Lisa Feldman Barrett: Counterintuitive Ideas About How the Brain Works | Lex Fridman Podcast #129

Lex Fridman 0:00 The following is a conversation with Lisa Feldman Barrett, a professor of psychology at Northeastern University, and one of the most brilliant and bold thinkers and scientists I've ever had the pleasure of speaking with. She's the author of a book that revolutionized our understanding of emotion in the brain called How Emotions Are Made, and she's coming out with a new book called Seven and a Half Lessons About the Brain that you can and should pre-order now. I got a chance to read it already, and it's one of the best short whirlwind introductions to the human brain I've ever read. It comes out on November 17th, but again, if there's anybody worth supporting, it's Lisa, so please do pre-order the book now. Lisa and I agreed to speak once again around the time of the book release, especially because we felt that this first conversation is good to release now, since we talk about the divisive time we're living through in the United States leading up to the election. And she gives me a whole new way to think about it from a neuroscience perspective that is ultimately inspiring of empathy, compassion, and love. Quick mention of each sponsor, followed by some thoughts related to this episode. First sponsor is Athletic Greens, the all-in-one drink that I start every day with to cover all my nutritional bases that I don't otherwise get through my diet naturally. Second is Magic Spoon, low-carb, keto-friendly, delicious cereal that I reward myself with after a productive day. The cocoa flavor is my favorite. Third sponsor is Cash App, the app I use to send money to friends for food, drinks, and unfortunately, for the many bets I have lost to them. Please check out these sponsors in the description to get a discount and to support this podcast. As a side note, let me say that the bold, first-principles way that Lisa approaches her study of the brain is something that has inspired me ever since I learned about her work. And in fact, I invited her to speak at the AGI series I organized at MIT several years ago. But as a little twist, instead of a lecture, we did a conversation in front of the class. I think that was one of the early moments that led me to start this very podcast. It was scary and gratifying, which is exactly what life is all about. And it's kind of funny how life turns on little moments like these, that at the time don't seem to be anything out of the ordinary. If you enjoy this thing, subscribe on YouTube, review it with five stars on Apple Podcasts, follow on Spotify, support on Patreon, or connect with me on Twitter @lexfridman. And now, here's my conversation with Lisa Feldman Barrett.

Lex Fridman 2:45 Since we'll talk a lot about the brain today, do you think ... Let's ask the craziest question. Do you think there is other intelligent life out there in the universe?

Lisa Feldman Barrett 2:54 Honestly, I've been asking myself lately if there's intelligent life on this planet.

Lex Fridman 2:58 Oh, yeah.

Lisa Feldman Barrett 2:59 Uh, you know, I h- I th- I have to think probabilities suggest yes, and also, secretly, I think I just hope that's true. It would be really, um ... I know scientists aren't supposed to have hopes and dreams, but, uh, I, I think it would be really cool, and I also think it would be really sad if it, if it wasn't the case. If we really were alone, that would be ... That, that would seem profoundly sad, I think.

Lex Fridman 3:29 So it's exciting to you, not scary?

Lisa Feldman Barrett 3:32 Yeah, no, you know, I take a lot of comfort in curiosity. It's a great, it's a great, um, resource for dealing with, uh, stress. So-

Lex Fridman 3:45 (laughs)

Lisa Feldman Barrett 3:46 Um, I'm learning all about mushrooms and, uh, octopuses-

Lex Fridman 3:49 Yeah.

Lisa Feldman Barrett 3:49 ... and, you know, all kinds of stuff. Um, and so for me, this counts, I think, in the realm of awe. But also, I think I'm somebody who cultivates awe deliberately on purpose to feel like a speck, you know? I, I find it a relief occasionally.

Lex Fridman 4:07 To feel small.

Lisa Feldman Barrett 4:08 To feel small in a profoundly large and interesting universe.

Lex Fridman 4:13 So, maybe to dig more technically on the question of intelligence, do you think it's difficult for intelligent life to arise like it did on Earth? From everything you've written and studied about the brain, how magical of a thing is it in terms of the odds it takes to arise?

Lisa Feldman Barrett 4:33 Yeah, so, you know, magic is just-

Lex Fridman 4:37 (laughs)

Lisa Feldman Barrett 4:39 Don't get me wrong. I mean, I-

Lex Fridman 4:40 (laughs)

Lisa Feldman Barrett 4:40 ... I like a m- I like a magic show as much as the next person.

Lex Fridman 4:43 Yeah.

Lisa Feldman Barrett 4:43 I, my husband was a magician, uh, at one time, but, uh, you know, magic is just a bunch of stuff that we don't really understand how it works yet. So I would say, from what I understand, there are some major steps in the course of evolution that, at the beginning of life, the step from single cell to multicellular organisms, things like that, which are really not known. I think for me, the question is not so much, um, could it ... you know, what's the likelihood that it would happen again, as much as, um, what are the steps and how long would it take? And if it were to happen again on Earth, would, would we end up with the same, you know, menu of life forms that we currently have now? And I think the answer's probably no, right? There's just so much about evolution that is stochastic and driven by chance.

Lex Fridman 5:43 But the question is whether that menu would be equally delicious, meaning, like, there'd be rich complexity of the kind of ... Like, would we get dolphins and humans, or whoever else falls in that category of weirdly intelligent, seemingly intelligent? However we define that.

Lisa Feldman Barrett 6:01 Well, I think that has to be true if you just look at the range of creatures who've gone extinct. I mean-

Lex Fridman 6:07 Right.

Lisa Feldman Barrett 6:08 ... if you look at the range of creatures that are, are on the Earth now, it's incredible, and, you know, it's sort of trite to say that, but it actually is really incredible. Um, particularly... I don't know, I mean, animals... There are animals that seem really ordinary until you watch them closely and then they become miraculous, you know, like certain types of birds which do very miraculous things. Uh, um, build, you know, bowers and do dances and all these really funky things that y- are hard to explain, uh, with a standard evolutionary story. Although, you know, um-

Lex Fridman 6:45 Yeah. The-

Lisa Feldman Barrett 6:45 ... people have them.

Lex Fridman 6:46 ... birds are weird. They do a lot of... For mating purposes, they, they have a concept of beauty-

Lisa Feldman Barrett 6:52 Well-

Lex Fridman 6:52 ... that I haven't quite... Maybe you know much better, but it doesn't seem to fit evolutionary arguments well.

Lisa Feldman Barrett 6:58 It does fit. Well, it depends, right? So I think you're talking about The Evolution of Beauty, the, um, book that was written recently by, what is it, Fromm, fr- um, is that his name? Richard Fromm, I think, at Yale.

Lex Fridman 7:10 Oh, interesting. No. I didn't... Oh.

Lisa Feldman Barrett 7:11 Oh, it's a great book. It's very controversial though, because he is argu- he's making the argument that the, the question about birds and some other animals is why would they engage in such metabolically costly, um, displays when it doesn't improve their fitness at all? And the answer that he gives is the answer that Darwin gave, which is sexual selection. Um, not natural selection. But, you know, selection can occur for all kinds of reasons. There could be artificial selection, which is when we breed animals, right, which is actually how Darwin... That, that observation helped Darwin come to the idea of natural selection.

Lex Fridman 7:50 Oh, interesting.

Lisa Feldman Barrett 7:51 Um, and then there's sexual selection, um, meaning... And the argument that, that, um, I think his name is Fromm, uh, makes is that, um, that it's the pleasure... The selection pressure is the pleasure of female birds. Which, as a woman and, um, as someone who studies affect, that's a great answer.

Lex Fridman 8:12 That's okay.

Lisa Feldman Barrett 8:12 I actually think there probably is natu- I think there is an aspect of natural selection to it, which he maybe hasn't considered.

Lex Fridman 8:18 But you were saying, uh, the reason we brought up birds is, uh, the, the life we've got now seems to be quite incredible.

Lisa Feldman Barrett 8:24 Yeah. So you peek into the ocean, peek into the sky-

Lex Fridman 8:27 Oh, yeah.

Lisa Feldman Barrett 8:28 ... there are miraculous creatures. Look at creatures who've gone extinct and y- you know, in science fiction, uh, stories, you couldn't dream up something as interesting. So my guess is that, you know, intelligent life evolves in, in many different ways even on this planet. Uh, there isn't one form of intelligence. There's not one brain that gives you intelligence. There are lots of different brain structures that can give you intelligence. So my guess is that the menagerie might not look exactly the way that it looks now, but it would certainly be as, as interesting.

Lex Fridman 9:04 But

Lex Fridman 9:05 if we look at, um, the human brain versus the brains, or whatever you call 'em, the mechanisms of intelligence in our ancestors, even early ancestors, that you write about, for example, in your, in your new book, uh, what, what's the difference between the, the fanciest brain we got, which is the human brain, and, uh, the ancestor brains that it came from?

Lisa Feldman Barrett 9:31 Yeah, I think it depends on how far back you wanna go. You know?

Lex Fridman 9:34 You go all the way back, right? In y- (laughs)

Lisa Feldman Barrett 9:36 Well, I... (laughs)

Lex Fridman 9:37 ... in your book. But, uh, so (laughs) what's the interesting comparison would you say?

Lisa Feldman Barrett 9:41 Well, first of all, I wouldn't say that the human brain is the fanciest brain we've got. I mean, an octopus brain is pretty different and pretty fancy, and they can do some pretty amazing things that we cannot do. You know, we can't grow back limbs, we can't change color and texture, we can't comport ourselves and squeeze ourselves into a little crevice. I mean, these are things that we invent. These are like superhero abilities that we invent in stories, right? We can't do any of those things. And so the human brain is certainly, um... We can certainly do some things that other animals can't do, um, that seem pretty impressive to us. Um, but, but, uh, I would say that there, there are a number of animal brains which seem pretty impressive to me, that can do interesting things and really impressive things that we can't do.

Lex Fridman 10:29 I mean, with your work on how emotions are made and so on, you, you kind of repaint the, the view of the brain as, um, as less glamorous, I suppose, than, (laughs) than you would-

Lisa Feldman Barrett 10:41 Well-

Lex Fridman 10:41 ... otherwise think.

Lisa Feldman Barrett 10:42 No.

Lex Fridman 10:42 Or, like, I, I guess you draw a thread that connects all brains, uh, together in terms of homeostasis and all that kinda stuff.

Lisa Feldman Barrett 10:50 I, yeah, I wouldn't say that the, that the human brain is any less miraculous than anybody else would say. I just think that there are other brain structures which are also miraculous. And I also think that there are a number of things about the human brain which we share with other, other vertebrates, other animals with backbones, but, um, that are m- that we share these miraculous things. But we can do some things in abundance, and we can also do some things with our brains together, working together, that other animals can't do, or at least we haven't discovered their ability to do it.

Lex Fridman 11:31 Yeah, the social thing. How... I mean, that's one of the things you write about. Uh, what's, uh... How do you make sense of the fact, uh, like, the book Sapiens and the fact that we're able to kinda connect, like network our brains together, like you write about. I'll try, I'll try to stop saying that. (laughs) Uh, w- is that, is that like some kind of feature that's built into there? Is that unique to our human brains? Like, how do you make sense of that?

Lisa Feldman Barrett 12:02 What I would say is that our ability to coordinate with each other is not unique.... um, to humans. There are lots of animals, uh, with- who can do that, and we, um ... But what we do with that coordination is unique because of some of the structural features in our brains. And it's not that anim- other animals don't have those structural features, it's we have them in abundance.

Lex Fridman 12:33 Mm-hmm.

Lisa Feldman Barrett 12:33 So, you know, the human brain is not larger than you would expect it to be for a primate of our size. If you took a, a chimpanzee and you ex- you grew it to, uh, uh, the size of a human, that chimpanzee would have a brain that was the size of a human brain. So there's nothing special about our brain in terms of its size. There's nothing special about our brain in terms of the, um, the basic blueprint that builds our brain from an embryo, is the basic blueprint that builds all mammalian brains and maybe even all vertebrate brains. Um, it's just that because of its size, and particularly because of the size of the cerebral cortex, which is the, um, a part, um, that people mistakenly, uh, attribute to rationality. (laughs) You know?

Lex Fridman 13:29 Why mistakenly?

Lisa Feldman Barrett 13:30 Well-

Lex Fridman 13:30 Isn't that all, where all the clever stuff happens?

Lisa Feldman Barrett 13:32 Well, no, it really isn't. And I will also say that lots of clever stuff happens in animals who don't have a cerebral cortex, but-

Lex Fridman 13:39 Right.

Lisa Feldman Barrett 13:40 ... um, but, uh, but because of the size of the cerebral cortex and because of some of the features that are enhanced by that size, that gives us the capacity to do things like build civilizations, um, and coordinate with each other, not just to manipulate the physical world, but to add to it in very profound ways. Like, you know, other animals can cooperate with each other and use tools. Um, we draw a line in the sand and we make countries and we even, then we create, you know, uh-

Lex Fridman 14:18 Well-

Lisa Feldman Barrett 14:19 ... we create citizens and immigrants.

Lex Fridman 14:20 But also ideas. I mean, the-

Lisa Feldman Barrett 14:22 Well-

Lex Fridman 14:23 ... countries are centered around the concept of, like, ideas is-

Lisa Feldman Barrett 14:26 Well, my, well, what do you think a citizen is and, and an immigrant? Those are ideas.

Lex Fridman 14:30 Right.

Lisa Feldman Barrett 14:30 Those are ideas that we, um, impose on reality and make them real, and then they have very, very serious and real effects, physical effects on people.

Lex Fridman 14:41 What do you think about the idea that a bunch of people have written about, Dawkins with memes, which is like, ideas are breeding ... Like, we're just like the canvas for ideas to breed in, uh, in our brains, so that this kind of network that you talk about of brains is just a little canvas for ideas to then-

Lisa Feldman Barrett 15:01 Yeah, I-

Lex Fridman 15:01 ... compete against each other and so on?

Lisa Feldman Barrett 15:02 I, I think as a rhetorical tool, it's cool to, uh, think, you know, think that way, so, um, I think it was Michael Pollan, I don't remember if it was in The Botany of Desire, but it was in one of his early books on, um, on, um, botany and gardening where he, uh, wrote about, um, anim- uh, he wrote about, uh, you know, plants y- sort of utilizing humans for their own, you know-

Lex Fridman 15:30 (laughs) Yeah.

Lisa Feldman Barrett 15:30 ... evolutionary purposes-

Lex Fridman 15:32 Yeah.

Lisa Feldman Barrett 15:32 ... which is kind of interesting. You can think about a human gut in a sense as a, uh, propagation device for the seeds of-

Lex Fridman 15:41 Right.

Lisa Feldman Barrett 15:41 ... you know, tomatoes and what, what have you, so it's kind of cool. Um, so I think, I think rhetorically, it's an interesting device, but, you know, ideas are, as far as I know, um, invented by humans, propagated by humans. Um, so, you know, I, I don't think they're separate from human brains in, in any way, although it, it would be, it is interesting to, to think about it that way.

Lex Fridman 16:06 Well, of course, the ideas that are using your brain to communicate and write excellent books, uh, and they basically picked you, uh, Lisa, as a, an effective communicator, and, and thereby are winning. So, that's an interesting worldview (laughs) to, to think that there's particular aspects of your brain that are conducive to certain sets of ideas, and maybe those ideas will win out.

Lisa Feldman Barrett 16:33 Yeah, I think the way that I would say it really though is that there are many species of animals that influence each, each other's nervous systems, that regulate each other's nervous systems, and they mainly do it by physical means. They do it by chemicals, scent. They do it by, you know, so, so termites and ants and bees, for example, use chemical scents. Mammals like, um, like rodents use scent, and they also use, uh, hearing, audition, and that little bit of vision. Um, primates, you know, non-human primates add vision, right?

Lex Fridman 17:08 Mm-hmm.

Lisa Feldman Barrett 17:08 Um, and, um, I think everybody uses touch. Humans, as far as I know, are the only species that use ideas and words-

Lex Fridman 17:18 Yeah. What are-

Lisa Feldman Barrett 17:19 ... to regulate each other, right? I can text something to someone halfway around the world.

Lex Fridman 17:23 That's fascinating.

Lisa Feldman Barrett 17:24 They don't have to hear my voice.

Lex Fridman 17:26 Yeah.

Lisa Feldman Barrett 17:26 They don't have to see my face, and I can have an effect on their nervous system. And ideas, the ideas that we communicate with words, I mean, words are, in a sense, a way for us to do mental telepathy-

Lex Fridman 17:38 Mm-hmm.

Lisa Feldman Barrett 17:38 ... with each other, right? I mean, I'm not the first person to say that, obviously, but how do I control your heart rate? How do I control your breathing? How do I control your actions with words? It's because those words are communicating ideas.

Lex Fridman 17:54 So you also write, I think ... L- let's go back to the brain. You write that Plato gave us the idea that the human brain has three brains in it, three forces, uh, which is kind of a compelling notion. Uh, you disagree. First of all, what are the three parts of the brain, and, uh, why do you disagree?

Lisa Feldman Barrett 18:16 So, Plato's description of the psyche, which for the moment we'll just assume is the same as a mind, there are some scholars who would say, you know, a soul, a psyche, a mind, those aren't actually all the same thing in ancient Greece, but we'll just for now gloss over that. So, Plato's idea was that, and it was a, it was a description of really about moral behavior and moral responsibility in humans. So, the idea was that, you know, the human psyche can be described with an, um, uh, metaphor of two horses and a charioteer. So, one horse for instincts, um, like, um, feeding, and fighting, and fleeing, and, uh, reproduction. I'm trying to control my salty language.

Lex Fridman 19:07 (laughs) .

Lisa Feldman Barrett 19:08 Um, which apparently they print in England. Like, I actually tossed off a fairly-

Lex Fridman 19:14 What, which? F, S? Which one?

Lisa Feldman Barrett 19:15 Yeah. F, F, um-

Lex Fridman 19:16 Oh, okay.

Lisa Feldman Barrett 19:17 ... yeah.

Lex Fridman 19:17 Yeah. (laughs)

Lisa Feldman Barrett 19:17 I was like, "You printed that?"

Lex Fridman 19:19 (laughs) .

Lisa Feldman Barrett 19:19 I couldn't believe you printed that. (laughs)

Lex Fridman 19:20 Without like the stars or whatever?

Lisa Feldman Barrett 19:22 No, no, no. There was full print.

Lex Fridman 19:23 Okay, then.

Lisa Feldman Barrett 19:23 Yeah, they also printed the- a B word, and it was really quite-

Lex Fridman 19:27 Wow.

Lisa Feldman Barrett 19:28 ... yeah. Anyways.

Lex Fridman 19:28 Well, we should (laughs) , we should, we should, uh, learn something from England.

Lisa Feldman Barrett 19:32 Indeed. Anyways, but instincts, and then the other horse represents emotions, uh, and then the charioteer represents rationality which controls, you know, the two beasts, right? And, um, fast-forward, you know, couple of centuries, and, uh, in the middle of the 20th century, uh, there was a very popular view of brain evolution which suggested that you have this, uh, reptilian core, like a lizard bra- an inner lizard brain, uh, for instincts, and then wrapped around that evolved, on, layered on top of that, evolved a limbic system for, uh, in mammals. And so the- the novelty was in a mammalian brain, which, uh, bestowed mammals with, uh, gave them emotions, the capacity for emotions. And then, um, on top of that, uh, evolved, uh, a cerebral cortex, um, which in, in largely in primates, but, but, um, very large in, in humans. Um, and it's not that I personally disagree, it's that as far back as the 1960s, but really by the 1970s, it was shown pretty clearly w- with evidence from molecular genetics, so peering into cells in the brain to look at the molecular makeup of genes that the brain did not evolve that way. And the irony-

Lex Fridman 21:07 Hm.

Lisa Feldman Barrett 21:07 ... is that, um, you know, the, the idea of the, the three-layered brain with an inner lizard, you know, that hijacks your, uh, hijacks your behavior and causes you to do and say things that, uh, you would otherwise not or maybe that you will regret later, that idea, um, became very popular, was popularized by, uh, Carl Sagan in, uh, The Dragons of Eden, which won a Pulitzer Prize in 1977, when it was already known pretty much in evolutionary neuroscience that the whole, uh, narrative was a myth.

Lex Fridman 21:47 So, well, the narrative is on the- the way it evolved, but do you ... I mean, again, it's that problem of, um, it being a useful tool of conversation to say like there's a lizard brain, and there's a, like if I get overly emotional on Twitter, that was the lizard brain, and so on. Uh, but do you-

Lisa Feldman Barrett 22:10 No, I don't think it's useful.

Lex Fridman 22:11 But-

Lisa Feldman Barrett 22:11 I think it's a, I think that-

Lex Fridman 22:13 But is it a, is, is it, uh, is it useful or is it accurate?

Lisa Feldman Barrett 22:17 I don't think it's accurate, and therefore I don't think it's useful.

Lex Fridman 22:20 Got it.

Lisa Feldman Barrett 22:21 So, I, here's what I would say. You know, I think that, um, the way I think about philosophy and science is that they are useful tools for living, and in order to be useful tools for living, they have to help you make good decisions. The triune brain as it's called, this, this three-layer brain, the idea that your brain is like an already baked cake-

Lex Fridman 22:52 (laughs) .

Lisa Feldman Barrett 22:52 ... and, uh, you know, the cortex, cerebral cortex is just layered on top like icing.

Lex Fridman 22:57 Mm-hmm.

Lisa Feldman Barrett 22:57 Um, the idea, that idea is the foundation of, uh, the law in most Western countries. It's the foundation of, uh, economic theory, and it large- and it's a great narrative. It sort of fits our intuitions about how we work, but it also, um, it's, in addition to being wrong, it lets people off the hook for-

Lex Fridman 23:25 Right.

Lisa Feldman Barrett 23:25 ... uh, for nasty behavior, you know? Um, and it also suggests that emotions can't be a source of wisdom, which they often are. Um, in fact, you, you would not-

Lex Fridman 23:37 Yeah.

Lisa Feldman Barrett 23:37 ... want to be around someone who didn't have emotions. That would be, that's a psychopath.

Lex Fridman 23:42 Right.

Lisa Feldman Barrett 23:43 I mean, that's not someone you, you know, wanna, wanna really, uh, have, have that person deciding your outcome. So, I guess my, and I could sort of go on and on and on, but my point is that, um, I don't think, I don't think it's a useful narrative in the end.

Lex Fridman 24:03 What's the more accurate view of the brain that we should use when we're thinking about it?

Lisa Feldman Barrett 24:08 I'll answer that in a second, but I'll say that even our notion of what an instinct is or what a reflex is is not quite right, right? So, if you look at evidence from, um-... ecology, for example, and you look at animals in their ecological context, what you can see is that even things which are reflexes, eh, are very context sensitive. Um, the, the brains of those animals are executing so-called instinctual actions in a very, very context-sensitive way. And so, you know, even when a physician, you know, takes the... You know, it's like the idea of your patellar, uh, reflex, where they hit, you know, your patellar tendon on your knee and you, you kick. The m- the force with which you kick, and so on, inf- is influenced by all kinds of things. It's, it's... A reflex isn't like a robotic, uh, response. And, um, so I think a better way is a way that... to think about how brains work, is the way that, um, m- matches our best understanding, our best scientific understanding, which I think is really cool, uh, because it's really counterintuitive. So, how I came to this view, and it's... I'm certainly not the only one who holds this view, I was reading work in... on neuroanatomy, and the, the view that I'm about to tell you was sugges- strongly suggested by that. And then I was reading work in signal processing, like by eng- you know, electrical engineering, and th- similarly, it, the work suggested that, that, the research suggested that the brain worked this way. And I'll just say that I was reading across multiple literatures, and they were... who don't speak to each other, and they were all pointing in this direction. And so far, although some of the details are still up for grabs, the general gist, I think, is pr- I've not come across anything yet which really violates... and I'm looking.

Lex Fridman 26:10 Mm-hmm.

Lisa Feldman Barrett 26:11 Um, and so the idea is something like this. It's very counterintuitive. Um, so the way to describe it is to say that your brain doesn't react to things in the world. It's not... Uh, to us, it feels like our eyes and... are, um, are windows on the world. We see things, we hear things, we, we react to them. Um, in psychology, we call this stimulus response. So, um, your face is a s- your voice is a stimulus to me. I receive input and then I react to it. Uh, and I might react very automatically, you know, system one.

Lex Fridman 26:49 (laughs)

Lisa Feldman Barrett 26:49 Uh, and, uh, oh, but I also might ex- execute some control where I maybe stop myself from saying something or doing something, and, um, more, in a more reflective way, execute a different action, right? That's system two. The way the brain works though is it's predicting all the time. It's constantly talking to itself, constantly, uh, talking to your body, uh, a- and it's constantly, um, predicting what's going on in the body and what's going on in the world, and making predictions. And, um, the information from your body and from the world really confirm or correct those predictions.

Lex Fridman 27:32 So, fundamentally, the thing that the brain does most of the time is just predict- like, talking to itself and predicting stuff about the world, not, like, this dumb thing that just senses and responds, senses and responds.

Lisa Feldman Barrett 27:46 Yeah. So the way to, the way to think about it is like this. You know, your brain is, uh, trapped in a dark silent box-

Lex Fridman 27:52 Yeah. That's very romantic of you. (laughs)

Lisa Feldman Barrett 27:56 ... um, which is your skull, and the only information that it receives from your body and from the world, right, is through the senses, through the sense organs, your eyes, your ears. And, um, you have, uh, sense- sensory data that comes from your body, uh, that you're largely unaware of, uh, to your brain, which we call interoceptive, as opposed to exteroceptive, which is the world around you. And... But your brain is receiving sense data continuously, which are the effect of some set of causes. Your brain doesn't know the cause of these sense data. It's only receiving the effects of those causes, which are the data themselves.

Lex Fridman 28:46 Mm-hmm.

Lisa Feldman Barrett 28:46 And so your brain has to solve what philosophers call an inverse inference problem. How do you know... When you only receive the effects of something, how do you know what caused those effects? So, when there's a flash of light, or, um, a change in air pressure, or a tug somewhere in your body, how does your brain know what caused those, um, events so that it knows what to do next to keep you alive and well? And the answer is that your brain has one other source of information available to it, which is your past experience. It can reconstitute, in its wiring, um, past experiences, and it can combine those past experiences in novel ways, and so we have lots of names for this. In psychology, we call it memory, we call it perceptual inference, we call it simulation, um, uh, it's also we call it concepts or conceptual knowledge, we call it prediction. Basically, if we were to stop the world right now, stop time, your brain is in a state, and it's representing what it believes is going on in your body and in the world, and it's predicting what will happen next based on past experience, right, probabilistically, what's most likely to happen. And it begins to, um, prepare your action-... and it begins to prepare your, the, it prepare your experience, based... So it's anticipating the sense data it's going to receive, and then when that, those data come in, they either confirm that prediction and your action executes, because the plan's already been made-

Lex Fridman 30:44 Mm-hmm.

Lisa Feldman Barrett 30:44 ... or, um, it, uh, or there's some, uh, sense data that your brain didn't predict that's unexpected, and your brain takes it in, we say encodes it. We have a fancy name for that. We call it learning. Your brain learns and it updates its, uh, storehouse of knowledge, which we call an internal model, and, uh, that you, so that you can predict better next time. And it turns out that predicting and correcting, predicting and correcting is a much more metabolically efficient way to run a system than constantly reacting all the time, because if you're constantly reacting, it means you have no... You can't anticipate in any way what's going to happen-

Lex Fridman 31:23 Mm-hmm.

Lisa Feldman Barrett 31:24 ... and so the, the amount of uncertainty that you have to deal with is, uh, overwhelming to a nervous system.

Lex Fridman 31:30 (laughs) Metabolically costly. I like it. (laughs)

Lisa Feldman Barrett 31:32 And so what is a reflex? A reflex is when your brain doesn't check against the sense data, that the-

Lex Fridman 31:41 Ah.

Lisa Feldman Barrett 31:41 ... potential cost to you is so great, maybe because you know your life is threatened, that your brain makes the prediction and executes the action without checking.

Lex Fridman 31:55 Yeah, so but prediction's still at the core. That's the beautiful vision

Lex Fridman 31:58 of the brain. I wonder, from almost an AI perspective, but just computationally, is the brain just mostly a prediction machine then? Like is the perception just a nice little feature added on top, like the, both the, the integration of new perceptual information, I wonder how big of, of an impressive system is that relative to just the big predictor, model constructor?

Lisa Feldman Barrett 32:25 Well, I think that we can, we can look to evolution for that, for, uh, one answer, which is that when you go back, you know, 550 million years, give or take, we, you know, the world was populated by creatures, really ruled by creatures without brains. Um, and, um, you know, that's a biological statement, not a political statement.

Lex Fridman 32:48 (laughs)

Lisa Feldman Barrett 32:48 So ruled with creatures-

Lex Fridman 32:49 You're calling dinosaurs dumb? You're talking about like-

Lisa Feldman Barrett 32:51 Oh, no, I'm not talking about dinosaurs, honey.

Lex Fridman 32:53 Yeah.

Lisa Feldman Barrett 32:53 I'm talking way back-

Lex Fridman 32:55 Way up.

Lisa Feldman Barrett 32:55 ... further back than that. Um, really these-

Lex Fridman 32:57 The animals.

Lisa Feldman Barrett 32:57 ... they're these little, little, um, creatures called, uh, amphioxus, which is the modern, it's a, or a lancet. That's the modern animal, but it's an animal that scientists believe is very similar to, um, our common, the common ancestor that we share, uh, with invertebrates, um, because, uh, be- basically because of m- uh, the tracing back the molecular genetics in cells. And that animal had no brain. It had some cells that would later turn into a brain, but in that animal, there's no brain. But that animal also had no head, and it had no eyes, and it had no ears, and it had really, really no senses for the most part. It had very, very limited sense of touch. It had an eye spot for, um, not for seeing, but just for, um, in training to circadian rhythm to light and dark, um, and it had no hearing. It had a vestibular cell so that it could keep upright in the water. At the time, approx- we're talking evolutionary scale here, so, you know, give or take some 100 million years or something, but at the time... You know, what are the vertebrate, like when a, when a backbone evolved and a brain evolved, a full brain, that was when a head evolved with sense, with sense organs and when, um, that's when your viscera, like internal systems involved. So the answer I would say is that, um, that senses... Neuros- motor neuroscientists, people who study the control of motor behavior, believe that, um, senses evolved in the service of motor action.

Lex Fridman 34:42 Mm-hmm.

Lisa Feldman Barrett 34:42 So the idea is that, like what triggered the, what triggered, what was, what was the big evolutionary change? What was the big pressure, uh, that made it useful to have eyes and ears and a visual system and an auditory system and a brain basically, and you know, and the answer that, um, is, you know, commonly entertained right now is that it was predation, that when at some point an a- an animal evolved that deliberately ate another animal-

Lex Fridman 35:16 (laughs)

Lisa Feldman Barrett 35:16 ... and this launched an arms race between predators and prey, and it became very useful to have senses, right? So these, these little amphiox-

Lex Fridman 35:26 Oh, I see.

Lisa Feldman Barrett 35:27 These little amphioxi, i- i- you know, don't really have, they, they don't have an en- they're not aware of their environment very much, really. They, um, uh, and so being able to look up ahead and, you know, ask yourself, you know, is the, you know, "Should I eat that or will it eat me?" um, is, is a very useful thing. So the idea, um, is that sense, sense, sense data is not there for consciousness. It didn't evolve for the purposes of consciousness. It didn't evolve for the purposes of experiencing anything. Um, it evolved, uh, in the ser- to be in the service of motor control. However, maybe it's useful. Um, this is why, you know, scientists sometimes, uh, avoid questions about why things evolved, that this is-

Lex Fridman 36:25 Mm-hmm.

Lisa Feldman Barrett 36:26 ... the philosophers call this teleology. You might be able to say something about how, uh-... things evolve, but not necessarily why. We don't really know the why. Uh, that's all speculation.

Lex Fridman 36:40 But the why is kind of nice here. The, the interesting thing is, uh, that was the first element of social interaction is, "Am I gonna eat you or are you gonna eat me?" (laughs) And for that, it's useful to be able to see each other, sense each other. Uh, that's kind of fascinating that, that there was a time when life didn't eat each other. (laughs)

Lisa Feldman Barrett 37:03 Or they did by accident, right? So an amphioxus, for example, will, um, it kind of like gyrates in the water and then it plants itself in the sand like a blade of gr- like a living blade of grass-

Lex Fridman 37:17 Yeah.

Lisa Feldman Barrett 37:17 ... and then it just filters-

Lex Fridman 37:18 Yeah.

Lisa Feldman Barrett 37:19 ... l- uh, whatever comes into its mouth, right? So it is, it is eating, but it's not actively hunting.

Lex Fridman 37:25 Hunting.

Lisa Feldman Barrett 37:26 And when, um, the concentration of food decreases, it, the amphioxus can sense this and so it basically wriggles itself randomly to some other spot which pro- probabilistically will have more food than wherever it is. So it's not really, you know, it's not guiding its actions, um, on the basis of... It's not... We, we would say there's no real intentional action-

Lex Fridman 37:55 Mm-hmm.

Lisa Feldman Barrett 37:56 ... um, in that, in that, in the traditional sense.

Lex Fridman 37:58 Speaking of intentional action, and if the brain is pre- if prediction is indeed a core component of the brain, let me ask you a question that scientists also hate is, uh, about free will.

Lisa Feldman Barrett 38:11 (laughs)

Lex Fridman 38:11 So, uh, how does, uh... Do you think about free will much? H- how does that fit into this, into your view of the brain? Why does it feel like we make decisions in this world?

Lisa Feldman Barrett 38:25 This is a hard que- we scientists hate this 'cause it's a hard, it's a hard question.

Lex Fridman 38:28 Yeah.

Lisa Feldman Barrett 38:28 We don't know the answer to it.

Lex Fridman 38:29 Have you taken a side? (laughs)

Lisa Feldman Barrett 38:30 I think I have.

Lex Fridman 38:31 Do we have free will?

Lisa Feldman Barrett 38:31 I think I have taken a side, but it, it... I don't put a lot of stock in my own intuitions or anybody's intuitions about the cause of things, right? Our ex- one thing we know about the brain for sure is that the brain creates experiences for us. Our, my brain creates experiences for me, your brain creates experiences for you in a way that lures you to believe that those experiences actually reveals the way that it works.

Lex Fridman 38:56 Right.

Lisa Feldman Barrett 38:56 But it doesn't. So the...

Lex Fridman 38:59 So you don't trust your own intuition about- (laughs)

Lisa Feldman Barrett 39:01 Not really.

Lex Fridman 39:02 ... about free will? (laughs)

Lisa Feldman Barrett 39:02 Not really. No, I mean, no. But, but I am also somewhat persuaded by, you know, I think Dan Dennett wrote at, at one point like, um, uh, you know, the philosopher Dan Dennett wrote at one point that, um, it, it's... I c- I can't say it as eloquently as him, but it... People obviously have free will. They are obviously making choices. So it's, you know... And so there is this observation that we're not robots and we can do some things like a little more sophisticated than an amphioxus. So, um, so here's what I would say. I would say that your predictions, your internal model that's running right now, right? That your ability to understand the sounds that I'm making and attach them to ideas is based on the fact that you have years of experience knowing what these sounds mean in a particular statistical, uh, pattern, right?

Lex Fridman 40:00 Mm-hmm.

Lisa Feldman Barrett 40:01 I mean, that's how you can understand the words that are coming out of my...

Lex Fridman 40:05 Mouth?

Lisa Feldman Barrett 40:06 Right. I think we did this once before too, didn't we? When we were...

Lex Fridman 40:09 I, I don't know. I would have to access my memory module.

Lisa Feldman Barrett 40:12 I think when I was in your, when I...

Lex Fridman 40:14 The class thing?

Lisa Feldman Barrett 40:14 Yeah, I think we did it just like that, actually. So bravo.

Lex Fridman 40:18 Wow.

Lisa Feldman Barrett 40:19 Yeah.

Lex Fridman 40:19 I have to go look, look back to the tape.

Lisa Feldman Barrett 40:21 Yeah. Anyways-

Lex Fridman 40:22 (laughs)

Lisa Feldman Barrett 40:23 ... the, um, the idea though is that your brain is using past experience and it can, and it can use past experience in, um... So it's remembering, but you're not consciously remembering. It's basically re-implementing prior experiences as a way of predicting what's gonna happen next.

Lex Fridman 40:41 Yeah.

Lisa Feldman Barrett 40:42 And it can do something called conceptual combination, which is it can take bits and pieces of the past and combine it in new ways. So you can experience and make sense of things that you've never encountered before because you've encountered something similar to them.

Lex Fridman 41:00 Mm-hmm.

Lisa Feldman Barrett 41:03 Um, and so a brain, in a sense, is not, um, just, um, doesn't just contain information. It is information-gaining, meaning it can create i- new information by this generative process. So in a sense you could say, well, that maybe that's, that's a source of free will. But I think really where free will comes from, or the kind of free will that I think is worth having a conversation about, is, um, involves cultivating experiences for yourself that change your internal model. When you were born and you were raised in a particular context that your mode- your brain wired itself to your surroundings, to your physical surroundings, and also to your social surroundings. So you were handed an internal model basically. Um, but, uh, when you grow up, the more control you have over your where you are and what you do, um, you can cultivate new experiences for yourself. And those new experiences, uh, can change your internal model, and you can actually, um, practice those experiences in a way that makes them automatic. Make... Meaning it makes it easier for the brain, your brain to make them again. And I think that that is something like...... what you would call free will. You aren't responsible for the model that you were handed, that someone, you know, your- your caregivers, uh, cultivated a model in your brain. You're not responsible for that model, but you are responsible for the one you have now. You can choose, you choose what you expose yourself to. You choose, uh, how you spend your time. Not everybody has choice over everything.

Lex Fridman 43:02 Right.

Lisa Feldman Barrett 43:02 But everybody has a little bit of choice, um, and- and so I think that is, uh, something that I- I think is arguably called free will.

Lex Fridman 43:13 Yeah, there's- there's like the- the ripple effects of the billions of decisions you make early on in life ha- have, are so great that, uh, even if it's not, even if it's like all deterministic, just the amount of possibilities that are created and then the focusing of those possibilities into a single trajectory, uh, that s- somewhere within that, that's free will, even if it's all deterministic. That might as well be of just the number of choices that are possible and the fact that you just make one trajectory through those set of choices seems to be like something like they'll be called free will. But it's still kind of sad to think, like there doesn't seem to be a place where there's magic in there, where it is all just the computer.

Lisa Feldman Barrett 44:08 Well, there's lots of magic, I would say so far, because we don't really understand, uh, how all of this is exactly played out at a- uh, I mean, scientists are working hard and disagree about some of the details under the hood of what I just described, but I think there's quite a bit of magic actually. And also there's- there's also, um, stochastic firing of neur- neurons don't, they- they're not purely digital in the sense that there is... There's also analog communication between neurons, not just digital. So it's not just with- not just with firing of- of axons, and some of that, uh, they- there's- there are other ways to communicate. And also, um, uh, there's noise in the system, and the noise is there for a really good reason. And that is, um, the more variability there is, the more potential there is for your brain to be able to be information-bearing. So, um, basically, you know, there are some animals that have clusters of cells, their only job is to inject noise.

Lex Fridman 45:24 (laughs)

Lisa Feldman Barrett 45:25 Uh, you know, into their, um, neural patterns.

Lex Fridman 45:28 So maybe noise is the source of free will. (laughs)

Lisa Feldman Barrett 45:30 So you can think about-

Lex Fridman 45:31 Yeah.

Lisa Feldman Barrett 45:31 You can think about stochasticity or noise as- as a source of fre- free will, or you can think of- of, um, conceptual combination as a- a source of free will. You can certainly think about, um, cultivating, uh, you know, you can't reach back into your past and change your past. You know, people try by psychotherapy and so on. But what you can do is change your present, um, which becomes your past, um-

Lex Fridman 46:02 Well, let me-

Lisa Feldman Barrett 46:03 Right. So-

Lex Fridman 46:03 Let me think about that sentence. (laughs) Uh-

Lisa Feldman Barrett 46:05 So one way to think about it is that you're continuously... This is a- a colleague of mine, a friend of mine said, "So what you're saying is that people are continually cultivating their past?"

Lex Fridman 46:15 (laughs)

Lisa Feldman Barrett 46:15 And I was like, "That's very poetic." Yes.

Lex Fridman 46:17 (laughs)

Lisa Feldman Barrett 46:18 You- you are continually cultivating your past as a means of controlling your future.

Lex Fridman 46:26 So you think, uh, yeah, I guess the- the construction of the mental model that you use for prediction ultimately contains within it your perception of the past, like the way you interpret the past or even just the entirety of your narrative about the past. So you're constantly rewriting the story of your past. Oh boy. Yeah. That's, one, poetic and also just awe-inspiring.

Lex Fridman 46:51 What about the other thing you talk about? You've mentioned about sensory perception as a thing that like is just you have to infer about the sources of the thing that you have perceived through your senses. S- so, uh, let me ask the- another ridiculous question. Is- is anything real at all? Like how do we know it's real? How do we make sense of the fact that just like you said, there's this brain sitting alone in the darkness trying to perceive the world? How do we know that the world is out there-

Lisa Feldman Barrett 47:26 I-

Lex Fridman 47:26 ... to be perceived?

Lisa Feldman Barrett 47:27 Yeah. So I don't think that you should be asking questions like that-

Lex Fridman 47:30 Yeah.

Lisa Feldman Barrett 47:31 ... without passing a joint.

Lex Fridman 47:32 Right. No, for sure.

Lisa Feldman Barrett 47:33 Yeah.

Lex Fridman 47:33 I actually did-

Lisa Feldman Barrett 47:34 But- yeah.

Lex Fridman 47:34 ... before th- this, so I apologize. Yeah.

Lisa Feldman Barrett 47:36 Okay. No, wait, well, that's okay.

Lex Fridman 47:37 (laughs)

Lisa Feldman Barrett 47:37 For- you apologized for not sharing.

Lex Fridman 47:38 Yeah.

Lisa Feldman Barrett 47:38 That's okay.

Lex Fridman 47:39 (laughs)

Lisa Feldman Barrett 47:39 So, I mean, here's what I would say. What I would say is that the reason why we can be pretty sure that there's a there there, is that the- the structure of the information in the world, what we call statistical regularities in sights and sounds and so on, and the structure of the information that comes from your body, it's not random stuff. There's a structure to it. There's a- a spatial structure and a temporal structure, and that spatial and temporal structure wires your brain. So an infant brain is not a miniature adult brain. It's a brain that is waiting for wiring instructions from the world, and it must receive those wiring instructions to develop in a typical way. So for example, when a newborn is born, when a newborn is born? When a- when a baby is born, um, the baby can't see very well because the visual system in that baby's brain is not complete.... the, the retina of your eye, which actually is part of your brain, has to be stimulated with photons of light. If it's not, the baby won't develop normally, the- to be able to see in a neurotypical way. Same thing is true for hearing. The same thing is true really for all your senses. So, the point is that, um, that the physical world, the sense data from the physical world wires your brain so that you have an internal model of that world so that your brain can predict well to keep you alive and well, and allow you to thrive.

Lex Fridman 49:19 That's fascinating that the brain is waiting for very specific kind of a set of instructions from the world. Like, not- not the specific, but a very specific kind of instructions.

Lisa Feldman Barrett 49:31 Yeah. So, you- scientists call it expectable input.

Lex Fridman 49:35 Right.

Lisa Feldman Barrett 49:35 The brain needs some input in order to develop normally.

Lex Fridman 49:39 Yeah.

Lisa Feldman Barrett 49:39 And s- we're- and we are genetically, you know, we, as I say in the book, we- we have the kind of nature that requires nurture.

Lex Fridman 49:48 Hmm.

Lisa Feldman Barrett 49:48 We- we can't develop normally without sense input, sensory input from the world and from the body, and what's really interesting about humans and some other animals too, but really seriously in humans, is the input that we need is not just physical. It also social. We, in order for an- an- an infant, a human infant to develop normally, that infant needs eye contact, touch. It needs certain types of smells. It needs to be cuddled. It needs... Right? So, um, without social input, the brain, it's- that- that infant's brain will not wire itself in a neurotypical way. And again, I would say the- there are lots of, um, cultural patterns of caring for an infant. It's not like the infant has to be cared for in one way.

Lex Fridman 50:49 Mm-hmm.

Lisa Feldman Barrett 50:49 Um, whatever the social environment is for an infant, um, that it, will- will be reflected in that infant's, um, internal model. So we have lots of different cultures, lots of different ways of rearing children, um, and that's an advantage for our species, although we don't always experience it that way. That is an advantage for our species. Um, but if you, if you just, you know, feed and water a baby without all the extra social doodads, um, what you get is a profoundly impaired, uh, human.

Lex Fridman 51:25 Yeah, but nevertheless, you're kind of saying that the physical reality has a, has a consistent thing throughout that keeps feeding these, uh, set of sensory information that our brains are constructed for, but...

Lisa Feldman Barrett 51:43 Yeah, the cool thing though is that if you change the consistency, if you change the statistical regularities, so prediction error, your brain can learn it. It's expensive for your brain to learn it, and it takes a while to- to- for the brain to get really automated with it, but, you know, you, um, had a wonderful, uh, conversation with David Eagleman who just published a book-

Lex Fridman 52:02 Yeah.

Lisa Feldman Barrett 52:02 ... about this-

Lex Fridman 52:03 Yeah.

Lisa Feldman Barrett 52:04 ... and gave lots and lots of really very, very cool examples, some of which are- I actually discussed in How Emotions Are Made, but not obviously to the extent that he did, um, in his book, which is- it's a fascinating book and it's- but it- it speaks to the point that your internal model is always under construction, and therefore you always can modify your experience.

Lex Fridman 52:30 I wonder what the limits are, like, uh, if we can- if we put it on Mars or if we put it in virtual reality or if we sit at home during a pandemic and we spend most of our day on Twitter and TikTok. Like, I wonder what- where the breaking point, like, the limitations of the brain's capacity to, uh, to- to properly continue wiring itself.

Lisa Feldman Barrett 52:54 Well, I think what I would say is that there are different ways to specify your question, right? Like, one way to specify it would be the way that David, um, phrases it, which is, can we, can we create a new sense?

Lex Fridman 53:09 Right.

Lisa Feldman Barrett 53:09 Like can- can we create a- a new sensory modality? Um, how hard would that be? What are the limits in doing that? Um, and, um, but another way to, um, say it is, what- what happens to a brain when you remove some of those statistical regularities, right? Like, what happens to a brain... What happens to an adult brain when you remove some of the statistical patterns that were there and they're not there anymore?

Lex Fridman 53:37 Are- are you talking about in the environment or in the actual, like, you remove eyesight, for example? Or do- do-

Lisa Feldman Barrett 53:43 Well, either way. I mean, basically one way to limit the inputs to your brain are to stay home and, um-

Lex Fridman 53:51 Right.

Lisa Feldman Barrett 53:51 ... protect yourself.

Lex Fridman 53:52 I see, yeah.

Lisa Feldman Barrett 53:53 Another way is to put someone in, um, solitary confinement. Another way is to stick them, uh, in a nursing home. Another... Well, not all nursing homes, but you know-

Lex Fridman 54:04 Right.

Lisa Feldman Barrett 54:05 ... but there are some, right? Which really are s- where peop- people are somewhat impoverished in the interactions, the- and the sensory st- the variety of sensory stimulation that they get. Another way is that you lose a sense, right? But the point is, I think that, you know, the human brain really likes variety, to- to say it in a, you know-

Lex Fridman 54:31 (laughs)

Lisa Feldman Barrett 54:31 ... a- like a- a pro- you know, sort of Cartesian way. You know, e- um, variety is a good thing, um, for a brain and, um, uh...... there are risks that you take, uh, when you restrict, uh, what you expose yourself to.

Lex Fridman 54:54 Yeah. You know, there's all this talk of diversity, the brain loves it to the fullest d- definition and degree of diversity.

Lisa Feldman Barrett 55:01 Yeah. I mean-

Lex Fridman 55:02 That's-

Lisa Feldman Barrett 55:02 ... I would say the only thing... Basically, human brains thrive on diversity. The only place where we seem to have difficulty with diversity is with each other.

Lex Fridman 55:12 (laughs) Yeah.

Lisa Feldman Barrett 55:13 Right? But we... Who wants to eat the same food every day?

Lex Fridman 55:16 Yeah.

Lisa Feldman Barrett 55:16 You never would. Who wants to wear the same clothes every day?

Lex Fridman 55:19 Yeah.

Lisa Feldman Barrett 55:19 I mean, my husband, if you ask him to close his eyes, he won't be able to tell you what he's wearing.

Lex Fridman 55:23 Yeah.

Lisa Feldman Barrett 55:23 He just... Right? He'll buy seven shirts of exactly the same style-

Lex Fridman 55:27 Yeah.

Lisa Feldman Barrett 55:27 ... in different colors, but they are in different colors, right? It's not like he's wearing (laughs) -

Lex Fridman 55:31 Well, how would you then explain my brain, which is terrified of choice and therefore, I wear the same thing every time?

Lisa Feldman Barrett 55:39 Well, you must be getting your diversity-

Lex Fridman 55:41 That's right.

Lisa Feldman Barrett 55:41 Well, first of all, you are a fairly sharp dresser, so there is that.

Lex Fridman 55:44 (laughs)

Lisa Feldman Barrett 55:44 But, um, so you're getting some reinforcement for dressing-

Lex Fridman 55:47 Yes.

Lisa Feldman Barrett 55:47 ... the way that you do. But, no, your brain must get diversity in-

Lex Fridman 55:51 In other places.

Lisa Feldman Barrett 55:52 ... in other, in other places, but I think we... You know, the, the... So, there... The two most expensive things your brain can do, metabolically speaking, is, um, is move your body, um, and, uh, learn something new. So novelty, that is diversity, right? Co- comes at a cost, a metabolic cost, but it's a cost, i- it's an investment that, that gives returns. And in general, people vary in how much they like novelty, unexpected things. S- Some people really like it, some people really don't like it, and there's everybody in between. But in general, um, we don't eat the same thing every day, we don't usually do exactly the same thing in exactly the same order, in exactly the same place every day. Um, the only place we have difficulty, uh, with diversity i- in, is in each other.

Lex Fridman 56:47 (laughs)

Lisa Feldman Barrett 56:48 And then we, we have considerable problems there, I would say, as a species.

Lex Fridman 56:53 Let me ask, uh, I don't know if you're familiar with Donald Hoffman's work about the s- like questions of reality. Um, what, what are your thoughts of the possibility that the very thing we've been talking about, of the brain wiring itself from birth to a particular set of inputs, is just a little slice of reality? That there is something much bigger out there that we humans, with our cognition, cognitive capabilities, are just not even perceiving. That, uh, the thing we're perceiving is just a crappy, like, Windows 95 interface onto a much bigger, richer set of complex physics that we're not even in touch with.

Lisa Feldman Barrett 57:38 Well, without getting too metaphysical about it, I think we know for sure, it doesn't have to be the, you know, crappy version of anything, but we definitely have a limited... We have, we have set of senses that are limited in very physical ways, and we're clearly not perceiving everything there is to perceive. That's clear. I mean, it's just... It's not that hard. We can't... Without special... Why do we invent scientific tools? It's so that we can overcome our senses and, and experience things that we couldn't otherwise.

Lex Fridman 58:10 Right.

Lisa Feldman Barrett 58:10 Whether they are, you know, different parts of the, uh, visual spectr- you know, the light spectrum or, um, things that are too s- microscopically small for us to see or too far away for us to see. Um, so clearly, we're only getting a slice, um, and that slice, you know, the, the interesting or potentially sad thing about humans is that we, whatever we experience, we think there's a natural reason for experiencing it.

Lex Fridman 58:43 Mm-hmm.

Lisa Feldman Barrett 58:44 And we think it's obvious and natural and it must be this way and that all the other stuff isn't important, and that's clearly not true. Many of the things that we think of as natural are anything but. We've cr- they're certainly real, but we've created them. Um, they certainly have very real impacts, but we've created those impacts. And we also know that there are many things outside of our awareness that have a g- have tremendous influence on what we experience and what we do. So, there's no question that that's true. I mean, just it's, it's, um-

Lex Fridman 59:19 But-

Lisa Feldman Barrett 59:19 ... but the extent is how fanta-

Lex Fridman 59:21 Yeah, exactly.

Lisa Feldman Barrett 59:21 You're, you know... Really, the question is, how fantastical

Lisa Feldman Barrett 59:23 is it?

Lex Fridman 59:23 Yeah, like what... You know, a lot of people ask me, I've... Uh, am I allowed to say this? I think I'm allowed to say this. Uh, I've eaten shrooms a couple of times, but I haven't gone the full... I'm talking to a few researchers in psychedelics. It's an interesting scientifically place. Like, what is the portal you're entering when you take psychedelics? Or another way to ask is, like, dreams? What are-

Lisa Feldman Barrett 59:45 Yeah. So let me tell you what I think, which is based on nothing.

Lex Fridman 59:48 (laughs)

Lisa Feldman Barrett 59:48 Like, this is based on my, like... Right? So I don't-

Lex Fridman 59:51 Your intuition. What's your-

Lisa Feldman Barrett 59:52 It's based on my, it's based on m- my... I'm guessing now, um, based on what I do know, I would say, but I think that... Well, think about what happens... So you're running... Your brain's running this internal model.

Lex Fridman 1:00:04 Right.

Lisa Feldman Barrett 1:00:04 And it's all outside of your awareness for the... You see the, you feel the products, but you don't, you don't sense the... You have no awareness of the mechanics of it, right?

Lex Fridman 1:00:13 Right.

Lisa Feldman Barrett 1:00:13 It's going on all the time. Um, and so one thing that's going on all the time that you're completely unaware of, is that, um, when your brain... Your brain is basically asking itself, figuratively speaking, not literally, right? Like, "How i- the sen- give- the last time I was in this sensory array with this stuff going on in my body and I... And that, this chain of events which just occurred, what did I do next? What did I feel next? What did I see next?" And so... It doesn't come up with one answer, it comes up with a distribution of it, possible answers, and then there has to be some selection process.And so you have a, a network in your brain, a s- sub-network in your brain, a population of neurons that helps to choose. It's not... I'm not talking about a homunculus in your brain or anything silly like that. Um, uh, this is not the soul, it's not the center of yourself or anything like that, but there is, um, i- um, uh, a set of neurons that weighs the probabilities, uh, um, um, and, and helps to select, uh, n- or narrow the field. Okay? And that, that network is working all the time. It's actually called the control network, the executive control network, or you can call it a frontoparietal, because the regions of the brain that make it up are in the frontal lobe and the parietal lobe. There are also parts that belong to the sub-cortical parts of your brain. It doesn't really matter. The point is th- that there is this network and it is working all the time. Whether or not you feel in control, whether or not you feel like you're expending effort doesn't really matter. It's on all the time, except when you sleep. When you sleep, it's, it's a little bit relaxed.

Lex Fridman 1:02:02 (laughs)

Lisa Feldman Barrett 1:02:03 And so think about what's happening when you sleep. When you sleep, the ext- the external world recedes, the sense data from... So basically your model becomes a little bit the tethers from the world are loosened. And this network which is involved in, you know, maybe weeding out unrealistic things is a little bit quiet. So use... Your dreams are really your internal model that's unconstrained by the immediate world except... So you can do things that you can't do in real life in your dreams, right? You can fly. Like I, for example, when I fly on my back in a dream, I'm much faster than when I fly on my front. Don't ask me why. I don't know, but...

Lex Fridman 1:02:53 When you're laying on your back in your dream-

Lisa Feldman Barrett 1:02:55 No. When I'm in my dream and flying in a dream, I am much faster flyer in the air.

Lex Fridman 1:03:00 You fly often?

Lisa Feldman Barrett 1:03:02 Hmm, not often but I...

Lex Fridman 1:03:03 You, you talk about it like you f- I don't think I've flown for many years.

Lisa Feldman Barrett 1:03:06 Well, you must try it.

Lex Fridman 1:03:08 I've, I've f- I've, uh, flown, I've fallen.

Lisa Feldman Barrett 1:03:11 That's scary.

Lex Fridman 1:03:12 Yeah. But you've flown? You're talking about like-

Lisa Feldman Barrett 1:03:14 Yeah, I fly.

Lex Fridman 1:03:14 ... airplane?

Lisa Feldman Barrett 1:03:15 I fly in my dreams.

Lex Fridman 1:03:16 Oh, that's impressive.

Lisa Feldman Barrett 1:03:16 And I'm way faster, right?

Lex Fridman 1:03:18 On your back?

Lisa Feldman Barrett 1:03:18 On my back. Way faster. Um, now you can say, "Well, you know, you never flew in your life." Right. It's conceptual combination.

Lex Fridman 1:03:25 Right.

Lisa Feldman Barrett 1:03:25 I mean, I've flown in an airplane and I've seen birds fly and I've watched movies of people flying and I know Superman probably flies... I don't know if he flies faster on his back, but-

Lex Fridman 1:03:36 He's w- he's... I don't know, I've never seen Super-

Lisa Feldman Barrett 1:03:37 He's always flying on his front, right?

Lex Fridman 1:03:39 ... Front, so...

Lisa Feldman Barrett 1:03:39 But, yeah. But anyways, my point is that, you know, all of this stuff really, um, all these experiences really become part of your internal model. The thing is that when you're asleep, your internal model is still being constrained by your body. Your, your brain is always attached to your body. It's always receiving sense data from your body. You're mostly never aware of it, uh, unless you run up the stairs, uh, or, or, you know, uh, m- maybe you, um, are ill in some way. But you're mostly not aware of it, which is a really good thing, because if you were, y- you know, you'd never pay attention to anything outside your own skin ever again. Like right now, you seem like you're sitting there very calmly, but you have a, a virtual-

Lex Fridman 1:04:23 Whole thing going on.

Lisa Feldman Barrett 1:04:24 ... drama, right?

Lex Fridman 1:04:25 Yeah.

Lisa Feldman Barrett 1:04:25 It's like a-

Lex Fridman 1:04:25 Yeah.

Lisa Feldman Barrett 1:04:26 ... like a, like an opera going on-

Lex Fridman 1:04:28 Yeah.

Lisa Feldman Barrett 1:04:28 ... inside your body. And so I think that one of the things that happens when people, um, take psilocybin or take, uh, you know, ketamine, for example, is that the tethers-

Lex Fridman 1:04:44 (laughs) Are completely removed.

Lisa Feldman Barrett 1:04:46 ... are completely removed.

Lex Fridman 1:04:48 Yeah.

Lisa Feldman Barrett 1:04:48 Yeah.

Lex Fridman 1:04:48 That's fascinating. And then... But-

Lisa Feldman Barrett 1:04:50 And, and that's why it's helpful to have a guide, right? Because the guide is giving you sense data to steer that internal model so that it doesn't go completely off the rails.

Lex Fridman 1:05:02 Yeah, I know, there's so... again, that wiring to the other brain that's the guide is at least a tiny little tether.

Lisa Feldman Barrett 1:05:09 Exactly.

Lex Fridman 1:05:10 Yeah.

Lex Fridman 1:05:11 Let's talk about emotion a little bit if we could. Emotion comes up often and I have never spoken with anybody who, um, who has a clarity about emotion from a biological and neuroscience perspective that you do. And I'm not sure I fully know how to... as a, as a... I mention this way too much, but as somebody who was born in the Soviet Union and romanticizes basically everything, talks about love non-stop, you know, emotion is a... I don't know what to make of it.

Lisa Feldman Barrett 1:05:49 (laughs)

Lex Fridman 1:05:49 Uh, I don't know what... I, I... So maybe, uh, let's just try to talk about it. I mean, from a neuroscience perspective, we talked about a little bit last time, your book covers it, how emotions are made, but what are some misconceptions we writers of poetry, we romanticizing humans have about emotion that we should, uh, move away from if we're to think about emotion from both a scientific and an engineering perspective?

Lisa Feldman Barrett 1:06:20 Yeah. So there is a common view of emotion in the West. The caricature of that view is that, um, you know, we have an inner beast, right? Your limbic system, your, your inner lizard. Um, we have an inner beast and that comes baked into the brain at birth. So you've got circuits for anger, sadness, fear. It's interesting that they all have English names, these circuits are-

Lex Fridman 1:06:47 (laughs) Yeah.

Lisa Feldman Barrett 1:06:48 But, um, the, the... And they're there and they're triggered by things in the world and, um, then they cause you to do and say and...... thing, you know. So when your fear circuit is triggered, you widen your eyes, you, um, gasp, your, uh, you know, heart rate goes up, you prepare to flee or, um, uh, to freeze. Um, and these are, these are modal responses. They're not the only responses that you give, but on average, they're the prototypical responses. That's the view, and, um, that's the view of emotion in the law. That's the view, um, you know, that emotions are these profoundly unhelpful things that are obligatory, kind of like reflexes. Um, the problem with that view is that it doesn't comport to the evidence, um, and it doesn't really matter. The evidence actually lines up beautifully with each other. It just doesn't line up with that view, and it doesn't matter whether you're measuring people's faces, facial movements, or you're measuring their body movements, or you're measuring their peripheral physiology, or you're measuring their brains, or their voices, or whatever. Pick any s- any, um, output that you want to measure and, you know, any system you want to measure, and you don't really find strong evidence for this. And I say this as somebody who, who not only has reviewed really thousands of articles and run, you know, big meta analyses, which are statistical summaries of, of published papers, but also as someone who has sent teams of researchers to, um, small-scale, uh, cultures, uh, you know, remote cultures, which are very different from, uh, urban, uh, large-scale cultures like ours. Um, and one culture that we visited, and I say "we" euphemistically because I, I myself didn't go because I only had two research permits, and I gave them to my students because I felt like it was better for them to have that experience and more formative for them to have that experience, but I was in contact with them every day by satellite phone, and, um, this was, um, to visit the, um, Hadza hunter-gatherers in Tanzania, who are not, um, an ancient people. They're a modern culture, but they live in circumstances, um, hunting and foraging, circumstances that, um, are very similar, in similar conditions to our ancestors', uh, uh, hunting-gathering ancestors, uh, w- when expressions of emotion were supposed to have evolved-

Lex Fridman 1:09:34 Mm-hmm.

Lisa Feldman Barrett 1:09:34 ... at least by one view of... Okay. So, uh, you know, for many years, I was sort of struggling with, um, this set of observations, right, which is that I feel emotion and I see, I perceive emotion in other people, but scientists can't find a single marker, a single biomarker, not a single individual measure or pattern of measures, that will- can predict how someone- what kind of emotional state they're in.

Lex Fridman 1:10:06 Mm-hmm.

Lisa Feldman Barrett 1:10:07 How could that possibly be? How g- how can you possibly make sense of those two things? And through a lot of reading and a lot of- and immersing myself in different literatures, I came to the hypothesis that the brain is constructing these instances out of more basic ingredients. So when I tell you that the brain- when I suggest to you that what your brain is doing is making a prediction, and it's asking itself, figuratively speaking, "The last time I was in this situation and this, you know, physical state, what did I do next? What did I see next? What did I hear next?" It's basically asking, "What in my past is similar to the present?"

Lex Fridman 1:10:57 Mm-hmm.

Lisa Feldman Barrett 1:10:59 Things which are similar to one another are called a category. A group of things which are similar to one another is a category, and a mental representation of a category is a concept. So your brain is constructing categories or concepts on the fly continuously. So if you really want to understand what a brain is doing, you don't- using machine learning, like classification models, is not going to help you because the brain doesn't classify. It's doing category construction.

Lex Fridman 1:11:29 Mm-hmm.

Lisa Feldman Barrett 1:11:31 And the categories change, or, or you could say it's doing concept construction. It's using past experience to conjure a concept, which is a prediction.

Lex Fridman 1:11:41 Mm-hmm.

Lisa Feldman Barrett 1:11:43 And if it's using past experiences of emotion, then it's constructing an emotion concept. Your concept will be the content of it, is, is, um, changes depending on the situation that you're in. So for example, if your brain uses past experiences of anger that you have learned, either because somebody t- labeled them for you, taught them to you, you observed them, uh, in movies, and so on, in one situation, could be very different from your concept of- for anger than another situation. And this is how anger, instances of anger, are what we call a population of variable instances. Sometimes when you're angry, you scowl. Sometimes when you're angry, you might smile.

Lex Fridman 1:12:38 Mm-hmm.

Lisa Feldman Barrett 1:12:39 Sometimes when you're angry, you might cry. Sometimes your heart rate will go up, it will go down, it will stay the same. It depends on what action you're about to take because the way predict- And I should say, the idea that physiology is yoked to action is a very old idea in, in, uh, the study of the peripheral nervous system. That's been known for really decades. And so...If you look at what the brain is doing, if you just look at the anatomy and- and you- what- here's the hypothesis that you would- that you would come up with. And I can go into the details. (smacks lips) I've published these details in- in scientific papers, and they also appear somewhat in, um, How Emotions Are Made, my first book. They are not in the, you know, Seven and a Half Lessons, because that book is- is really not pitched at that level of explanation.

Lex Fridman 1:13:32 Right.

Lisa Feldman Barrett 1:13:33 It's just giving... It's really just a set of little essays. Um, but the evidence... But what I'm about to say is actually based on- on- on scientific evidence. When your brain begins to pre- make- form a prediction, the first thing it's doing is it's making a prediction of how to change the internal systems of your body, your heart, your cardiovascular system, the control of your heart, control of your lungs, right? A f- a- a flush of- of cortisol, which is not a stress hormone, it's a hormone that gets glucose into your bloodstream very fast, because your brain is predicting you need to do something metabolically, um, uh, expensive. And so- so either that means either move or learn.

Lex Fridman 1:14:23 Mm-hmm.

Lisa Feldman Barrett 1:14:23 Okay? And so your brain is preparing your body, the internal systems of your body, to execute some actions, to move in some way. And the... And then it infers, based on those motor predictions and what we call viscera motor predictions, meaning the- the m- the changes in the viscera that your brain is preparing to, um, to execute. Um, i- your brain makes an inference about what you will sense-

Lex Fridman 1:14:59 Mm-hmm.

Lisa Feldman Barrett 1:14:59 ... based on those motor movements. So, your experience of the world and your experience of your own body are a consequence of those predictions, those concepts. When your brain makes a concept for emotion, it's constructing an instance of that emotion.

Lex Fridman 1:15:21 And-

Lisa Feldman Barrett 1:15:21 That is how emotions are made.

Lex Fridman 1:15:24 And those concepts load in... The predictions that are made include, um, contents inside the body, contents outside the body. I mean, it includes other humans, so just this construction of a concept includes s- th- variables that are much richer than just some sort of, um, simple notion.

Lisa Feldman Barrett 1:15:51 Yeah, so our colloquial notion of a concept where, um, i- you know, um, w- where I say, "Well, what's a concept of a bird?" And then you list a set of features off to me, that's- that's people's understanding, you know, typically of what a concept is. But if you go, uh, into the literature in, um, cognitive science, what you'll see is that the way that scientists have understood what a concept is has really changed over the years. So, people used to think about a concept as, um... Philosophers and scientists used to think about a concept as a dictionary definition for a category. So, there's a set of things which are similar out in the world, and, um, uh, your concept for- for that category is a dictionary des- definition of the features-

Lex Fridman 1:16:39 Right.

Lisa Feldman Barrett 1:16:39 ... the necessary and sufficient features-

Lex Fridman 1:16:41 Right.

Lisa Feldman Barrett 1:16:41 ... of that- of those instances. So, for a bird, um, you know, it would be...

Lex Fridman 1:16:47 Wings, feathers.

Lisa Feldman Barrett 1:16:49 Right. A beak.

Lex Fridman 1:16:50 Yeah.

Lisa Feldman Barrett 1:16:50 It flies, whatever. Okay. Um, that's called the classical category. And scientists discovered, observed that actually n- not all instances of birds have feathers and not all instances of birds fly. And so the idea was that you don't have a single representation of necessary and sufficient features stored in your brain somewhere. Instead what you have is a prototype, a prototype meaning, um, you still have a single representation for the category, one, um, but the features are like of the most typical instance of the category or maybe the most frequent instance, but not all instances of the category have all the features, right?

Lex Fridman 1:17:32 Right.

Lisa Feldman Barrett 1:17:32 They- they have some graded similarity to the prototype. And then, uh, you know, what, um... I'm gonna like incredibly simplify now a lot of work to say that then a series of experiments were done to show that, in fact, what your brain seems to be doing is coming up with a single exemplar or instance of the category and reading off the, um, features when I ask you for the concept. So, if we were in a pet store and I asked you, "What are the features of a bird? Tell me the concept of bird," you would be more likely to give me features of a good pet.

Lex Fridman 1:18:26 Mm-hmm.

Lisa Feldman Barrett 1:18:26 And if we were in a restaurant, you would be more likely, you know, like a budgie, right, or- or a canary. If we were in a restaurant, you would be more likely to give me the features of a bird that you would eat, like a chicken.

Lex Fridman 1:18:38 Yeah.

Lisa Feldman Barrett 1:18:39 And if we were in a park, you'd be more likely, um, to give me, uh, um... In this country, uh, you know, the features of a sparrow or a robin. Whereas if we were in South America, you would probably give me the features of a peacock because that's more common, um, or it's- or is more common there than here that you would see a peacock in such circumstances. So, the idea was that really what your brain was doing was conjuring-... a concept on the fly that meets the function that the category is being put to.

Lex Fridman 1:19:18 (laughs)

Lisa Feldman Barrett 1:19:18 Okay?

Lex Fridman 1:19:19 Yep.

Lisa Feldman Barrett 1:19:19 Okay. Then people started studying ad hoc concepts, meaning, um, concepts that, where the instances don't share any feat- any physical features, but the function of the instances are the same. So for example, think about all the things that can protect you from the rain. What are all the things that can protect you from the rain?

Lex Fridman 1:19:48 Uh, umbrella, uh, sh- sh- like, this apartment.

Lisa Feldman Barrett 1:19:54 Right.

Lex Fridman 1:19:55 Um...

Lisa Feldman Barrett 1:19:56 Your car.

Lex Fridman 1:19:57 Not giving a damn.

Lisa Feldman Barrett 1:19:59 (laughs)

Lex Fridman 1:19:59 (laughs) Like, (laughs) like a m- like a mindset.

Lisa Feldman Barrett 1:20:03 Yeah.

Lex Fridman 1:20:04 (laughs)

Lisa Feldman Barrett 1:20:04 Right. Right. So the idea is that the function of the instances is the same in a given situation, even if they look different, sound different, smell different. This is called an abstract concept or a conceptual concept.

Lex Fridman 1:20:22 Yeah.

Lisa Feldman Barrett 1:20:22 Now, the really cool thing about conceptual categories or conceptual cons- or k- yes, conceptual category or conceptual as a category of things that are held together by a function-

Lex Fridman 1:20:35 Mm-hmm.

Lisa Feldman Barrett 1:20:35 ... which is called an abstract concept or a conceptual category, 'cause the things don't share physical features, they share functional features, there are two really cool things about this. One is, that's what Darwin said a species was.

Lex Fridman 1:20:51 (laughs)

Lisa Feldman Barrett 1:20:51 So Darwin is known for discovering natural selection, but the other thing he really did which was really profound, which he's less celebrated for, is understanding that all biological categories have inherent variation. Inherent variation. Darwin wrote in The Origin of Species about, before Darwin's book, a species was thought to be a classical category, where all the instances of dogs were the same, had exactly the same features, and any variation from that, uh, perfect pl- platonic, uh, uh, instance was considered to be error. And Darwin said, "No, it's not error. It's meaningful." Nature selects on the basis of that variation. The reason why natural selection is, is powerful and can exist is because there is variation in a species. And in dogs, we talk about that variation in terms of the size of the dog and the, uh, amount of fur the dog has and the color and the sh- how long is the tail and how long is the snout. In humans, we talk about that variation in all kinds of ways, right? Including in cultural ways. So that's one thing that's really interesting about conceptual categories, is that Darwin was basically saying a species is a conceptual category.

Lex Fridman 1:22:32 Yeah.

Lisa Feldman Barrett 1:22:32 And in fact, if you look at modern debates about what is a species, you can't find anybody agreeing on what the criteria are for a species-

Lex Fridman 1:22:42 Mm-hmm.

Lisa Feldman Barrett 1:22:43 ... 'cause they don't all share the same genome. We don't all share, we don't. There isn't a single human genome. There's a population of genomes, but they're variable. It's not unbounded variation, but they are variable, right? And the other thing that's really cool about, um, conceptual categories is that, um, they are the categories that we use to make civilization. So think about money, for example. What are all the physical things that make something a currency? Is there any physical feature that all the currencies in all the worlds that's ever been used by humans share?

Lex Fridman 1:23:33 Well, certainly, right? But, uh, but what, what is it? Uh, is it definable? You know, the, so, uh, I- it's getting to the point that you're-

Lisa Feldman Barrett 1:23:41 There is no physical.

Lex Fridman 1:23:42 ... you're making this function.

Lisa Feldman Barrett 1:23:42 There, it's the function.

Lex Fridman 1:23:44 It's the function.

Lisa Feldman Barrett 1:23:44 Right. It's that we trade it for material goods. And the, and we have to agree, right? We all impose on whatever it is, salt, barley, little shells, big rocks in the ocean that can't move, Bitcoin, pieces of plastic, mortgages, which are basically a promise of something in the future, nothing more-

Lex Fridman 1:24:01 Mm-hmm.

Lisa Feldman Barrett 1:24:02 ... right? All of these things, we impose value on them, and we all agree that we can exchange them for material goods.

Lex Fridman 1:24:11 Yeah. And, uh, yes. Uh, that's brilliant. By the way, you're attributing some of that to Darwin, that he thought-

Lisa Feldman Barrett 1:24:16 No, I'm, no, I'm saying that what Darwin did-

Lex Fridman 1:24:18 You, 'cause it's a brilliant view of what a species is, is the function.

Lisa Feldman Barrett 1:24:21 I, yeah, what I'm saying is that what Darwin... Darwin really talked about variation in, um... So have you read, for example, the biologist Ernst Mayr, who was an evolutionary biologist, and then when he retired, became a historian and philosopher of biology? And his suggestion is i- uh, that Darwin, Darwin did talk about variation. He vanquished what's called essentialism, the i- s- idea that there's a single set of features that define any species. And, um, out of that grew, um, really discussions of wha- uh, the funct- you know, like some of the functional features that species have, like they can reproduce, uh, off s- they can have offspring. The individuals of a species can have offspring. Well, it turns out-... that's not a perfect, uh, d- you know, that's not a perfect, uh, criterion to use. But it's a functional criterion, right? So, what I'm saying is that in cognitive science, people came up with the idea, they discovered the idea of conceptual categories or ad hoc concepts, these concepts that can change based on the function they're serving, right? And, um, uh, e- that it's there. Darwi- it's in Darwin and it's also in the philosophy of social reality. You can... The way that philosophers talk about social reality, just look around you. I mean, we impose... We're treating a bunch of things as similar, which are physically different, and sometimes we take things that are physically the same and we treat them as separate (laughs) categories.

Lex Fridman 1:25:58 But it feels like the number of variables involved in that kind of categorization is nearly infinite. And that's-

Lisa Feldman Barrett 1:26:05 Hmm, no, I don't think so because there is a physical constraint, right? Like you and I could agree that, um, we can fly in real life, but we can't. That's a physical, that's a physical constraint that we can't break, right? You and I could agree that we could walk through the walls-

Lex Fridman 1:26:21 Right.

Lisa Feldman Barrett 1:26:21 ... but we can't. We could agree that we could eat glass, but we-

Lex Fridman 1:26:24 Oh, there's a lot of constraints, but I- I just-

Lisa Feldman Barrett 1:26:25 Yeah. Yeah, we could agree that the virus doesn't exist and we don't have to wear masks.

Lex Fridman 1:26:30 Right. Yeah.

Lisa Feldman Barrett 1:26:33 But, you know, physical reality still holds the Trump card, right? But still-

Lex Fridman 1:26:37 (laughs)

Lisa Feldman Barrett 1:26:37 ... there's a lot of-

Lex Fridman 1:26:38 (laughs) The Trump card. Well-

Lisa Feldman Barrett 1:26:39 (laughs)

Lex Fridman 1:26:40 (laughs) ... pun unintended.

Lisa Feldman Barrett 1:26:41 Pun completely unintended, but there you go, that's a predicting brain for you.

Lex Fridman 1:26:45 Yeah, yeah. (laughs)

Lisa Feldman Barrett 1:26:45 Um, uh, but, but there's a tremendous amount of leeway.

Lex Fridman 1:26:49 Yes.

Lisa Feldman Barrett 1:26:50 Yeah. That's the point. So, I-

Lex Fridman 1:26:52 But, but-

Lisa Feldman Barrett 1:26:52 ... what I'm saying is that emotions are like money.

Lex Fridman 1:26:55 (laughs)

Lisa Feldman Barrett 1:26:55 They... Basically, they're, they're like money, they're like countries, they're like, um, kings and queens and presidents. They're like everything that we construct, that we impose meaning on. We take these physical signals and we give them meanings that, um, they don't otherwise have by their physical nature. And because we agree-

Lex Fridman 1:27:16 Yeah.

Lisa Feldman Barrett 1:27:18 ... they have that function.

Lex Fridman 1:27:19 But the, the... So the beautiful thing, so maybe unlike money, I, I love this similarity, is it- it's not obvious to me that this kind of emergent agreement should happen with emotion because our experiences are so different for each of us humans and yet we kind of converge.

Lisa Feldman Barrett 1:27:36 Well, in a culture, we converge, but not across cultures. There are huge, huge differences. There are huge differences in what cate- what concepts exist, what they're, um, what they look like. Um, so what I would say is that, um-

Lex Fridman 1:27:51 What they feel like.

Lisa Feldman Barrett 1:27:52 ... what, what we're doing with our young children as we... Uh, as their brains become wired to their physical and their social environment, right, is that we are curating for them. We are bootstrapping into their brains a set of emotion, uh, concepts.

Lex Fridman 1:28:11 Mm-hmm.

Lisa Feldman Barrett 1:28:11 That's partly what they're learning. And we curate those for infants just the way we curate for them what is a dog, what is a cat, what is a truck. We sometimes explicitly label and we sometimes just use mental words. W- um, when, you know, your kid is, you know, throwing Cheerios on the floor instead of eating them, or your kid is crying when, you know, she won't put herself to sleep or whatever. You know, we use mental words and, um, a word is this... Words with... For infants, words are these really special things that they help infants learn abstract categories. There's a huge literature showing that children can take things that don't look... Infants, like infants, really young infants, b- pre-verbal infants can take... If you label... If I say to you, and you're an infant, okay, so I say, "Lex."

Lex Fridman 1:29:08 (laughs)

Lisa Feldman Barrett 1:29:08 "Lexi, this-"

Lex Fridman 1:29:10 Yeah.

Lisa Feldman Barrett 1:29:11 "... is a bling."

Lex Fridman 1:29:12 Yeah.

Lisa Feldman Barrett 1:29:13 And I put it down and the bling makes a squeaky noise.

Lex Fridman 1:29:16 Yeah.

Lisa Feldman Barrett 1:29:17 And then I say, "Lexi."

Lex Fridman 1:29:18 I'm getting excited, by the way. (laughs)

Lisa Feldman Barrett 1:29:20 "This is a bling." And I-

Lex Fridman 1:29:22 Yeah.

Lisa Feldman Barrett 1:29:22 ... put it down and it makes a squeaky noise. And then I say, "Lexi, this is a bling." You, as young as four months old, will expect this to make a noise.

Lex Fridman 1:29:37 Yeah. That's awesome.

Lisa Feldman Barrett 1:29:38 A squeaky noise. And if you don't-

Lex Fridman 1:29:40 Just like that.

Lisa Feldman Barrett 1:29:40 ... if it doesn't, you'll be surprised because it violated your expectation.

Lex Fridman 1:29:43 (laughs) Yeah.

Lisa Feldman Barrett 1:29:43 Right?

Lex Fridman 1:29:44 Yeah.

Lisa Feldman Barrett 1:29:44 I'm building for you an internal model of a bling.

Lex Fridman 1:29:49 Yeah.

Lisa Feldman Barrett 1:29:49 Okay. Infants can do this really, really at a young age and so there's no reason to believe that they couldn't learn emotion categories and concepts-

Lex Fridman 1:29:59 Yeah, exactly.

Lisa Feldman Barrett 1:29:59 ... in the same way and in, in one c... And what happens when you go to a new culture? When you go to a new culture, you have to do what's called e- emotion acculturation. So, my colleague, Batia Mesquita in Belgium studies emotion acculturation. She studies how when people move from one culture to another, how do they learn the emotion concepts of that culture? How do they learn to make sense of their own internal sensations and also the movements, you know, the raise of an eyebrow, the tilt of a head? How do they learn to make sense of cues from other people using concepts they don't have but have to make on the fly?

Lex Fridman 1:30:40 So, there's the difference between cultures. Let me, uh, open another door I'm not sure I want to open, but, uh, difference between men and women. Is there, um, difference between the emotional lives of those two categories of biological systems?

Lisa Feldman Barrett 1:30:58 So, here's what I would say, you know, we did a series of studies, um, uh, in the 1990s, where we asked men and women to tell us about their emotional lives and women described themselves as much more emotional than men. They believed that they were more emotional than men, and men agreed.

Lex Fridman 1:31:16 (laughs)

Lisa Feldman Barrett 1:31:17 Women are much more emotional than men.

Lex Fridman 1:31:18 Okay.

Lisa Feldman Barrett 1:31:19 Okay? And then-We gave them little handheld computers. These were little Hewlett-Packard computers. They fit in the palm of your hand, couple of pou- they weighed a couple of pounds, so this was like pre-PalmPilot even, like this was, you know, 1990s and like early. And, um, we, um, asked them... We would, you know, ping them like 10 times a day and just ask them to report how they were feeling, which is called experience sampling. So we experience sampled and, um, and then at the end... And then we looked at their reports and what we found is that men and women basically didn't differ, and there were some people who were really... had many more instances of emotion. So they were, you know, um, they were treading, uh, water in a tumultuous sea of emotion.

Lex Fridman 1:32:16 (laughs)

Lisa Feldman Barrett 1:32:16 And then there were other people who were like floating tranquilly, you know, in a lake-

Lex Fridman 1:32:20 (laughs) Yeah.

Lisa Feldman Barrett 1:32:21 ... um, it was really not perturbed very often and-

Lex Fridman 1:32:23 Yeah.

Lisa Feldman Barrett 1:32:23 ... and everyone in between. But there were no difference between men and women. And the really interesting thing is at the end of the sampling period, we asked people, um, "So reflect over the past two weeks and tell us..." so, you know, we've been now pinging people like again and again and again, right? "So tell us h- how emotional do you think you are?" No change from the beginning. So men and women believe that they are... they believe that they are different и when they are looking at other people, they make different inferences about emotion. If a man... if a man is scowling, like if you and I were together and some- so somebody's watching this.

Lex Fridman 1:33:04 Uh-huh.

Lisa Feldman Barrett 1:33:04 Okay? And, um-

Lex Fridman 1:33:06 Hi.

Lisa Feldman Barrett 1:33:06 Yeah. Hey. We w- we were saying hey.

Lex Fridman 1:33:08 (laughs)

Lisa Feldman Barrett 1:33:08 Hey, hi, yeah, hi. Um-

Lex Fridman 1:33:10 By the way, people love it when you look at the camera. (laughs)

Lisa Feldman Barrett 1:33:14 Um, if you and I make exactly the same set of facial movements, when people look at you, both men and women look at you, they are more likely to think, "Oh, he's reacting to the situation." And when they look at me, they'll say, "Oh, she's having an emotion." She's-

Lex Fridman 1:33:35 Yeah.

Lisa Feldman Barrett 1:33:35 ... you know. Yeah. And I wrote about this actually, um, uh, right before the 2016 election.

Lex Fridman 1:33:46 You know what? Maybe I- I could confess. Let me try to carefully confess.

Lisa Feldman Barrett 1:33:51 But you are really gonna.

Lex Fridman 1:33:53 Yeah. That, um, that when I... that- that there is an element when I see Hillary Clinton, that there was something annoying about her to me. And I... just that feeling and then I tried to reduce that to, "What- what is that?" Because I think the same attributes that are annoying about her, when I see in other people wouldn't be annoying. So I- I was trying to understand what is it because it- it certainly does feel like that concept that I've constructed in my mind.

Lisa Feldman Barrett 1:34:30 Well, I'll tell you that I think... well, let me just say that, um, that- that what you would predict about, for example, the performance of the two of them in the debates, and I wrote an op-ed for The New York Times actually, um, before the second debate, and it- it played out really pretty much as I thought that it would o- o- based on research. It's not like I'm like