Oh, I appreciate that. Yeah, you know, I was like, I don't know if that's like a compliment to you or I'm insulting other comedians who are great people.

I think a lot of comedians have that same desire. Not every single one, not every single one. Some want to justify their complacency by yelling, but they're like, why do things got to change? That's a whole type of comedian. But a lot of people are like, wait, why are we doing it this way? Like, isn't there a better way to be doing things? And they have a seek and a quest to them.

Yeah, yeah. There is a type of comedy that's very much about, like, you see that thing, I see that thing. The way you're thinking about it is not the only way to think about it. And that intrigues people, I think. So yeah, I don't think there's any reason to despair. It's important. If there's a takeaway there, it's that everybody has approximately the same system doing approximately the same things, we are all going to make errors. So it's really easy to notice somebody else's error and look at it and say like that person, they should have done a better job. Why don't they know that this is how

things work in the world? I can't turn off my tendency to have those thoughts too when I encounter somebody that believes something that I think is obviously false. I now know that for every one of those I notice in someone else, I'm possessing one of those in a different domain that I just haven't discovered. You can't know everything. So me and my sister were raised in the same environment. We encountered the same environment. We encountered the same things. Uh, I, uh, as a, as a younger, more obnoxious child, uh, would sometimes scoff when she didn't understand something about a computer. Uh, but when I went out in the world and had social interactions and they didn't go as expected, she would scoff at me. I know more about computers because that's what I was attending to.

She knows more about people. And I think it's a really interesting strength of our species that we have these systems that leverage the fact that when you know something, you tend to want to know more about it. One of the strengths of our species is that we specialize in ways that other species do not. You can have a comedian, you can have a scientist, you can have a psychologist and a baker and all of the other careers that exist in the world. There's not another animal species that specializes to that degree. And that is our strength. That's why I have an incredible computer and phones.

And I was like, all of the technology's planes are because people specialize to an extreme degree. It allows us at a population level to be able to cumulatively have way more knowledge than even our close evolutionary relatives. There's no chimps creating planes or computers or anything like that. So it's important when you notice that somebody else doesn't know something to appreciate that the fact that we are built that way allows us a really big leg up. It allows us to cumulatively create all kinds of things that other species cannot. So it's important that that guy doesn't understand something about computers. He knows something that you don't.

And all of us as a population are stronger for that. But also some of the things that that person thinks they know about computers or about planes. Let's take a scary example. Yeah. The biggest plane expert in the world. Right. She might know everything. the biggest plane expert in the world, right? She might know everything.

You know, this is the person who goes and speaks before Congress about, or this is the person who's designing the new plane or et cetera. This is the person at the FAA who's regulating the other, you know, the plane manufacturers, right? She might not actually know what she thinks she knows, right?

She might have- Almost certainly that's true. Yeah. That's scary. There's going to be, if the feeling of certainty, which you said earlier, right, is not actually correlated with whether or not the thing you believe is true. Right. Yeah. And I go talk to, you know, the woman who knows more about planes, everybody else in the world. And I say, should I feel safe on this flight right now? And she says, yes, I'm certain of it.

Well, wait, then you say then you say you show me and you talk me through it. I was like, this is why I don't believe anybody about anything. Um, uh, appeals to, uh, confidence, uh, do the opposite, uh, to my, uh, tendency to adopt those beliefs or not. So, um, I should back up. It's not quite as, as dark, uh, as, uh, that last description made it sound, feedback in the world has something to do with ground truth. It's just not totally correlated. So if you have an idea and you see evidence that confirms it, and then you see more evidence, and then you see more evidence,

that's a sign that you might be on the right path. Where things might go awry is when you're getting the evidence from maybe a not randomly sampled place. If you, for example, have an idea in mind, you Google it and you get some YouTube videos that definitely don't represent a random sample of opinions in the world. The people that have YouTube channels often, it's very specialized. They often are creating that content because they're trying to spread information about a minority belief or opinion.

maybe the earth is flat and you go online and see a few videos of evidence that appears to confirm your idea that the earth is flat, the risk is given what we know about our system, you could prematurely become very certain and thus no longer be interested or open to fairly considering other evidence in the world because you think you're done. That system that's designed to prevent us from wasting time on things that we do approximately know can go wrong when you happen to just by chance get a few pieces of confirming evidence. It may more often go awry with new technologies that people are using to accumulate their information. So that curiosity system is not triggered because you've seen too much, you feel like

you've exhausted it in your initial couple YouTube videos that you watched. Right. And you are not open to the actual information. It's almost too late for you. Yeah, that's the problem. The problem is that once you become very certain, it's very hard to shake that certainty from people. And I was like, this is something that we did that study. We thought we were going to design an intervention to shake people of unjustified certainty. And so far, nothing, nothing we've

tried has worked. We've backed up and now we're trying to understand how you might be able to prevent people from becoming unjustifiably certain to begin with. But yeah, we haven't had a whole lot of luck. That's an interesting, that's a really interesting way to put it. Because, you know, we did a segment on Adam Ruins, everything about the moon landing and about why the moon landing could not have been faked. Because the film technology did not exist at the time. I love that argument. I love that clip. So, and I, we made that because I came across that argument from a filmmaker and I was like, this is fascinating.

I've never heard this argument that like literally they did not have, you know, it was a six hour live broadcast. And, you know, a part of the argument that moon landing conspiracy theorists make is that it was slowed down film. That's how they did the jumping, right? That's how they did the slow motion jumping on the moon. They slowed down film. And he said, actually, there's no, there was no technology at the time that could have achieved that on a six hour uninterrupted live broadcast. And he went through all the ways they could have done it. And they're all ludicrous and would have involved NASA inventing film technology that did not arise for 40 more years and then everyone keeping it a secret, which is, I would say, implausible.

At the very least, look, we can have an argument, maybe. But at the very least, it's a very interesting, fun argument. Right. And so we did it on the show and look, I didn't expect to turn around moon landing conspiracy theorists. Right. Um, that was not really an expectation I had, but I was still struck by how much they did not. They, they did not even seem interested in engaging with the argument, um, with the, uh, they were like, oh no, no, this, they were just like, Adam's just a moron. He's saying nonsense. Let's look back at the same shit we've been saying the whole time, you know? Um, and that, and that really struck me. So the idea that, yeah, they wouldn't have curiosity about this other argument is a very,

that seems very apt to me. Yeah. And why would you if you think you figured it out? It is in, under some definitions of what it means to be rational, if you are very confident that you have figured it out, you shouldn't spend more time weighing other evidence. That's intentionally costly. That takes time. that's, that's intentionally costly. That takes time. Uh, it's, it's a rational for people to, um, get, get, get the idea in mind and then walk away. Um, uh, but it's not, uh, yeah, it can go,

it can go awry. Um, and when people are left with the wrong idea, it's very, I don't have a good solution for how you break them out of that. Uh, it's, it's, it's tough. It's easier to try to catch it before they become certain. Um, I can think of interventions, uh, them out of that, it's tough. It's easier to try to catch it before they become certain. I can think of interventions in that domain that have promised that haven't been tested yet. But once people become certain, they are very stubborn with their beliefs and it's very hard to shake them of those beliefs. And it's difficult to know how to,

yeah, occasionally people will break out of that belief, but it's almost hard to figure out why it would happen. I was talking to a friend who quit drinking. I also quit drinking. And she described, you know, how impervious she was to the argument that she should quit, you know, that even after she had said out loud, I have a drinking problem. She's like, I'm not going to do anything about that, though. And then one day, her name's Edith Zimmerman. She's a wonderful writer.

She said it was as though a different self of her, like, peeked out from the clouds and said, you could live a different way. Right. And it's like a mysterious moment to her. What caused the change? Because she was so beyond certain, you know, it was also a physical, physical addiction that was changing her mental processes, as I talk about in my Mind Parasite show. And it was it were we don't really have an understanding of when those moments occur. That's right. And I've talked to conspiracy or people who are like, oh, yeah, I used to believe in this conspiracy theory.

When you read the accounts from folks who write, I was trapped in this way of thinking and I got out. They don't seem to be certain of how it happened either. Right. to be certain of how it happened either. Right. So we know that in order to be curious, you have to know that there's something that you don't know. You have to appreciate that there's a piece of information that you could have that you don't.

You think that that would lead us to be able to figure out exactly how you give that moment to people, but it's very hard. It's very hard to figure out how to make them have that realization. I am a rational thinker, and I think reasonably, and therefore, that almost seems like an extra toxic form of certainty because of your identity, and if you're very sure that that's how you operate, that's an extra hard circumstance to figure out how you could persuade somebody to maybe consider reconsidering some of their beliefs. Yeah.

And that's the, that's like the, I remember reading, actually, we were talking about Ayn Rand earlier, right? Ayn Rand and her followers had this whole thing about like, we are the objectivists. We do everything by logic, right? And there was an essay I read years ago because what ended up happening was they,

was her inner circle exhibited cult-like behavior toward like in the later years. Like it was that people were being excommunicated and shunned and a lot of very irrational things were going on. You know, people were just doing whatever she said or she was investing. You had some some superficial similarities to Scientology or other things like that. It was it was a short lived, I think, period in which it was that. And that doesn't necessarily reflect on her ideas.

If anyone's listening is a fan of her ideas or her writing. I personally am not. But, you know, I'm not going to poo-poo all of that. But the essay was about, well, how is it possible that these people who had defined their lives and their self-identities around we are rational, you know, would end up, like, behaving in the deeply unrational way that, you know, cult-like behavior is where you are taking your orders from authority and you're refusing to see a wider world around you, etc. And so that seems like that can be extra pernicious to have that firm feeling about rationality. Yeah, I think that that's right. And all of these things also interact with other influencers of human behavior. A lot of the things that people do are not about discovering information.

It's about making relationships with people, forming social bonds, figuring out who is part of my group and who is not. How do I value signal in order to attract people that have the same values as me to me, uh, your clothes, the way that you talk, uh, a lot of human behavior is probably happening for, for, for those goals. So gaining social capital, or I was like, I call it, I call it things that people do to feel fancy, um, uh, Twitter behavior, a lot of the Twitter behavior that I would like feel like is irrational. I'm like, Oh no, that's probably some sort of weird value signaling. You're like

getting, getting points with your, your people, um, and, uh, figuring out who your people are. So I ran, uh, that whole history and situation. Uh, I suspect that some of the people's behavior was not oriented towards rationality and figuring out what's true in the world, even though that was the values that they were espousing that brought them together. It was about like impressing each other. And yes, sometimes bad things happen when that's prioritized over everything else. Well, let's talk about, for instance, like, you know, social structure and academia, for instance, because academia is the field that, you know, broadly structure and academia, for instance, because academia is the field that,

you know, broadly, that is, that is the most hyper rational that where everything is, you know, based on peer review and evidence, and you're depending on the different fields, they have different standards, but it's, you know, this is the life of the mind, this is where we try to, you know, we're all here in the pursuit of the truth. But it's also a social world in which, you know, you have dominant personalities and you have people behaving badly and you have social currency and you have all those sorts of things. You've been outspoken about sexual harassment in academia.

I have. You gave a talk that made waves recently, which I watched, which is both about AI and problems on the internet and knowledge formation. But then you also spoke about why men should not have that fear of that fear of, oh, no, like they're coming for us and we're all going to be booted out by the anti-sexual harassment mob. Right. Yeah. I mean, just let me in on a little bit of your thinking about that. Yeah. So academia, not that long ago, women were literally not allowed to be part of that at all. And in a lot of fields, we have a very long way to go in terms of there being representation along a lot of dimensions that are important for really having diversity in those fields.

I was harassed by a professor at the University of Rochester in my first year of graduate school who went on to disrupt the educations of 15 other women after me. I wasn't able to say anything about it because professors who are making decisions about graduate students' lives have the power to completely derail their careers. It's like they control the research funding, they control access to lab resources, and nobody knows who the graduate students are. Whether or not you get to get your PhD or go on to the next level

depends upon the professor saying, this is somebody who's good, this is somebody who you can trust, this is somebody who you should interview. And because of the very strict power hierarchies, the extreme power that professors wield over students, there have been some really bad relatively, thankfully, infrequent. It became clear to me that a lot of my male colleagues who were identifying as allies that were very supportive of me, uh, and the other women when we came forward, uh, they were supportive, but they were also under the impression that minor things were turning into lawsuits. And this made them fearful, even if they were trying to be supportive. So the comments I made at the end of my NeurIPS talk were intending to correct the record on that. it became clear to me that most of the men in academia, most of them had this

incorrect notion that people's careers are sometimes destroyed over minor misunderstandings hitting on women in academia, jokes, trying to be nice. There were a lot of men who told me outright that they were feeling very anxious about interacting with women. They didn't know how to do it. Their hearts were in a good place. They were trying to figure out how to accommodate women, but they were falsely under the impression that they should like give them space or not mentor them because they don't want to like mess it up. And they weren't realizing the asymmetry that creates in terms of educational opportunities, professional opportunities for women.

So I thought I should correct that. And I did. Nobody's career. I was like, if I'm incorrect, it's like if somebody can point me to a professor whose career was destroyed over false allegations or even like dirty jokes or hitting on someone, when they apologize, the people that do very bad things,

they always apologize for something minor. And often that something minor is technically true, but those apologies are lies of omission uh they're apologizing for a thing that absolutely happened hitting on a student making a dirty joke uh and they're not mentioning all of the many many more egregious things that they did that are the reason why there's an internal inquiry or there's a lawsuit they leave out like the assault uh when there assault when there's evidence of assault. So those apologies were misleading all of the other men in the field. And that's not fair to them either. I was like, that's not fair for men in academia to be falsely under the impression that

your career could be destroyed over a minor misstep. It's not fair to the junior women that still are really not represented at the senior ranks that really, really need equal access to opportunities to meet with their male professors and mentors that need to be included in social gatherings where there's going to be talk of science. Women need to be there too. And the men also can benefit from having women be part of the conversation. Every dimension along which you increase diversity, you increase the hypotheses that are on the table. It leads to higher rates of innovation. It's better for everybody to have as many people from as many backgrounds at the table. And in

academia, we have a really long way to go. Yeah. I mean, I, look, in my own work, right, I've experienced, there are so many topics that we would not have been able to do that I'm so glad that we did, that we were not able to do. We would not have been able to do if we didn't have, you know, people of diverse backgrounds and of all types of diverse backgrounds in the room, because we wouldn't know the things, we wouldn't have had the conversations. Not because, oh, white men can't talk about this. Not that, but because the idea would not have been brought up. That's correct, yeah.

No one would have pitched the idea. No one would have had the experience that led to the thought. And so it manifestly makes the work better. I can only imagine the same thing is true if you're talking about researching human behavior. Exactly. You need all of the kinds of humans from all the backgrounds in the room.

I was like, this conversation is fun because now it's looping back around. Everybody knows different stuff. None of it is quite right. But you need, if the goal is getting at what is true in the world, you need all of the different variations on concepts represented in the room to make progress towards that goal. That's really important in all fields, but especially in science or any of the disciplines that are aimed at getting at truth. You need poor people and rich people and you need people of all ethnicities.

You need men. You need women. You need everyone in that room because all of those dimensions are going to tend to be coming from all of those different dimensions affect the experiences that you have and the experiences you have are directly related to the beliefs that you have. Yeah. And is there an asymmetry? Those men in academia who don't understand what is actually at stake or don't understand what these issues actually are. Is there an asymmetry in their knowledge that is coming from the academic worldview?

Is there something about, again, the commitment to rationality or understanding that is getting in the way of addressing that or seeing it? When I read things that Steven Pinker has written and tweeted on the topic, I would say yes. Pinker has written and tweeted on the topic, I would say yes. He's a great example of somebody who prides himself on being a rational intellectual thinker, but fails regularly, pretty catastrophically, to consider the perspectives of other people. Mm-hmm. Yeah, that idea of I am I am the uber rational person and I have thought everything through and therefore my pronouncements on any issue are, uh, are therefore rational are, is a, is a problem among, uh, the, the most eminent statesmen of academia, you know, the folks

like that's, I, I relate to that criticism, certainly. But I also, it's so strange because I, the idea that, oh, anything could happen to you. You know, like, people are being attacked for anything now. And men have to walk around in eggshells. I mean, look, the entertainment industry, all industries are going through this transition at the same time. We're all having the same realizations, you know, and it's expressed in different ways in the entertainment industry versus academia, for instance. But like I work with – I have women working under me and I have now as I did then like a commitment to mentoring and to like doing my part to, to bringing more folks in. Right. I've never felt that I, uh, that, that I was going to be attacked for something spurious or that my career could be destroyed if I say the wrong thing and

that therefore I have to keep my distance. I've never felt fear around it. And, and, uh, that doesn't seem that hard to not sexually harass in the workplace. Like, am I alone in this? Well, so your access to that truth, I would say that that is much closer to the truth. That is definitely closer to the truth of the matter. One of the points that I made at the NeurIPS talk is the amount of effort and risk somebody has to take to levy a complaint against a professor is huge. I was like, the chances are, most likely outcome for filing a complaint is that you will be retaliated against and destroyed and nothing else is going to happen. So the- And that continues to happen now.

That continues to happen now. Nothing typically happens to the professor at, um, uh, the, the, the, uh, and that continues to happen now that continues to happen now, nothing typically happens to the professor at all. And in our case, uh, the professor, uh, sent an unwanted picture of his genitalia to a student. I was like, there was no question about what happened. Uh, he, uh, uh, had loud sex within earshot of students who didn't want to hear but could. There's a whole laundry list of things that are not debated as to what he did. He still has his job. Nothing happens most of the time.

The effort it takes to prepare and then file a lawsuit, it takes a lot of money. And it's on par with completing a PhD dissertation, but it's a lot more emotionally exhausting. So this is a message that I was worried might be interpreted the wrong way, but it doesn't seem like it was. What I tell my male colleagues that are anxious about this is unless you are deeply doing wrong by the women around you, you are incredibly safe. So you can feel free to interact with them. I would guess that your access to what is true might directly hinge on the fact that you are interacting with women. If you're in a field where there just aren't any around, uh, if, uh, you are not

mentoring them, uh, out of, um, yes, there's no way to convey, uh, my sarcasm. So I will just say it. I was like, I, I had, uh, professors that, uh, were good hearted people that with a totally straight face seriously said to me, uh, after with our complaint about what had happened to us in graduate school. I had professors that said, you know, I'm so sorry for what you went through. I had no idea. This is why I don't mentor women students. I would never want them to feel uncomfortable. I don't know about their experiences. And I was like, wait a minute, you uncomfortable. I don't know about their experiences. And I was like, wait a minute, back up. You don't mentor women's students. I was like, yes, I don't, I don't, I didn't know about your experience. I don't know what women go through. And so I don't want to mentor them.

And I'm like, I, um, uh, wait a minute. Um, uh, that was, that's out of respect for the women. Like that's, you are not, that is discrimination. Uh, people are sensitive sensitive i don't accuse them that's discrimination it's discrimination and if i back them up very gently and say like okay say what you just said but say like i don't mentor like name another minority i don't mentor black people and they're like oh that's racist like yes it is now now no say what you just said they're like i don't mentor women that's sexist like That is sexist. And I understand that they didn't understand that it was sexist. But that's why there's no women in, in senior positions at universities is because of this just failure to appreciate

things that are true in the world. And you can't, if you don't have that, that experience. I mean, let's bring it back to knowledge formation. Like, like perhaps, perhaps the reason that I don't have fear is because I, I work with many women who if I look, we do a comedy show, we all make jokes. I'm sure I've made a joke that is insensitive to the experiences of women because I happen to not be one. However, the women who we have in the room make another joke in which they like push back and they say, oh yeah, you're an asshole. Right. In the context of it, because we're comedians. Right. And like, we're, we're going back and forth in that way. And because, because we have that open line of communication, right. And, and of course we have communications where people say, Hey, when that thing was said, like, that wasn't a great thing

to be said, you know, and could we be mindful of that in the future? And we say, yes, of course. Oh, of course. You know, because we respect each other and we communicate like human beings. But you never work with a woman as a peer who is able to give criticism or feedback or who you respect enough to care about when they give those things. That might be if you're cut off to that type of person, then you might have fear about, oh, wait, what if I say the wrong thing? I don't know what's going to happen because you're so unfamiliar with those people's needs and way of being. And you have no empathy because you're not around them. So you don't have the knowledge and therefore you are wrong. That's right.

That's exactly right. People form beliefs with respect to what they have access to. And they form beliefs constantly based on the evidence they see. If you're a senior dude professor hanging out doing your evolutionary psychology and you don't see a lot of women around, all people are constantly looking for causal explanations as to how the world works. It might be understandable for you to say like, oh, there's no women here. I've never observed anything overtly sexist. I don't think that I am sexist. It must be that they don't want

to be here. I'm going to wave my hands around and come up with an evolutionary theory for why they're not here. I'm going to publish that in a journal that's reviewed by people that have my perspective, that share my perspective, that will think this is super cool. And I'm – even though I don't – that's wrong. That's not right. Women want to be in senior levels in all disciplines that I have had contact with. And there are systematic things that keep them out. that keep them out.

I am sympathetic with how an old white dude at a university could be left with that belief given that there's no women around to correct the record. They're doing the best job that they can. They're wrong, but they're doing the best job that they can with the information they have.

I think the era of Twitter, I was like the people having a voice bigger than they did earlier in their careers than they did is one of the powerful forces that's turning things around. The solution ultimately needs to be we need representation at the senior boss levels. Yeah, I relate to that. I mean, when I hear you still hear older comedians say, oh, less women want to be comedians. That's why there's less women comedians. I'm like, oh, you haven't been to like an open mic in 30 years.

You haven't been to an improv class. You haven't like actually met anyone who's trying to be a comedian because there's a million of them. And they'll tell you what the issue is. You know what I mean? They'll be like, yeah, it's tougher. And if you have those conversations, you understand that. And you can like work to correct that. But if you're just like

you know, performing in theaters and bumping into people in the green rooms of late night talk shows, you might be like, oh yeah, there's no women around. I guess they just don't feel like it. They like having babies and I don't know, being pressed. They're not here because they're doing it. They don't want to be it. They like having babies and I don't know, being pressed. They're not here

because they're doing, they don't want to be here. They opted out. I love how we've connected this to the original topic of like how you form that knowledge because it's what you experience has such a direct impact

on the knowledge that is in your brain. Yes, that's exactly correct. It all loops around. Let's just go back to your field for one last question. What are you most excited about in the future of your field? There's so many open questions on what you're studying, but what makes you really, really excited that this is what's right over the horizon? excited that like this is what's right over the horizon oh my goodness so many things uh i am very excited with uh the new interest in uh connecting ai research with human belief formation research um ai uh is a field that's trying to create intelligence you might think that they would

be like better versed in psychology and cognitive science research and neuroscience research than they are. But there's been some borrowing and a little bit of crosstalk over the years. But AI and computer science up until very recently had been very, very separate. There's starting to be more meaningful connections and collaborators between psychologists and neuroscientists and AI researchers. And why I think that's interesting is because the big vision, if you want to create intelligence, you want to create something that is going to interface well with humans. I was like, you need to make sure that you're designing systems that are not going to accidentally amplify incorrect beliefs or biases that people will tend to have as they're interacting with things like Facebook and Google. But also,

you want to design intelligent systems that can do better than people do. How cool would it be to have a system that could alert you as to when you are maybe developing premature certainty and should hold off for more information? All of this is not, we're not near doing anything like this. Self-driving cars still can't tell the difference, still don't know that they need to like press on the brakes for a toddler, but they had not, not, not, not a bird. A bird will get out of the way. We're not there yet, but thinking about people in formal terms

and then designing systems that are influenced and better than people. You're blowing my mind because it's like almost one of the strange things that human minds do is we create models of other minds, right? That like here's a being that's acting in a certain way and I can expect it will act in a certain way because it's got a mind in there too and I can make predictions about it. And a very basic one of those is, yeah, you don't need to hit the brakes when you're driving towards a pigeon. Right.

Because you know that pigeons get out of the way. But you're like cats, maybe I could hit a cat. You could. And then a baby I'm definitely going to hit. Right. Because you know that pigeons get out of the way, but you're like cats, maybe I could hit a cat and then a baby I'm definitely going to hit. Right. And so for AI, it doesn't just need to model like, here's what I am approaching, which it's already bad at or not good enough for production currently. It also needs to model what other minds are going to do, what actions they might take, not to mention the minds in the other cars. Yeah. So it's –

You don't think about that as a theory of mind task, but it absolutely is. If you slam on the brakes for a bird, the driver behind you won't be expecting it, and you will hit a car accident. And, my God, we talk about self-driving cars a lot on this show, but it really ties back to everything that you said at the beginning, because it's a situation where driving is a sneakily, it seems like a rule-based system. Like, especially in the U.S., we do follow rules mostly. So it almost seems like it's just like logic gates opening and closing and like everybody's like following the rules, right? But then it's also happening in reality.

following the rules, right? But then it's also happening in reality. And as you've said, reality is almost infinitely detailed. An almost infinite number of strange things can happen in reality that because humans are attuned to reality, we were evolved to be in reality, we can often compensate for. There's the long tail of edge cases of strange, spooky things that can happen. The different surface on the roads, you know, like the, oh, it's slightly, there was a slight misting of rain this morning. So my tires might work differently. Like all these different tiny

little bits of information that are so difficult to program into an artificial intelligence. And I think that's something else so interesting about artificial intelligence is that now the way that they create so many of them is via these neural networks. And the people who create them actually don't know why they come up with the solutions that they come up with, right?

Right. I don't like, I was saying that too. And then I said that to Ode Oliva, who is a professor at MIT and the director of the MIT, the IBM Watson research group over there. And she chastised me because I said, I was like, oh, these systems, they're black boxes. And she said, brains are black boxes. And that's what we do for a living is we probe them. Of course, you can dive in and understand the black box. And to say that is kind of a cop out.

So she's totally 100% right. I have since realized that these, you know, deep learning is employing algorithms that people have sort of, when they've encountered a problem, thrown up their hands and said like, oh, too bad that this algorithm is being racist, but there's nothing that we can do about that. That's not really a thing. It's opaque. But it's interesting that we're creating, when you're saying the confluence between artificial intelligence and brains and like those researchers coming together, when we're, some of these artificial intelligence systems, we create them and then we actually need to study them to understand how they work.

That's so cool. But yes. And that is also true of brains. Yes. So that might actually be, I'm down on AI a lot on this show. And look, honestly, that is what I am down on is corporate America claiming that AI can do things that it can't do. And you should be down on that.

Yeah. Exactly. I'm behind you. In ways that might hurt people, kill people, or make our lives worse. Right. I think that I'm down on the culture of the tech industry around AI.

I'm not down on the research. The research is fascinating. And so one of the most positive things you could say about it is that, like what's proof that it's making advances? It's creating things that are approaching the complexity of our own brains in that we don't fucking understand them.

We don't understand our brains and we don't understand artificial intelligence. So in that way, artificial intelligence seems more human intelligent like because we understand it less. We're getting close. I think that I think you're totally right. I think that's a sign that we're making some kind of weird progress. Our artificial intelligence systems also what they do depends upon the data that we give them. And there is no system that's taking in data and statistics from so many different places like humans do. And I think that's part of the difficulty and also the

interestingness is coming from that nuance. It's interesting that you gave up the self-driving cars example. I find it interesting that if there was going to be a set of rules that you follow for driving, like maybe the DMV manuals would be in those. I went to undergraduate in L.A. I did my graduate degree in upstate New York at, in Rochester, New York, near the Canadian border. In both of those places, I checked the DMV manuals. They say approximately the same things for like how you should treat stop lights. You should stop definitely when it's red. And when it's green, you go. And if it's yellow, you should

slow and not go into the intersection. But people's actual behavior was totally different in those two places. Yes. Uh, in Rochester, New York. Do you know this? Cause you were also in snowy places. I actually never drove in New York. So, yeah.

So when the light turns green in Rochester, New York, I just moved there from LA. Um, and I noticed immediately, uh, that, uh that when the light turns green, people don't go. They hang out for just a beat. And this was very frustrating for me. If you do that in LA, you get honked at. In Rochester, it was clear that that was the standard. So I found myself sitting behind cars at green lights that wouldn't go for a moment. And then the snow started falling and I had the experience of driving again.

Light turns green. The car in front of me didn't go, and a car went sliding through the intersection after because you can't stop suddenly when the roads are icy like you could in L.A. It did not mention this in the New York state DMV manual. Um, but, uh, everybody converged on, uh, without, I think any conscious realization, uh, the rule that you're not going to go as soon as you see it be green, because you're going to see if there's a car that tried to stop that then slides through. Um, uh, if you go at green, then you're going to hit them.

You'll cause an accident. Everybody knows that. I don't think that it's conscious. Everybody has learned from the statistics of their environment. And even though the rules that they were taught say the same thing on that issue, the behavior is totally different. And it's adaptive to what is the best course of action given the environment. And this is why when people say, well, humans are such bad drivers, that's why we need AI as soon as possible. Well, actually, that is a very example of humans

being especially skilled at this task, right? Because it's a localized, specific piece of reality detail, this infinite detail that reality has, that the driving culture of Rochester, New York has adapted for and like put in place a cultural norm, which resulted in loss of less loss of life more generally. That's, that's right. That's right. And I worry about, um, uh, the fact that there are so many of these instances, there's one that I noticed that's a difference in behavior, um, from, uh, people in Rochester, New York versus, uh, people in LA that's clearly difference in behavior from people in Rochester, New York versus people in LA that's clearly adapted to the circumstances of the locale.

I don't know if people working on self-driving cars are aware of those differences. I don't know the degree to which those sort of differences might be integrated. If you have a self-driving car that was designed in L.A. that is interacting with human drivers in Rochester, New York, it's not going to behave in the same way that is in accordance with the expectations of the drivers around it. Right. And that's – I mean Teslas are designed in not quite Los Angeles but in SoCal. Right. And I'm sure there are Teslas in Rochester, New York. And right now they're just doing the lane assist, whatever.

But that's, I mean, that's a real problem. I saw never, I never saw one. I was like, and I wonder, I like, I was looking, I was looking. Oh, there's some, oh, there's some, there's some rich guy in Rochester, New York for sure. But yeah, how, it's such an interesting question. How can, like a self-driving car also has to account for culture? There's this wonderful book called Traffic about how our traffic systems work. I'm blanking on the name of the author, but it's called Traffic.

And he talks about traffic culture and how in the United States, people say, oh, people drive so crappy in LA or whatever. We actually have a very rule-following traffic culture. People stop for stoplights even if you're driving at 2 a.m., the biggest asshole in America at 2 a.m. in the middle of a country road sees no one coming for half a mile in any direction will stop and wait for the green light, right? It is a really built in cultural norm. Other countries don't have that norm, right? Not like some other countries don't have that norm. And ones that don't have higher rates of traffic deaths. Right. Or there's there's different way. When I was in Mumbai a number of years ago, the neighborhood I was in was the traffic culture was entirely there were no signals of any kind. It was entirely is a car going to hit me? If no, if yes, I must stop. And if no, I can go. And it was entirely making eye contact with other people and saying, I'm going to go. Right. It was very touch and go. There was no

rule following. And that is completely different, right? That culture affects so much. And it's another example of how much detail there is and how difficult that is to handle. That is so fascinating to see how those two, because a factor of human brains is that they are social and that they communicate with each other and that we change our behavior based on what other human brains are doing. And so seeing how that interacts with artificial intelligence, that is very exciting.

Right. And I think that's an important thing to keep in mind as we're designing AI systems that are interacting with humans. I see a lot of talk in papers that seems to suggest that the authors are understanding people as largely being all approximately the same when that's really not true at all. Our strength as a species is that we can hyper specialize.

We are substantially different from each other, even in domains where we like laid out a set of rules and they appear to match. There's a lot of diversity in terms of what we believe and what we believe directly impacts what we do. Yeah. And the AI systems that are interacting with humans, it's very important that they take that under consideration. Well, your work is so fascinating. I really appreciate you coming on the show to talk to us about it.

I love talking to you, Adam. I was like, we're going to keep doing it. That's so this was a pleasure. And thank you so much for having me on. This was a really fun conversation. Thank you so much. I was like, that's, that's, um, uh, so, uh, this, this was a pleasure. Uh, and, uh, thank you so much for having me on. This was a really fun conversation. Thank you so much. I've had a blast. Well, thank you once again to Celeste Kidd for coming on the show.

I was fascinated by that conversation. I hope you enjoyed it as much as I did. That is it for us this week on Factually. I want to thank our producer, Dana Wickens, our engineer Ryan Conner, our researcher Sam Roudman, Andrew WK for our theme song. You can follow me on Twitter at Adam Conover. You can sign up for my mailing list

or check out my tour dates at adamconover.net. And until then, we'll see you next week on Factually. Thank you so much for listening.