• njikomes

Donald Hoffman: Consciousness, Perception, Evolution & the Nature of Reality

Updated: Apr 7

Full episode transcript below. Beware of typos!

Nick Jikomes

Professor Donald Hoffman, thank you for joining me.


Donald Hoffman 4:18

Thank you for inviting me. It's very, very great pleasure to be here.


Nick Jikomes 4:22

Can you start off by just telling everyone a little bit about who you are and what your scientific background and research is all about?


Donald Hoffman 4:30

Well, I'm a professor emeritus now at the University of California, Irvine in cognitive sciences department. I taught there since 1983. And before that, I was I got my PhD at MIT in the brain and cognitive science department and also the Artificial Intelligence Lab. My PhD was in computational psychology. And before that, I went to UCLA got a degree in quantitative psychology at UCLA. So it's sort of like My educational, academic background.


Nick Jikomes 5:03

So psychology is a really, really big area. And I talk to people fairly often that sort of span, span the full spectrum there all the way from neuroscientists interested in psychology type questions to therapists and psychiatrists that are treating patients in different ways, all the way to people like you. So your PhD is in computational psychology. What exactly is that?


Donald Hoffman 5:31

Well, it really brought together the computer ideas from artificial intelligence, because I was in the Artificial Intelligence Lab at MIT, plus the neuroscience psychology side of things from the brain and cognitive science department. It was then called the psychology department a couple years after I graduated, they changed their name to the current name. So it was really the integration of computational ideas from artificial intelligence, with the emerging data that we were getting from neuroscience and rigorous Third reason, your cognitive science and psychology, trying to go beyond just paper and pencil stories about the brain and cognition with really trying to build mathematical and computationally precise models of the so it was, it was a great time I, I had two wonderful advisors of David Marr, and Whitman Richards and David Marr really pioneered this idea of bringing together artificial intelligence and neuroscience, he was one of the big pioneers of that, especially in the field of vision. And so I it was a privilege to get to work with David, for, for a while he he died while I was his graduate student, he, you know, died of leukemia. But, but But still, the idea was to combine mathematical models, computational models, together with the neurobiological data that we're trying to understand, and the psychological data. So


Nick Jikomes 7:03

interesting. So another, another thing that's gonna permeate the conversation, I think, is evolution and thinking about things in terms of in terms of evolution, and what sort of evolutionary logic can help us explain different aspects of human psychology? And so can you speak a little bit about evolutionary psychology? For those that aren't? aren't that well versed in it? What is evolutionary psychology? And why is it important for any kind of psychologist maybe, to have some grounding in evolution?


Donald Hoffman 7:36

Well, the theory of evolution by natural selection is one of the pillars of modern science, right? If you think about the big pillars, the big theories that guide modern sciences, it'd be quantum field theory, general relativity, special relativity, of course, and evolution by natural selection. And most of us think about evolution by natural selection, as describing that is the best scientific theory we have for describing how our bodies evolved, how we got the heart, how, why we have two legs, as opposed to six and things like that. So we tend to think of it as the theory about how different species get their bodies. But evolutionary psychology points out that it's not just your bodies, but your brains that have evolved and with the evolution of your brains, also your cognition, and your emotions, and your thinking ability. These are all from an evolutionary point of view, aspects of the human being that we should try to understand as adaptive responses to the environment that were shaped by natural selection. So it's not just that your heart and your muscles were shaped by natural selection, but your mind and your emotions and your thoughts, your cognitions are shaped. And so this then, you know, leads to the idea that we should try to mathematically model using evolutionary game theory, for example, how precisely people have evolved to interact, to compete mentally, how our various cognitive abilities evolved. And in the case of light work, how our perceptual abilities evolved, how sensory systems evolved. So so it takes evolution and says, if we're going to think about evolution by natural selection as a general theory about how organisms gradually over time evolved to be the way they are, we need to apply it to the mind as well. And so evolutionary psychology is is a fairly robust attempt at doing that. And my you know, I'm not saying that evolution by natural selection is the final word on things. In fact, my attitude about scientific theories is that there always works in progress. And, you know, when we look back on science 100 years ago, we can always point at the limitations of what They had and, and what they should have known and what they didn't know and what we now know. And I'm sure the same will be true a century from now that they'll look at our best theories and our best ideas and say, Well, you know, that was good for the time, but boy, they should have really been aware that there are these limitations to the theories. And we know a lot more now. So, so I'm not putting out evolution of natural selection as the gospel or anything like that. But I'm just saying, the reason I take it seriously, is because right now, we have no better theory to work with. And so it's our responsibility as scientists to take the best theories that we have on the table right now. Push them to the limits, see what what they entail, and also try to break them. So try to learn from them useful information in how the mind works, and so forth. And then try to break them and see, okay, you know, what's the next step after we're done with the current version of evolutionary psychology? What will be the next theory that transcends it? So I'm not doctrinaire about evolutionary psychology, but But right now, current state of Sciences is such that there's no better framework. So that's the framework that I'm using.


Nick Jikomes 11:05

So another person who takes the same general attitude as you in terms of how to understand the workings of the mind is Steven Pinker. And you referenced him in your book against reality. And I'll just read read a part that where you reference, Dr. Pinker, and and I want to get your your take on this. So you say that, in one of his books, he makes a surprising claim, quote, our minds evolved by natural selection to solve problems that were life and death matters to our ancestors, not to commune with correctness. And then you continue that, that observation is central, our minds were shaped by natural selection to solve Life and Death problems, full stop, they were not shaped to commune with correctness. So you really emphasize that point. What is that point? Can you elaborate on it for us?


Donald Hoffman 11:55

That's right. So Steven Pinker is brilliant. And he's a good friend, actually, even when I was a my last year as a graduate student at MIT was Steve's first year as an assistant professor at MIT. And I had the privilege of becoming a friend, being a student in one of those classes. And, and he's been a great help to me over the years. So I'm, you know, greatly indebted to Steven Pinker. And frankly, it was his book, how the mind works, that that really got me to study evolutionary psychology more seriously. So, I mean, I was aware of it and actually tried to get us to hire someone in the early 90s. In fact, Lita cosmetics, I tried to get our department to hire her, but But I hadn't really myself jumped in and studied it. And it was Steve Pinker's book that really got me to jump in and study it in the late 90s, and 99, or 2000. And the point about evolution by natural selection is the framework of the theory is that it's not about animals are competing. And the the one that's the best will, you know, is the absolute best of whatever he does is rather all you have to do is be better than your competition, you have to be a little bit more fit than the competition a little bit better, getting food a little bit better at reproducing a little bit better fighting whatever it might be, so you don't have to be the best. So it's not about optimality. It's just better than the competition, just enough to get past. And so there's nothing about really optimality in in evolution by natural selection is more what we call satisficing, finding solutions that are better than the competition and get the job done. And in particular, although most of us intuitively think that, you know, of course, we should see reality as it is. But how else could our perceptions be useful if we don't see reality? Evolutionary theory just basically says, sensory systems are there to enhance your fitness, full stop. If you think that to enhance your fitness, sensory systems should evolve to show you the truth. That is something you need to prove. That's not an assumption of the theory, the assumption of the theory is sensory systems evolved to shape organisms that are more fit, that can then other means reproduce, more successfully, have more offspring. So it's an interesting hypothesis to say, Well, the way sensory systems do that is they shape you to see aspects of the truth, maybe not the whole truth. In fact, knowing what things see all of objective reality, whatever it might be, but that, you know, we might be shaped to see some of objective reality. That's an interesting hypothesis. That is not written into evolutionary theory, that hypothesis that that's something you'd have to prove all the evolutionary hypotheses She says is reshaped, but with sensory systems that guide adaptive behavior, period. And so that's, that's what pinker was pointing out there. And, and I agree with him full, full heartedly and and, you know, we have many examples of this. And we, for example, there are good reasons for us to try to deceive others. And so, you know, you know, hunter gatherer society, if everybody go out, goes out and gathers, and comes back, or hunts and comes back and shares, you know, if I didn't get so, I didn't succeed today, but you know, my friend, Joe, you brought down a, you know, a bison, then, you know, I might ask Joe to give me some of his bison. And tomorrow, when I do better than Joe, I'll give him something on mine. So we might have some kind of cooperation that way. But in a, it's easy to show mathematically, that in a situation where everybody is cooperating in that way, then cheating is a very fit strategy, if I go out there and pretend to hunt and gather, but really what I'm doing is going down by the river, finding a nice shade tree and just relaxing and, and they'll come back at the end of day and go, Oh, Joe, you know, I had a hard day couldn't find anything, I have some of yours. And if Joe, you know,


is a kind guy just keeps doing that for me, then then I'm going to be far more fit than Joe because Joe's out there, you know, putting his life on the line wasting his energy, you know that I'm not wasting, you know, hunting down stuff. And so that strategy is more fit. And so you can actually show that the the deceiver the cheater is, if everybody else is cooperating, the defector, the cheater is more fit, and will reproduce more than than that. So. So then you get this evolutionary game or arms race in which if Joe can detect when I'm blessing him who can detect when I'm cheating, and I'm lying to him, then he won't be taken advantage of as as often. So he would be a little bit more fit. So now you find that there's going to be selection pressures for the cooperators to begin to detect the cheaters. But then, if I'm a cheater, if I learn how to be a better cheater, and I can still persuade the cooperators, then I could still be more fit. So you get this what we call an evolutionary arms race where the cheaters become better at deceiving. And the cooperators get better at detecting the cheaters and punishing them. You get all the emotions of anger and retribution and justice and so forth coming up as a result of this evolutionary process. And Robert rivers is quite famous in evolutionary circles for pointing out that the best deceiver, the best cheater who's deceiving is the one who is self deceived, who doesn't even know that they're lying. Because if you know it, you might suddenly be trained and you have blush of shifty eyes, your posture might change you might give away. But if you believe it deep down, deep down, you are completely self deceived, and you believe your life, then you're the best deceiver. And so so there are evolutionary pressures for us to be deeply deceived about our own motivations. So this is just one of one of many examples we can give where selection pressures don't lead to the truth, they could lead to destruction, not only of others, but of yourself.


Nick Jikomes 18:34

I think I remember vaguely, someone in my life saying that, like almost every deep scientific or philosophical question is explored at some point in the show Seinfeld. And so what you're describing is the George Costanza strategy. You know, it's not alive, if you believe it. Exactly. And so yeah, there's a really a lot of interesting dynamics that you can think about in terms of this evolutionary framework, especially, especially relevant to humans, or the human to human interactions we have and the emotions that govern those, I want to put a pin in that stuff, because it's often a little bit tricky for people to wrap their minds around. And I want to I want to build up some other concepts for people. So you know, thinking in terms of evolutionary psychology, how and why the mind evolved to be the way it is. How would you think about something like beauty, right? So a common aphorism is that beauty is in the eye of the beholder. And I'm wondering what someone like you makes a statement like that. And I know that in the book, you give a really interesting ecological example that ties into this had had to do with something called a jewel beetle. So So what's going on there?


Donald Hoffman 19:44

That's right. So from an evolutionary point of view, attractiveness is an evolved feature. In our ability to detect attractiveness is an evolved feature And so when I look at somebody, and in that initial half second initial glance, I often will get a hit, everybody gets a hit, you know of the attractiveness of the person. And it's not, even if you're not trying to do it, you just see it, you just automatically see it. We're wired up to make these evaluations automatically. And the, what's really going on there, what is it that makes you automatically see someone as more attractive than someone else, from an evolutionary point of view, what you're doing unconsciously, is one of the most sophisticated computations of your entire life, you are bringing in, through your senses, many, many visual auditory smell, and other cues about that person, dozens of cues, and you're doing an extremely important computation with all those cues, you are using those cues to estimate the reproductive potential of the person in front of you. What is the probability that this person could have or successfully raise kids, or have and successfully raise kids. And so that's not conscious. In the general case, you're not conscious of that. But that's what this is all about. So it's really an estimate of the reproductive potential of the person in front of you. Now, of course, to really know that you would have to know exactly what is the health of the person because you know, if the person has cancer, you know, terminal cancer, well, you know, their reproductive potential is very little, if they have certain genetic defects, the reproductive potential is less, and so forth. So but you can't, you can't see if they have cancer cells, you can't read their DNA. So you have to use other cues. And so you use cues like the the quality of skin, the some features of the eyes, something called the limbal ring that we could talk about, if you wish, something that we looked at in my lab, there are dozens of these queues that you're automatically looking at, and they would all tell the rights, the same story. So you, you are sophisticated enough to look at the queues and look at the ones that say this person has high reproductive potential versus the ones that say, no, they don't, and to adjudicate all of those and come up with a final answer. And that final answer you feel as the degree of attractiveness from very, very attractive to less attractive. And so that's the, that's just one idea about how evolutionary psychology evaluates these things. Most of us think, you know, a person looks great or not, and don't realize that there's this very sophisticated computation going on. And from it was this framework of evolutionary psychology of attractiveness. That actually led me and my graduate students to study something called the limbal ring in the eye and to show that that that feature of the eye is among the dozens or hundreds of cues that are picked up unconsciously, to evaluate attractiveness.


Nick Jikomes 23:04

So let's dwell on the eyes for a moment. The eyes are really interesting in the context of human behavior. Because a we're very visual organisms, we're visual primates, a lot of our cortical, or a lot of our brain real estate is devoted largely to vision. People intuitively naturally, automatically fixate on one another's eyes, right, that tells us a lot of information about where someone is looking. People are often very attracted to someone's eyes, as opposed to other people's eyes. And they may or may not sort of even be able to articulate what they're attracted to. And there's a lot of a lot of interesting information there that people often subconsciously exaggerate through various cultural mechanisms. So you mentioned one feature of the eye, and I'm wondering if you could just kind of point out for people, what are some of the things that people are consciously or not evaluating and another human beings eyes? And what is it that actually tied to in terms of the underlying biology?


Donald Hoffman 24:01

Well, we there's a lot that we look at in the eyes, we probably look at the human face more than any other object in our lifetime. And we look at the eyes more than any other part of the human face. So we spent a lot of time looking at people's eyes. And so it's not a surprise that there are a lot of features of the eyes that affect our estimates of the attractiveness, the reproductive potential of a person. So and it differs, female evaluations of male attractiveness, look at the eyes differently than male evaluation of female attractiveness in the eyes. So for example, you want to look for signs? Right, because one, one feature that, you know, declines with with age is your reproductive potential. As you hit 40 5060 know, the reproductive potential drops off and so if you're too young or too old, your reproductive potential declines and so there's going to be an optimal age range. And so you're really trying to estimate attractiveness. This is related to the optimal age range. So, and it turns out that youth is more critical in females than, than males in terms of reproductive potential, right, the females have a shorter window of reproductive potential than males. And so, so these are the kinds of selection pressures that will, females will not be as pressured to look for you within a male as males will be pressured by selection to look for, for youth and females. And so So here's one, a few concrete examples. If you look at a baby's face, the eyes are a big part of that face, right? If you so there, they have a small face, but they have big eyes. And as we get older, the ratio of the size of the eyes to the whole size of the face declines. So larger eyes are a sign of youth. And so male ratings of female attractiveness are enhanced by if the eyes are made a little bit big. So makeup that makes the eyebrows a little bit bigger, that puts eyelashes that emphasize with the eyes to make them look a little bit bigger, will make females look more attractive. Males with eyes that look too big, will look boyish, and therefore not interesting. So so so there's going to be a big difference there. People diameter. So your pupils dilate and contract if there's not enough light, they dilate if there's a lot of light they contract. But they also dilating contract due to your emotions. And, you know, if you find someone attractive, your pupils can dilate. And and so male ratings of female attractiveness go way up. If the picture of the female has dilated pupils, the more dilated the better males find that very, very attractive.


Nick Jikomes 27:00

I mean, are you? Are you saying that there have been experiments where identical photographs have been manipulated by researchers so that the eyes are slightly bigger and somehow somehow the the man reading it says this one's more attractive, even though he doesn't realize why.


Donald Hoffman 27:18

Exactly. So in one very funny and famous example, a psychology textbook had two covers that looked it was a beautiful woman's face, a young woman's face on it. But in one, in half of the books, they had dilated the pupils of the eyes, and the others, they were small. And they just put these books out there. And they just asked me a man to to pick up the dirt. They all look the same in beefy looked on, they're just all look the same. But they asked me which one would you like to buy. And the man always with for reasons they didn't understand, they always picked out the books that had the females with the dilated pupils. So this is this goes on under the radar. And you know, I spent a lot of time consulting for various corporations for many years, telling them this stuff, and they use it. You're being manipulated all the time, in advertisements and point of sale and everywhere. If I've helped them do this, they know how to make things look attractive, there's no I'm telling you a couple of these things, there are dozens and dozens of these things that can be used to. So So companies are using evolutionary psychology in their products, the product design, the advertisements, the videos that they put online, and so forth. So So yeah, this is, this is not this is a case where evolutionary psychology has proven us, it's worth so well that companies are paying big bucks to do stuff


Nick Jikomes 28:44

I see. And now, you know, sometimes we see things or other animals see things that you might call super stimuli. So things that resemble certain features that normally in most circumstances indicate some high potential fitness payoff, but are nonetheless not attached to an actual fitness payoff. You gave a really interesting example that I had not heard about before of these things called jewel beetles. And so what is that example? And what is that actually telling us?


Donald Hoffman 29:16

So this is the phenomenon called supernormal stimuli. In evolutionary theory, and these are the cases where what seems to be going on is that the visual system, the sensory systems of organisms do need to do things quickly. And with as few calories as possible. So we tried to, you know, find heuristics and tricks, shortcuts to get to the answer. So instead of giving, you know, a detailed analysis of exactly the the optimal AI and so forth, we might evolve. Some shortcuts. The bigger the eye, the better. The bigger the pupil the better. In other words, there may be the sort of shortcut added that are built into the computation so you can get the computation done right and quickly. And and even if you get it wrong a little bit, the fact that you can do it quickly and most of the time, you're right, it should be good enough. So there's a trade off between the most accurate kind of representation versus how quickly and how cheaply you can do it. So and we see this in the case of the jewel beetle. So this is a Beetle in the outback of Western Australia. The bills are dimpled, glossy and brown. The male's fly the females are flightless. But the male's fly around in looking, you know, for eligible females, and if he finds one he likes and mates. And but it turns out that in the outback, there were also you know, men who like to drink beer, and they have these beer bottles called studies that were also dimpled, glossy and just the right shade of brown. And so the guys were tossed the empties out into the Outback. And it turned out that the jewel beetle males would flock all over these bottles trying to mate. So they have full body contact with the bottles. And they would try to make and the females weren't interesting, because this was a super female, apparently, to these male juveniles. So the species could have gone extinct would they had to, you know, remove the bottles, change the bottles to save the species. So here's a remarkable thing that, you know, the jewel beetle had been around, you know, countless 1000s, maybe even millions of years, the male's had successfully found and mated with eligible females for for that whole time, you'd think that evolution had taught the males to know and recognize what a real female Yes, and apparently not, it just gave the male jewel beetles a hack a little trick. A female is anything dimpled, glossy and brown, the bigger the better. And and so when they found this dimpled, glossy brown bottle that was bigger than any normal female, there, that's a supernormal stimuli that was just far more attractive to them. And so they'd and so even being on the bottle and trying to mate not successfully didn't override these cues that were that they were using to determine attractiveness. And so we have the you know, the whole makeup industry is based on supernormal, some super supernormal stimuli. So women will put on lipstick that makes lips a shade of red, for example, that would never occur in nature. But men find more attractive, right? So so there's something going on there, right? Red lips are certainly healthier than blue lips, they indicate a degree of oxygenation in the blood and therefore health. But so but what you have in makeup is you take the cues that males are using to find females attractive, and making them go even beyond whatever occur in nature. The same thing happens with the kinds of cosmetic surgery that you might know, artificial implants and so forth will also do this kind of thing. So, so we use supernormal stimuli all the time, in makeup, and and so forth. So yeah,


Nick Jikomes 33:18

interesting. Yeah, no, I'm also struck by you know, the, the use of things like eyeglasses and contact lenses, not only to correct visual defects, but perhaps to to do other things that we might not even realize that we're doing.


Donald Hoffman 33:33

Yes. So you're just talking like context without any electronics in them or anything like that?


Nick Jikomes 33:39

Yeah, I mean, just, you know, you might wear a pair of glasses and actually, you know, magnifies the size that your eyes look to someone else or contacts that emphasize save the color or the change that outer ring in the in the eye.


Donald Hoffman 33:52

That's right, so one of the things that we discovered in my lab, and this is work with Darren, Patrick and, and others that did his PhD on this, we discovered that there's if you look closely at the eye, where the white of the eye meets the color part of the iris, in many people, you can see a little ring there. It's harder to see in really dark brown or black eyes, but if lighter brown eyes, blue eyes gray eyes, you can see in many cases this so called limbal ring. And it turns out that the limbal ring is very pronounced in babies and as we age it tends to be less pronounced. And being able to see a limbal ring requires also that the cornea be clear. So if you have cataracts or something like that in the cornea, you won't be able to see the limbal ring so, so it we hypothesize that, that people will find distinct limbal rings more attractive than non distinct and in bigger males in particular would find larger liberal rings in females more attractive, because they would indicate youth. So you get both youth and health as a signal from the liberal ring. So we did the experiments in one experiment, we would have males. So we looked at female ratings of male attractiveness, but the one that's really striking is the male ratings of female attractiveness, we have two pictures, the same female face, side by side, and the and they look identical, but we artificially changed the little ring and one of them in one face, the two little rings in the eyes of one face, we'd have some to make a little bit bigger, but it wasn't over the top right, it was so subtle that you wouldn't have to look forward to see that we done. But we just asked the males and the females to rate which of the two faces was more attractive. And people would look at them and go, What are you doing this? They're the same face? What do we say? Well, you're just, you know, humorous, just go ahead and just choose one. So they would they thought they were randomly choosing and yet, they significantly chose the face with the larger limbal rings over the other one, even though I had no idea what was going on. So this is again, the case that that shows that this was a super normal stimulus in the sense that we had taken ghost faces and enhanced the rings bigger than that face had, and perhaps a little bit bigger than it would normally have had. Anyway, so it was a slightly supernormal stimuli. And we unconsciously pick it up. And so so really, when you interact with a person, in the first 500 milliseconds, the first half second of interaction with a person, you're just unconsciously taking in dozens and dozens of cues, unconsciously and doing a very sophisticated analysis. And the end result is the probability that this person, you know, could successfully bear or have, you know, father or mother and raise offspring. So, so that's what we're really doing. But we just feel it as attractiveness, that attractiveness, which seems just like a knee jerk reaction is a very sophisticated summary of incredibly complex computation is going on.


Nick Jikomes 37:12

Now, we, we, we obviously know what we mean, when we say attractiveness with respect to other people, if I'm attracted to someone, versus being attracted to someone else, we all know what that means. We're also as people are attracted to things that aren't people were attracted to activities we might engage in, we're attracted to, literally inanimate objects and other kinds of in other senses of the word attracted. What exactly does that? How do we think about that in terms of fitness payoffs? So if I perceive something in my environment that's completely inanimate? What what does that have to do with something like fitness payoffs, if it's not another person that I have any reproductive potential towards?


Donald Hoffman 37:55

Right, so there's very, there's a lot of different answers for a lot of different contexts on this, but I'll just mention a couple. So in the case of various plants and fruits, right, again, what we're doing is our attractiveness is or the degree to which we find something interesting, might be dependent on how likely it has to be beneficial to us to eat. So we can look at various colors on fruits and textures of fruits. And we're very sensitive to that we can tell by looking whether it's right or not, and whether we want to eat it or not. So we have that kind of belly. When we look at larger scenes, like like big landscape, VISTAs, there's some evidence that we tend to find more attractive, those scenes that would be more likely to help us survive. So you might want to have some trees but not be in the middle of a forest. You might want to have a river and you might you want want to see it bending out of the beams so that there's more to explore there, you might want to be seeing everything from a slight Hill, but not Mount Everest, where you can actually see any enemies or predators that might come your way. So So in other words, our a feeling of a liking and disliking a scene, again, has to do with its effects on our reproductive potential and our fitness in terms of can we see predators, can we see prey? Can we avoid predators? Can we avoid being prey? And and you know, is there a place to explore for resources that no so, so we might have, you know, also an attraction to the austerity of a desert but, but in many cases, that's a different kind of thing in this this is a lot of our attraction is based on what would lead to you know, survival and reproduction consequences. Now, there are other levels, the on that as well. So we might, there are social pressures now, I mean, I may find that I start liking a certain kind of art because my in-group says that that's the kind of art I should like. And so I, so there are social pressures as well. So then that is fitness up, because conforming to my group gives me access to the group gives me feedback, you know, positive health in the group, and so forth. So we have in groups and out groups, that can also affect our judgments here. So that's another. So this this is it's a very, very rich field, it has everything to do with not only mating, but also eating the the areas we like to live, this, the groups that we want to associate with, and the fighting that we have between political parties, and international conflicts, and so forth. Evolutionary psychology is a powerful framework for understanding all of this.


Nick Jikomes 40:55

So and a lot of the stuff that we're really gonna keep digging into here, is going to boil down to our, our conscious perceptions, and what exactly are we seeing? What exactly are we perceiving when we perceive something? Ultimately, what we're going to talk about is what we're perceiving real? And what does that even mean to say that it's real or not? So before we do some of that stuff, can you just define a couple terms for people from from your standpoint? What is consciousness? And what is perception? What are those things mean to you?


Donald Hoffman 41:27

Well, so I'll start with perception. And I would say that it's standard in the field of psychology, evolutionary psychology and cognitive science, to think about sensory systems as organs in our bodies that take information from the environment, and represent that information in a way that can guide adaptive behavior. Those those that would be called perception. So perception occurs when we interact with the environment gather information from the environment, in, in a format that is useful for us to actually interact with that environment. So there's a perception perception action loop that goes on in this point of view. So we have perceptual systems, but But presumably, so does an amoeba, right, it doesn't have perhaps eyes, but it has some kinds of sensory systems allow it to maybe have more direct contact or, or voltage gated channels or something like that, that are that are being triggered. Now consciousness, the way to talk about consciousness is very much the way I would try to tell someone about the color red. Right? So if you were blind, I would have a hard time telling you what red Yes, right? I will, how do you say to a blind person? Well, it's it's like it's hot with it's, it's more like hot than cold or something like that. But you would realize that or anything that you can say, is really going to fall for short, you if you want to know what red is, you've got to see it. And that's the same thing with most of our sensory systems. Like, if I if someone had never tasted vanilla before, if I said, or they couldn't taste, and I tried to explain what the taste of vanilla was like, good luck. You can't, you can't explain it, you just you'd be. And so when I tried to say what I mean, by consciousness, I have to point to it in the same way, their point to the notion of your red apple to someone who is not blind. And, and so I would say so here's, here's the pleasure that I'll give you for what I mean by consciousness, when you open your eyes, and enclose them, you just had a change in your conscious visual experience. Whatever just changed, when you had your eyes open versus money closed, that's a change in your conscious visual experience. So I'll give you that pointer, and then I leave it to you to or if you have a siren is going by. And then suddenly, that turns off, that change is a change. That's not a free experience. But that silence versus the siren, that's an auditory conscious experience. And so So there are all these conscious experience of touch, taste, smell, and so forth. That these are experiences that that we that we have firsthand. And that's the best way I can point them out. So they're their experiences that we all seem to just take for granted.


Nick Jikomes 44:37

So one of the one of the interesting features of conscious experience that we all have and often take for granted is that it's it's it's unified. It's integrated, right. So when I when I look at you on the screen here and see everything in my field of vision. I'm not I'm just sort of seeing one holistic image of everything. I'm not I'm not perceptually, decomposing everything into its separate shapes and colors, and all of these things, I can talk about those components, I can create abstractions that try and pick apart the different aspects of the scene. But there's this sort of unified wholeness to our conscious experiences that exist, moment to moment, we feel like one entity having having one such experience nearly all the time. But there are interesting examples where consciousness can be split, and you started to talk about some of those in the book. So can you talk a little bit about some of the split brain experiments that can be done and what that says about the the unitary nature of consciousness?


Donald Hoffman 45:38

That's right, so your brain has two hemispheres a left and a right hemisphere. And you have about 86 billion neurons in your brain 43 on each side, and the two hemispheres are connected by a band of fibers, that's about 200 to 220 million axons, these are neural fibers. So it's like think about like a cable, a computer cable connecting two qubit, two computers together. So the left right hemisphere had this very, Twitter, million 220 million axon cable connecting the two hemispheres, the pass information back and forth between the two hemispheres. And actually, a friend of mine, Joe Bowgun, was a surgeon, he who, in various severe cases of epilepsy, where a person had an epileptic focus in one hemisphere and a bad part of the brain that was sending off random signals that would cause them to have seizures, when they couldn't be treated with drugs, they would, in desperation, try the surgery where they would open up the skull. And Joe and other there were a couple surgeons who did this, but Joe Bowgun, who was a friend of mine who did this, we take a scalpel and cut the corpus callosum, sometimes all of it, or sometimes just the part of that he thought would be enough to, to deal with. And the idea was that suppose the, the bad part of the brain, what we call the epileptic focus was in the left hemisphere. Well, when it goes bad, it's it sends the whole left hemisphere into an electrical storm. And that's what an epileptic seizure is, it's an electrical storm, the sword just takes you down. The reason why you're jerking around is because your motor cortex has all this random neural activity going through it. And so, and then it goes to the right hemisphere, because it goes over the corpus callosum, the information goes across the corpus close on the soil, the right hemisphere goes down. And so the whole brain goes down. So the, the idea of the surgery was, if we cut the corpus callosum, maybe only one hemisphere will go down, and the other hemisphere can stay conscious, or at least stay alert, so that you don't fall down hurt yourself, or if you're driving, you know, kill yourself and things like that. Well, the surgery was a dramatic success, it really reduced the frequency and difficulty of these epileptic acid episodes. So it was a clinical success. And a bunch of experiments then began to show that very interesting thing you can, it turns out that the right hemisphere is getting information from the left part of your visual field. So if I'm looking straight ahead that you everything to the left of so right now I'm looking at your nose, everything to the left of your nose is my left visual field, because I'm looking at your nose, everything to the right of your nose is my right visual field. So everything to the left of your nose is now going first to my right hemisphere. And everything to the right of you knows that that part of the visual world is going to Our left hemisphere. Now in normal people with the corpus callosum, the two hemispheres then quickly communicate and you also your eyes move around, and so forth. So so you can get the information to both hemispheres. But what they found was that they could do an experiment with these split brain patients, where they flashed on the screen, a phrase like key ring. So they have their person look@a.in, the middle of the screen. So here's a blank screen, and I just there's a dot, and we asked the person, please look at the dot. And then while they're looking, we flash up the phrase key ring where key is to the left of the dot, and ring appears to the right of the.so. We just flash it up for a 10th of a second, so that you don't have enough time to move your eyes. Right. So so when we do that, then the word key only goes to the right hemisphere. And the word ring only goes to the left hemisphere. And it turns out when you ask these people what they saw,


the person will say, Well, I only saw the word ring. And they'll say that because the left hemisphere for most people has the language speech centers. So when when I'm talking with you right now, is my left hemisphere is doing the talking. That's the side that controls speech. The right hemisphere can understand language It can cast, but it usually can't speak. But it can, it can pull up profanities, but it can't, can't talk. So the person when you ask them, What did you see, though, they'll say I saw the word ring. And you say, Well, what kind of ring wedding ring doorbell ring, a key ring, what what what kind of ring, and the person will go, I don't know, I just saw the word ring. And then if you ask the person with their left hand to you give a little box and put your hand in your left hand on the box and pull out what it is that you saw, and you have all sorts of things on there, there's a pencil, there's of spoon, there's a, there's a key, there's a ring, and so forth. And there's also a key ring, the left hand will go through, it'll pick up all sorts of stuff, and it'll come out with a key. And so you asked the person you. So if you do with the right hand, the right hand will pull out a ring, because the left hemisphere controls right hand. So if you have both hands work independently, the left hand will pull out a key while the right hand is pulling out a ring. And if you do this with a blindfolded person, right, the person just told you that they saw the word key, sorry, they saw the word ring. So they'll tell you with their mouth, I saw the word ring, when you blindfold them, have them pull up with their left hand, what they saw in the left hand pulls out a key and young blindfold and in looking, why did you pull out a key with your left hand, you said you saw rang, the person will either say I have no idea or they'll make up a story? Well, they'll confabulate. So it turns out that the two hemispheres can have completely different conscious experiences. One is having an experience of the word key, the other is having the experience of the word ring. And and it goes even deeper they they can have different personalities and different religious beliefs. A friend of mine vs. Rama conda, and his friends call him Rama. So Rama is a professor at UC San Diego, with one split brain patient, he found that the I believe it was the left hemisphere believed in God and the right hemisphere was an atheist. So you can have not only different conscious experiences, but different personality structures that are associated with the two hemispheres. And so this then raises a very interesting related question about consciousness and its relationship to the brain.


Nick Jikomes 52:16

You know, what is that? So I mean, that's a very dramatic example, where, you know, literally, these people have had a key component of their brain physically severed. And so that that helps us to understand why you get some of these traumatic split brain results that you see when you put people in these, these contrived situations in the laboratory. Is it possible that normal consciousness, even for a completely healthy, you know, awake person, like you or me, is split in ways that we simply don't even recognize or realize, to I mean, start to think about this, right? You can imagine, you know, if you just think about your own life, like you're not the exact same version of yourself in all social contexts, like I act differently, and I'm prone to speak differently, I do different things around my close friends, as compared to my colleagues or my family is something like that going on in the normal brain, potentially even to a degree that's, you know, beyond what we're even aware of.


Donald Hoffman 53:13

Yes, we have good evidence that there are all sorts of networks of activity on the brain and default mode, network attention network, there's various networks that are that are going on. And then the two hemispheres can have completely different personalities. And you might have a left hemisphere that wants to do one thing and a right hemisphere that wants to do something else. And you know, it's Friday night, and you're deciding whether to party or to study, if you're a student, for example, and maybe the left hemisphere wants to study on the right hemisphere wants to party, and you find that you are conflicted. They might literally be a conflict between two different personalities in one head, they're talking via the corpus callosum, and have different ideas about what to do on a Friday night. So there's, yes, there's, there's not just one, you we tend to think of yourself as just one you. But there's at least to us, possibly, in fact, probably with very, very different likes and dislikes. When when has been tested, the left hemisphere and right hemisphere, usually are systematically quite different in their personalities and what they like and dislike, as different as being atheist versus a religious, quite, quite different. And, and it seems to we can go down to smaller and smaller circuits in the nervous system and and find all sorts of interesting structure all the way down. So so you're not just one we tend to think of ourselves as one unified being but but from a neurobiological point of view. You have two hemispheres and each hemisphere has 43 billion neurons with various kinds of networks, default mode, network and so forth. And so you're a complex thing from neurobiological point of view.


Nick Jikomes 54:53

Interesting. And you know, one of the things that's worth talking about relatively early in the conversation here is You know, when when we're trying to understand something like consciousness in terms of its relationship to the brain. So when a neuroscientist is thinking about these things, they're often looking for these things that are called the neural correlates of consciousness. Can you explain what that what that means for people? And the extent to which? Well, first, let's just define what are neural correlates of consciousness. So when a neuroscientist is looking for such a thing, what are they doing?


Donald Hoffman 55:29

So there, I'll give a concrete example, just before I try to give a definition because I think the definition won't make much sense without the example. So there's an area of the brain called area v4, visual area for and roughly back here, on both hemispheres. And if you have a stroke, in say, the area v4 of your left hemisphere, you will have something called Hemi a chroma Tassia, you lose all color in the right visual world, you just see like a black and white picture Shades of Grey. And if you instead had the stroke in the right hemisphere, every before, then you would lose color in the left visual field, it would just be Shades of Gray, you'd so you'd have if I ever supposed to stroke as in my right hemisphere. So my left field is just black and light shades of gray. If I take a red apple and put it right in front of me, the the left half might look black and white, and the right half might look red. So I've moved over. Now it's all black and white, and shades of gray. When we were here, now it's red. So as I move the apple around, through my visual fields, I get different color experiences. So So here we have an interesting correlation between the activity of a part of the brain area before and a specific kind of, of color experience, no visual experience of color. And you can do something with a device called a transcranial magnetic stimulation device, it's a strong magnet. And you can literally just touch it to the skull and turn on the magnet, you but don't try this at home, you need to know what you're doing. Because you can actually fry your brain if you don't do this thing, right? So so you don't do this at all. But with a TMS unit and proper supervision, they can put it by area before and inhibit v4, and you will experience color drain from your left visual world. And then when you like when you turn off the inhibition, the color you see the color flow back into your visual world. So here, so now I can start to tell you what we mean by a neural correlate of conscious experience. So there's this specific conscious experience of color. And there's a specific area of the brain area v4 That seems to be correlated in this in this way. So technically, what we're trying to do are find the minimal neural circuitry that is correlated with specific kinds of conscious experiences, or the minimal circuitry that's available that's necessary for just being awake versus in a dreamless sleep back back. That's another kind of experience that we could talk about as well. So there's specific conscious experiences, and there's just being awake and alert versus flat out, you know, just gone unconscious, like under sedation. So they're the minimal neural structures that that are perhaps correlated, and most of my colleagues think, that are the sources of generation of the conscious experience.


Nick Jikomes 58:39

So at one point, in the book, you said that, quote, many experiments hunt for correlations in neural activity and consciousness, expecting that as the hunt succeeds and the list of correlations grows, a critical discovery will solve the mystery of consciousness, just as the double helix solve the mystery of life. So so let's just suppose right, 100 years go by 200 Years go by whatever it is, we learned more and more about the brain, we can define all of the cell types, we understand something in a lot of detail about the dynamics of brain activity. The idea is that at some point, the knowledge will be sufficiently dense, that we can create theories that will allow us to explain why we have qualitative experience at all right? This is really the so called hard problem of consciousness. And, you know, many people throughout the ages have have come come out on both sides of the aisle here. But do you believe that there will come a point where we know enough about the brain that we can actually solve that problem? Or is it actually insoluble?


Donald Hoffman 59:42

Well, first of all, say what most of my colleagues think that is that I would say 99% of my colleagues is in the cognitive neurosciences and also other fields, artificial intelligence that are looking at this kind of problem. I think that 99% of them assume that physical systems like brains, or perhaps computational systems, silicon chips and so forth, are part of fundamental reality. So space and time are fundamental. Physical objects in space and time are fundamental reality, as certain physical systems with the right kind of complexity, like brains, with their neural circuits, or perhaps computers with the right kind of programming of their circuits will have the right kind of complexity to generate conscious experiences. So this is, this is the standard view. So, it's what you might call a version of naturalism, space and time are fundamental particles, elementary particles are the fundamental reality of objects inside space and time, neurons are in computers are just vast, complex assemblies of interacting particles. And when you get the right kind of interaction of these particles, the right complexity, then conscious experiences emerge. And the example I gave a v4 would seem to confirm that right that, you know, it's if you stimulate, right, if you inhibit, before you lose color experiences, I should also mention if you stimulate v4, you'll get psychedelic color experiences. So you know, if you stimulate it, you get these wild color experiences coming out of it, as well. So clearly, you must be the v4 is causing the conscious experiences, right? If I take out before you lose them, if I, you know, stimulated you, you get them. Clearly. Area v4 is causing the color experiences. And I would say that most of my colleagues think that that's the case that we will, is a matter of time before we figure out how the physical systems generate the conscious experiences, right? So so, of course, saying that v4, when you stimulate before you get color experiences, when you ablate, v4 when you when you delete v4, ruin it, you lose color experience, that's not a theory about how v4 creates color experiences, but not at all. So there's no explanation there at all. That's just the data that you need to this is the good data that we're getting that needs to be explained. Why is it that v4 stimulation leads to color phosphine. So you call her experiences, and why is it that V v4 damage leads to color loss? So what's remarkable is I would say 99% of my colleagues think we'll have a story. But right now, yeah, and I can talk about the stories that are that are out there in the news, my friends, they're brilliant people. They're, they're interesting stories. But as I tell them to their face, there's not a single specific conscious experience that their theories can explain. So we can go into some of those. But here's the state of play right now. Everybody thinks that somehow brain activity or more general computer activity, or the right kind of complex activity, somehow gives rise to conscious experiences. But right now, we have no theory that can explain even one specific conscious experience, and really no theory that's believed by anybody but the person proposing it then in their graduate students. In some sense, it's, there's a lot of there's a lot of theories out there because no theory has really succeeded in explaining anything to everybody's satisfaction. So what my own tech is, this whole effort has missed an important fact that modern science has discovered in the last few decades. Right. So most of my colleagues in the neurosciences and so forth, computational artificial intelligence, are assuming an ontology of physicalism. So space time is fundamental. Physical objects in space time are fundamental. So that's the ontology and a methodology of reductionism, that as we go to smaller scales, in space time, we find more and more fundamental laws.


And that those two hypotheses, the ontology of space time and the methodology of reductionism have been spectacularly successful in science for several centuries. So they're incredible what they've done. But it's the glory of science, to be able to look at his own theories and transcend them to get new theories, right, for why we thought Newton was the final answer. In the 1890s, it was perfectly legally acceptable to think that physics was done and you could send smart graduate students to some other field because it was all over in physics. And then Newton, then Einstein comes along in 1905, and sort of Edmund plank comes along in 1900 and sort of blows that away. So our two biggest scientific pillars are now telling us something very important. This is evolution by natural selection on the one hand and also quantum field theory, together with gravity with the physicists are telling us is that space time is doomed. That's a quote, that's a quote from physicist named Nima or Connie Ahmed. Also Ed Witten, David gross, and others. This is not wild died weirdness from fringe physicists, these are state of the playing state of the art brilliant physicist right now are working, who have taken a very sober look at what our current physics is telling us. And they're saying, spacetime has had a good run. And it's over. We need a new, deeper framework than spacetime. And they're, they're actually finding deeper mathematical structures, so Nima or Connie Ahmed, and one Malta Cena at the Institute for Advanced Study at Princeton, so this is not a Podunk us somewhere. This is one of the most serious places for physics research in the world, they have something they call the cosmological politic. It's beyond space time. It's beyond quantum theory, there are no Hilbert spaces. It's so space and time are not fundamental objects in space and time are not fundamental. There's this deeper realm of the cosmological polytopes, and also something called the amplitude hedron. And these structures allow them to predict scattering processes at colliders like the Large Hadron Collider, better than the theories inside spacetime in the sense that, that the mathematics becomes more simple. And you see symmetries when you let go spacetime. So So physics is telling us, and this has just been in the last two or three decades that it's really become very clear that physics is telling us space time was had a good run, and it's over. So with spacetime, reductionism is dead, because you simply can't go to smaller scales.


Nick Jikomes 1:07:14

I see. So I to help me wrap my head around this. As a non physicist, when you say that some physicists are saying that space time is doomed. Does that mean that spacetime is a construct? You can think of it as a mathematical construct needed to get various equations to work out such that they actually predict experiments and observations in the physical world? Are you saying that there are observations that physicists have made that cannot be adequately explained? On last you suppose some other structure besides spacetime?


Donald Hoffman 1:07:49

Well, it's even worse than that. It's that the very theories themselves, like quantum field theory together with gravity, when you do the computation, it tells you that the very notion of spacetime ceases to even be sensible, when you get to a small scale 10 to the minus 33 centimeters, the so called Planck scale, the the very notion of six spacetime ceases to make any sense. And then when you look at quantum theory, quantum theory says you have to have a measuring device, and then the system that you're measuring. And suppose we have a room, we want to measure, say the the spin of the electron to arbitrary position. Precision. Well, quantum theory tells us that the measuring device is a physical system with its own degrees of freedom. And if you want to get because it's a quantum system, it's subject to an uncertainty principle. So if you want to get a more precise measurement of the position, or momentum or spin of an electron, you need to have more and more degrees of freedom in the measuring device itself, because itself has uncertainty. So as you add more and more degrees of freedom you're at, you're adding more and more mess to this measuring device. And at some point, what happens is, as you try to get more and more precision, the device itself gets so massive, that the room with the device, and the electron you're trying to measure collapses into a black hole, and your measurement is destroyed. And so there's a, what it means is that there are no local observables in space time. There's quantum theory with gravity, basically tells us there's no local observables in spacetime. There's nothing local that can be measured to very, very high precision. In fact, 10 to the minus 33 centimeters is all the precision that you can get in terms of spatial resolution, I would have been impressed if it was 10 to the minus 33 trillion, but 10 to the minus 33. I'm not very impressed at all spacetime is a very, very small tool. So it just gives up and what they found is if you now start Binding structures beyond spacetime like the cosmological Pollito pinapple to hedron, you can show how they project into space time, and can predict what you would see in space time. Like in the Large Hadron Collider when two gluons go in smash into each other and four gluons go sprang up, you can look at the probability is what they call the the amplitudes. For the various kinds of scattering events, if you do it with mathematics of spacetime, like using quantum field theory, the answer is pretty ugly, you get hundreds of pages of algebra, it's, it's very, very difficult. If you do it outside of space time in these new deeper structures, where you just let go spacetime, you can do something that might be hundreds of pages of algebra and spacetime, you can do in three or four terms. So the math becomes simpler. And then you also discover that there are symmetries in the scattering data that, for example, have something called a dual conformal symmetry that you can't see in spacetime, but you can see in these deeper structures, so So the physicists like Nima or Connie Ahmed, and others are all in now that's what their careers are doing. And then you can imagine why the the young physicists are all excited about this, right? Because it's something something new, they can discover something that so spacetime has had a good run, it's over the really bright people are now many really bright people are now scrambling to try to find what is beyond spacetime. What is it that goes beyond space time?


Nick Jikomes 1:11:28

So first, can I ask you, so that I'm, we're picking up like the pen or something on your desk on the microphone? One of the things so this is all fascinating stuff. I want to try and bring it back to some of the evolution perception stuff, and talk about it from a different angle, partly because this is this is so fascinating. And so beyond my area of expertise, that, you know, it's hard for me to even evaluate it, you know, evaluate some of these things that the physicists are doing. But so so with respect to space time, not from a physicist perspective, but say from a sensory neuroscientists perspective, can you talk about what spacetime is in terms of the brain and how it's generating perceptions? How would you actually think about describing what space time is from from the standpoint of a cognitive neuroscientists?


Donald Hoffman 1:12:26

So, most of my colleagues would say that the brain is estimating the true shapes of objects and distances of objects in space time. So we've been shaped my advisor, David Marr at MIT, in his book vision was very explicit that we've evolved to estimate the true shapes, curvatures of objects and their distances, and so forth. He Maher thought that no simpler creatures like flies, probably didn't have truths about reality, they just had sensory signals that sort of guided wing movements that were that were adaptive. But for us, we're actually estimating the truth. And there's a whole branch of work in my field, Bayesian estimation of visual features and auditory features of the world. And so so that's the standard view is that there is a real physical world out there that Apple really exists and has a shape and a color when it's not perceived. And my brain and your brain takes in all sorts of visual cues and estimates the true shape, and color and so forth, the distance to that Apple, so. So that's the standard view. I decided to check when by the way I was raised on my advisor, took their point of view, I believed it. But then I started looking at evolution natural selection, more more carefully. There's evolution is now a mathematically precise theory. This is called evolutionary game theory. And so you can use the tools of the mathematical tools of evolutionary game theory to ask a precise question. And the question is this. What is the probability that sensory systems have been shaped by natural selection? To see any aspects of true structures in the objective world? What is the probability that sensory systems have been shaped to see the truth or at least some of the truth? Clean technical question, right. But the nice thing is, you might say, well, that's that's the kind of question you have with your friends at Friday night after a few beers. And this is no this is a clean technical question. evolutionary game theory has the tools to answer that question precisely. And the answer is very, very straightforward. The probability that any sensory system has been shaped to see any Truth About objective reality, except for one truth that I can talk about, there's only one. But the problem is that has been checked to see any sensory truth, except the one I'll mention if you want, is zero. So the probability is precisely zero, that our perceptions of shape and color and motion and depth are true, then what evolution makes clear is that the very language of our perceptions of shape and color and position and, and sounds, and so forth, is simply the wrong language to describe objective reality. So it's not that we get the colors a little bit off, or the shapes a little bit off, is that there is no possible true description at all, in the language of our senses. That's what evolution entails. So so most of my colleagues will say, Oh, we don't see all the truth, but we just see the parts of the truth that we need. You need to look at evolutionary theory very, very clearly. And the reason the answer is the probability is zero that we see any aspect of reality is that fitness payoff functions, which govern our behavior, we, if we want, we can get into that. But the fitness functions which govern evolution, almost surely has no information about the structure of the world in them, they erase information about the structure of the world, I think,


Nick Jikomes 1:16:19

yeah, I think we should dig into that, because we need, we need an example or something here that that gets at what you just described, because it's it's extremely counterintuitive, it sort of flies in the face of all common, not only common sensical views of what the real world is, but but even, you know, very informed view. So you know, when I, you know, I'm just going to speak off the cuff, if I'm looking here at this cup, right, I can triangulate various aspects of the cup using a multitude of senses, that correlates perfectly well with what other people would use using any combination of their senses. And it's not difficult for me to say, well, I can understand that my perception is limited, right? My photoreceptors only pick up certain wavelengths, my somatosensory receptors in my skin are only going to be to pick up certain frequencies and textures and things, and so on and so forth. So I can buy very easily that the full or complete picture of reality might not be there. But because all of these different sensory systems across individuals seem to be triangulating the same basic thing. How is it that you can say that there's absolutely, or very little, almost zero chance that whatever, whatever the real cup is, whatever that is, is actually being described, even if imperfectly?


Donald Hoffman 1:17:39

So I'll answer two levels. First of all, just answer that. The evolutionary what's remarkable here is that evolutionary theory agrees with what the physicists are saying, space time is don't it's over. And objects with, with space time, go all the objects in space time, they're not fundamental. So what's remarkable is that the two pillars of modern science, both agree on this very counterintuitive, playing. So So now to your question, how can we understand this? So how is there have a metaphor that I think might be helpful. So imagine that you're, you put on a virtual reality headset, and you're in the metaverse and you're playing, say, tennis with your friend, who's also got a headset and body suit on. And so you, you pick up the tennis ball, and you hit it to your friend, and you hold up the tennis ball. Now, you and your friend both see the tennis ball. But now when you hold up their tennis ball, only when your head when you turn your head to headset in a certain way, do you see the tennis ball? Right? When you turn your headset that way? You no longer see the the tennis ball. Now, is there really a tennis ball there? No, this metaphor, the reality you're interacting with is a supercomputer with diodes and resistors and electronic voltages and lots of megabytes of software. That's the reality you're interacting with that all that hardware causes certain photons to be sprayed to your eye. And when they get sprayed to your eye, you create a tennis ball, but the tennis ball only exists when you look if I said, but the tennis ball is surely there when I don't look because look, I can hold up there my friend Joe, who's playing Virtual Tennis with me, Joe, can you see it and you go, Oh, yeah, I see it. The tennis ball is still there. Does that mean that the tennis ball really exists? Because Joe sees it? No, Joe's got his headset on. And he's creating his tennis ball. The tennis ball. The Josee is actually different from the tennis ball the icy because there is no single tennis ball. I create a tennis ball and Joe creates a tennis ball. We're interacting with some reality which in this metaphor is the supercomputer with all the voltages and magnetic field, so there is some reality beyond us. But it's not. You know, what I'm saying is think about VR if you think about virtual reality, and realize that when you You take your VR headset off and get out of the metaverse, you look around, I'm claiming you still have headset on evolution give you a headset,


Nick Jikomes 1:20:08

you're talking, you're basically just saying that the perceptual experience you're having at one moment is a rendition built from something but that rendition has essentially, there's, you can't derive the the something from from the rendition that you're experiencing.


Donald Hoffman 1:20:26

Exactly. And from an evolutionary logic, the reason is this, you don't need to see the truth to win the game. Right? Evolution is about being fed winning the game. To win, the only point from an evolutionary point of view is out competing, the others, that's all. So if you don't need to estimate the truth, and why bother, you can just out compete them. So. So to put it back into the VR game, suppose that I have some really super smart geek with that can actually look at the electronic voltages in the supercomputer. And so he actually, in this metaphor, he knows the truth, so we can actually toggle the voltages. And so we compete with him, I'm going to play tennis with him, but he has to toggle voltages really fast to hit the tennis ball. Whereas I can just use my, my VR headset. Well, he's seeing the truth, but good luck playing the game, I'm gonna win him, I'll beat him six love it, you know,


Nick Jikomes 1:21:20

that's actually a really good example. And I at least didn't get to the part of the book where if that's in there, so you could have the the smartest being in the universe, who actually can see all of the transistors and stuff inside the computer knows what state each logic gate needs to be at, to do the rendition. But if he's got to operate that apparatus at that level of detail, he's simply going to miss the game.


Donald Hoffman 1:21:45

That's right. Another example just to bring it home and make it make it seem sort of obvious. Suppose you're writing an email. But what you're really doing is toggling voltages in the computer. Well, good luck, if you if you had to toggle voltages in a computer to craft an email your friends wouldn't hear from you. Instead, what we have are these nice little icons on our desktop, right? Little eye candy that lets us click on Files and open them up and type stuff into them and enclose the files. But you know, so we have this blue icon on your screen for the the email that you're writing. But that doesn't mean the email itself was blue and rectangular. And, you know, if the if the icon is in the lower right corner of your desktop, it doesn't mean that the the email itself was in the lower right corner of your, your, your laptop, so So the whole point is, we have a user interface explicitly, to hide the truth. The truth is too complicated. You don't need to know the truth, if you had to know the truth, you wouldn't survive. And so we see none of the truth.


Nick Jikomes 1:22:41

So okay, so here's, here's where maybe you start to lose me. So I totally, I totally hear what you're saying about, you know, it's simply too much information for us to handle to understand all of the physical states of all the logic gates inside of a computer, we literally just couldn't use the computer for anything we use it for if we had to have that level of detailed information. But the iconography that I do interact with on my computer screen here. Certainly, you know, if I click on the the Mail icon, the the Mail icon looks like it's an icon with respect to a another perception I have, which is a physical envelope I might hold in my hand, it's not an icon with respect to the physical state of the underlying machine. But it is a it is there's an information compression that has happened. But nonetheless, isn't there still a systematic relationship between the icon and the underlying physical state of the machine, even though the the output is vastly different?


Donald Hoffman 1:23:44

Well, there there is. So the blue icon on your desktop for the email, there is a systematic relationship between that and something going on in the computer. It's very complicated. There's nothing green in your computer that corresponds to it. There's nothing rectangular that corresponds to it. And if the green icon is in the middle of your screen, it doesn't mean it's in the middle of your computer. So all of the predicates that you would use, your sensory predicates are the wrong predicates. To talk about how it corresponds with the objective reality, the objective reality that corresponds to is really this abstract code. That's that's coordinating all these you know, flows of electrons and so forth. That's it's it's incredibly complicated. What's going on. The the problem with the desktop in terms of making that connection is it's trivial compared to the reality. And that's the same thing. I'm saying our best science is telling us effectively, space time is trivial. objects in space time are trivial. They're a dumbed down user interface that we have mistaken for the objective reality. We are we're just we haven't been bright enough to recognize that we've been stuck in a headset and we've assumed that you know, what we see is What's what's their space and time and objects in space and time? We're now waking up. What's remarkable is, we've been sort of beaten over the head by our own theories. So quantum field theory with gravity, as being the physicist over the head and said to them, wake up space time is doomed. And now evolutionary theory is saying the same thing to us wake up. evolutionary pressures would not have shaped you to see the truth, they would shape you to see a user interface as much, much simpler than whatever the truth might be.


Nick Jikomes 1:25:29

So yeah, so this is what you call the interface theory of perception, the idea that perception isn't a direct window into reality, whatever exactly, that means. It's, it is a it is like a graphical user interface, very much like a computer screen. But an important point I would like you to briefly comment on and you're very clear about making this in the book is that you, you don't take perceptions, literally, but you do take them seriously. And so what why is that important to emphasize here?


Donald Hoffman 1:26:00

Well, very, very bright people have responded to what I'm saying by saying, look, obviously, what Kauffman is saying is ludicrous. If you think that that train coming down the tracks, that Twitter miles an hour is just a little icon in your virtual reality interface, why don't you stop in front of it. And after it's killed, you will know that the train was real, and it really can kill. And, and so and I get, I get emails all the time, from people who's saying, I don't care what you've done, the earth, here's the one liner that takes you down right? Now, if you things can hurt you things you if you don't pay attention to them, that you can be you can be hurt, and so forth. So I get this all the time. And I wouldn't step in front of that train. For the same reason, I wouldn't carelessly drag my file icon, my blue rectangular file icon to the trash can icon. Not because I take the icon, literally my file isn't blue and rectangular. But I do take the icon seriously. If I drag that icon to the trash, I could lose a year of work from Sam writing a book. So So we're, we have to take our perceptions seriously, from an evolutionary point of view, we evolved these perceptions to stay alive. You better take them certainly if you don't take them seriously. You're the one that won't reproduce. So absolutely. We have to take our perception seriously from an evolutionary point of view. But it's a logical error to to then claim because I must take them seriously. I'm entitled to take them literally. That's that's a logical error. It's it seems part of the human psyche that we just are inclined to make that logical error, brilliant people as well as non academics who haven't looked at this, we're all inclined, I was inclined to do the same thing we're all inclined to say, because I can stub my toe on that rock, as Johnson said to Berkeley, because I can stub my toe on the rock, or because that train could kill me or because that be could stink me, it really must be real. Well, yeah, I must take that be very seriously, I must take the train stirs seriously, I must not kick that rock with without shoes on, of course. But that doesn't mean that I have to take that I get to take my senses. Literally, that's a logical error. And so physicalism which has been the foundation of the current work on consciousness, you know, the neural correlates of consciousness and then studying to try to see how consciousness emerges. That that whole approach to consciousness is based on a simple, trivial logical error. And in fact, physicalism itself is based on a trivial logical error, because we must take our senses seriously, we assume that we're entitled to take them literally in space and time and physical objects really are the ultimate nature of reality. And in fact, they're a trivial interface. And the reality is far more complicated. And science has the tools to step behind the interface and go so until we get until, say, physicalism, is the biggest impediment now to the progress of science. That's so physicalism has been the biggest accelerator of science until now. So the physical strain work was really good for science. But now our theories are telling us physicalism had a good ride, it was very helpful time to let that horse go get off that horse, there's a new horse, beyond space time and beyond objects in space time. So we need for science to actually start to progress more quickly. We literally have to let go of physicalism and move on to the next phase of science. And consciousness itself will just be one aspect of you know, the neural correlates of consciousness and understanding conscious experiences, what they are and how they related to brains. Will is just part of that whole thing. But physicalism right now is the biggest is going to be the biggest obstruction to the next phase of scientific research.


Nick Jikomes 1:30:02

So, so there's a number of branch points here. One thing, so one thing I want to summarize is earlier you said, you basically said that the odds that evolution would have baked in a perceptual apparatus that accurately that accurate gave us perceptions that that describe reality as it is, is vanishingly small. We didn't we didn't name it, but you were sort of using there something called the fitness beats truth theorem. And I don't think we should dwell here on on any of the math or the details at that level. But you also mentioned that you said something like, you know, it's vanishingly unlikely that, you know, what we're perceiving? Is anything resembling reality, as it is at at a core level, save for one thing, what is what was that one thing that you mentioned?


Donald Hoffman 1:30:56

Right? So the world could have many different structures, you could have notion of neighborhood like what we call topology, how close are things together, you could have a structure of order, like one is less than two is less than three is less than 10. Right? That's an order to an order. You could have partial order. There's all all sorts of different structures of metrics and things. So there's countless structures you could imagine that reality might have. There's only one structure, that evolutionary theory in its formulation assumes that that is carried between perception and reality. And that's what are called what we call sigma algebra, or sigma additive classes. These are structures that are required for probability. And the reason for this is the reason we need that for for any scientific theory is, if the probabilities of events in reality are arbitrarily related to probabilities of events, where we can measure them in space and time, then science would not be possible. So for science to be possible. Now, it may not be true. I mean, we can't prove that it's true that our sensory systems reserve sigma, sigma algebra as the respective math symbols, so that the technical term of that there are measurable functions so that our sensory systems have a measurable functions from the sigma algebra sigma div classes, in reality to the smaller sigma algebra and smaller segment of the classes that we use for our perceptions in space and time. If, if that relationship is arbitrary, if it was not what we call homomorphism, then science would not be possible. So the fact that it's a scientific theory means it's assuming that measurable structures are preserved or not, are respected, not preserved, but respected. And but but any other structure is not required by the theory. And so force. So if someone wants to argue that we see that structure, but even out they either have to show how evolution in this current framework would automatically evolve systems to see that structure accurately, or to be homomorphism. Or they would have to explain, at a deeper level, a change to evolutionary theory, that would add that structure as a part of the theory that would necessarily be preserved. And so so the burden really is on those who want to claim that evolution preserve the structures because here's the bottom line, there's a paper we have called fact fiction and fitness. And people just Google my name. And in fact, fiction and fitness, you can see the technical work, but we prove is that if you look at the set of all fitness functions, so suppose the world has a structure like a total order 12345 through 10, something like that, that's a total order. And we and then we have our payoff a fitness payoffs, which maybe go from zero to 100. Zero means you're dead. 100 means that's the best you could possibly have and things in between. So you could then ask for all the, for all the states in the world, for that structure in the world, what are all the fitness payoff functions that are possible, and the only restriction is they have to be measurable, right? They have to be the ones that that preserve probability. So that's the only constraint on this other natural evolutionary theory gives us no reason to prune any fitness functions out. So you have to look at all the possible fitness payoff functions, then you can ask, okay, here's the whole set of Fitness Pal functions, and usually pretty trivial to write them down. Very easy. Then you can say, which of these payoff functions preserve the total order, which preserve this partial order, which preserve a topology? How many of them preserving so for any structure, you can ask the technical question, how many of these fitness payoff functions out of out of this huge, big, big bunch of path functions? How many actually contain information about that structure? And in every case, you find the answer is probably all these URL, probably zero that the payoff function even has the information. So if you're tuned Do the power function, and it doesn't have the information. Good luck, you can't get the information in your sensory system. So that's that's the short answer to it. We can we can go into more complicated things, you know, I've got pushback, the technical pushback, but that's the big idea. I'll just summarize it, pay off functions generically do not have information about structure in the world. So if sensory systems are tuned to the payoff functions, generically, they won't have infinite, they won't know the structure of the world, because they can't be tuned to the structure the world is the information out there in the payoff.


Nick Jikomes 1:35:37

Yeah, I think I think I got it. And there's a couple ways I want to get at this first one, I asked, Do you have a hard stop at 3pm? Yeah, okay, perfect. Because, because we're gonna need more. So, okay, I wanna, I want to start talking about language, and I think this might connect in, might connect into, to some of the things that you were just saying. So language is very interesting, at a number of levels, obviously, we use it as a communication tool, obviously, it also has adaptive utility for human beings. So it has something to do with fitness. It has some to do with perception, right? Because we have to perceive the speech of each other in order to even have this conversation. So so if we, if we're thinking in the context of what you call the interface theory of perception, so my conscious perception is, it can be analogize to the computer screen I'm looking at here, and it's got a bunch of different iconography in it, that actually allows me to do stuff in the world fluidly. And, and adaptively. And and be on my way, without getting bogged down in the just vast sea of information that would otherwise overwhelm me. So, you know, thinking of computer screens and icons here, the icon of the mail, the mailbox, say it's an icon rather than a symbol in the sense that, right, an icon resembles something and a symbol is arbitrary. So the icon on my computer screen for the mailbox, actually, is an icon of a physical mailbox I might go look at, right? So it's a perceptual icon, another perceptual icon. And so there's this kind of relativity there, where the icons are essentially defined with respect to other perceptual icons, that are sometimes in my experience, that starts to remind me of language, right? When you think about language and what it is, and its structure. Well, the words that compose languages are defined with respect to other words, and there is some kind of syntax, some logic that tells us how words can be combined and recombined in ways that allow us to communicate. And I'm wondering, I'm wondering how you think about language, whether or not there's there's something deeper going on here? Does the does the structure and logic of linguistic human language? Tell us something? Is there some now deeper analogy there between that and the structure of our perceptions? Or is language just sort of a special case of sensory perception?


Donald Hoffman 1:38:05

There's been debate about whether language is unique to humans and so forth. And and I think it's probably not, I think that there's probably very sophisticated linguistic kinds of systems that other animals have, but we'll see. But it's neither here nor there. From from our point of view, humans use language as a way of communicating with each other. And we have to play a language game. So So for example, when you have a 15 month old baby, and you're about to teach him some of his first words. And so say a mother or father is with the baby, and there's a rapid sitting on a blanket in front of it. And the mother or father points. So Johnny is sitting there and mom points and says, rabbit, and Johnny looks. And if he's old enough, he gets it. Right. He would if you think about so we're starting to teach the child how to play this language game. He sees Mom's hand pointer down 10 point he sees the rabbit. What's really interesting, though, is this called a stance of definition. So we're teaching allow extensive definition. And we do that all the time. We don't do it just with with kids. I mean, suppose you've never had you know, some, some food and you go to some South American country and have some food you never had before. So VGA or whatever it might be right. And someone gives you something they say they point and say so BJ, you taste it in a Whoa, I've never tasted that before. Right. And so that you have, again, a spice you've learned so Vijay by ostensive definition, whatever the thing we have, so Vijay here in Southern California, so it's not a big deal, but some food that they have that we wouldn't normally have. So, what's what if you think about what's going on there, though, it's really difficult. There's when a child looks at that whole situation. And the mother says rabbit, well, how does the child know what part of the visual scene she's referring to? Is it the color the white color of the rabbit? Is it the left ear? Is it the freeness of the rabbit? Is it maybe the left leg of the rabbit and three square inches of the carpet? That it's on? What what is it? Is it the whiskers of on the nose? Or is it the fact that it has only four legs? So so the child has an infinite range of hypotheses that he could possibly have about what mom means by rabbit. And we actually know if mom takes 15 month old Johnny and points and says, quadruped Ed, we know that mommy is doing something really disturbingly wrong, you shouldn't, you shouldn't say mammal or Quadra pen, you should say, you know, rabbit, and you shouldn't say what the what the particular species of rabbit is, you know, whatever the particular kind, you know, Cottontail, you should just say, rabbit. So there's we have built into us, without being given a university course mom knows, don't say quadropod, don't say mammal, don't say Peter Cottontail, say rabbit and the child knows. What's remarkable is the child knows to only look for what we call the basic level logic. So So language is really complicated we, we have to have sort of a shared assumption about our user interface. And then as we start to talk, we have to have shared assumptions about what part of that user interface you're directing my attention to, so that I know what the word means. And so this is a non trivial thing, the language game is very, very non trivial. This, this problem of ostensive definition arises everywhere. And, and so it's a non trivial fact that we can learn to coordinate or at least we think that we're coordinating. Sometimes when you spend a lot of time with a person, you thought you understood them. And then you find out later on, oh, no, there was something very, very deeply wrong in my understanding of that person's world.


Nick Jikomes 1:42:11

So, so another, there's a couple things here that I think will be interesting to get into. So one, you know, you're describing the sort of view of perception as you know, perceptions about perceiving or experiences about perceiving fitness payoffs, or fitness potential in the environment, rather than, you know, reality, per se, in the traditional sense of the term. connecting it with language perception here, I'm wondering if this is an interesting example, that maybe helps illustrate at least a part of what you're saying. So we quickly learned as infants to perceive some and to stop perceiving other phonemes. So the one the one example I can name for people, as you know, in English, obviously, we have R and L that are an L sounds, and we need to distinguish those two to understand and to use English. But that's not the case in Japanese. So a Japanese infant is born just like all infants, a bit with the ability to distinguish all of the different sound waves that correspond to the phonemes of all the languages. And yet, within just a year or two, I think the the Japanese infant will become untuned, or unable to discriminate the r and the l syllables. And that's why a Japanese person speaking English that they learned late in life has a certain accent and vice versa. So literally, literally, someone learns not to perceive two different states of reality, two different sets of sound waves, because in that language, it's not useful, and therefore doesn't tell them anything about the fitness that they're like anything about the fitness they're gonna experience in their environment at least. Is that right? Is that it? Does getting this idea. Yeah, yeah.


Donald Hoffman 1:43:50

That's, that's a pointer that we ignore. What you see there is that we're perceptually being trained to ignore things that don't affect our fit that don't enhance our fitness. Hmm, yeah. So now that most of my colleagues would say, of course, that just means that we're only seeing those aspects of the truth that we need, not the aspects of the truth that we don't need. So that wouldn't be so I'm saying something that goes a little bit further, I'm saying that we don't need to see any of the truth, except measurable structures. That's it. Right? That's the only structure that needs to be preserved according to evolutionary theory. And the rest of it, we don't need to see any truth.


Nick Jikomes 1:44:26

And if if I've heard everything that you said for last few minutes correctly, what so what you're saying is, we don't need to see any of the truth. But what you're not saying, correct me if I'm wrong, is that there is no systematic relationship between our perception and some underlying physical reality. You're not saying that there's, there's sort of an arbitrariness to that. You're just saying that nothing about the things that we can label and talk about in terms of our perceptions, allows one to deduce the underlying physical state of the universe that's responsible for those


Donald Hoffman 1:44:59

things. Exactly right when I see that yellow tennis ball in the virtual reality, all the predicates, fuzzy, yellow round, if I try to use those and say how, okay, so the reality must have some fuzziness or yellowness around. Those are just the wrong predicates to describe what's going on. So you have to literally let go of those predicates completely.


Nick Jikomes 1:45:22

So. So another thing I want to connect into here is, so I understand the argument you just made. It makes it's internally consistent to me, I recognize what you're saying as as valid on its own terms. Now, you're, I believe, still faced with the so called hard problem of consciousness, you still have to bridge that gap somehow, if you're interested in answering that question. Someone else that I just spoke to on the podcast named Bernardo kastrup, would say that the the mistake you're making or the difference in his viewpoint from yours, if I'm understanding correctly, is that the qualitative aspects of experience are the things that are there. And so the the hard problem of consciousness dissolves? In that way, it sounds like what you're saying is, there is this other physical reality, but we just can't sort of derive it from the aspects of experience that that we can describe using language. Am I missing something there? Or how would you start to think about


Donald Hoffman 1:46:22

that? You actually, I'm good friends with Bernardo, and on this week, we agree. So let me explain why we completely agree. So I've been just taking our best physicalist theories, namely, evolution by natural selection and quantum field theory and saying what they entail. That's what we do in science. Our theories entail that reality, whatever it is, is entirely unlike anything that we perceive. spacetime is doomed. That's what so that's what our best science tells us. So now, but they are theories do not tell us what is beyond space time. Those theories just tell us you there's something beyond space time, I can't tell you what it is. But we do know that whatever that thing is beyond space time, if we projected back into space time, you'd better look like quantum field theory better look like evolution by natural selection. Because those things work. Within our spacetime framework, they're really good theories within our framework are that what they're just telling us is that framework stops, and there's something beyond you need. So you need some new good ideas of what's beyond there, whatever those new Good ideas are, they're not, you can't just smoke anything you want to make any story you want. It better be mathematically precise, and a better project into space time in precise ways that gives us quantum field theory or something better inside spacetime, and evolution by natural selection or enhancement of evolution by natural selection. So so we're free now, as scientists do posit, whatever we want to outside, we can, in fact, we can smoke, whatever we want to while we're figuring it out, wise will be creative. But when we've when we made our creative move, now we've got to be rigorous, turned into math and projected back and see if it works. So I'm proposing that we take conscious experiences themselves as the nature of fundamental reality, so conscious agents, by something called Conscious agents, and their conscious experiences as the fundamental reality. So it flips everything around instead of taking inanimate matter, inanimate space and time and in an inanimate physical object, like protons and electrons, and so forth. As the fundamental reality. I'm taking something beyond space time to be we'll call these entities called Conscious agents, not necessarily human can not by a longshot, an infinite variety, potentially infinite variety of conscious agents, with experiences utterly unlike anything I've ever imagined. This is that so so we just have to posit this realm of, of potentially unbounded variety of conscious experiences and agents having those experiences, we have to propose a dynamics. So I think we Bernardo and I are good friends. We've talked many, many times in person and remotely. And he's an idealist. And this is an idealist theory that I'm proposing. But the difference is, and you know, no, I like to work with her on this. He's got the mathematical chops to do this, is he's doing it philosophically. But ultimately, as a scientist, I want to do it mathematically. Because I would like to get mathematical theories that we can then project into space time and show how we get done quantum filter, and so forth, coming out of it, and evolution by natural selection. So I take and again, I could be wrong. And in fact, I'll put it this way that I don't think any scientific theory to date will be viewed as the truth. You know, 100 years, it'll be viewed as it was really good for them. It was a good building block and we found something better. And I think that that's going to always happen in science. There's never going to be a final theory of everything. In principle, because every scientific theory starts with certain assumptions. And it says if you grant me these assumptions, then I can explain all this other stuff. But those assumptions aren't explained. They're the miracles of the theory. And if you say, well, I'll give you I'll get a deeper theory that explains those assumptions. Great, that new theory has its own new assumptions. And so you'll have your new set of miracles. And this is a principled aspect of scientific explanation, that there can be no final theory of everything. And so what I'm proposing is a hypothetical next theory, but not by any means, in my view of the final theory, just here's a interesting next step in the evolution of scientific theories. Let's try a theory in which we have these things called Conscious agents with their conscious experiences, let's work on a dynamics of this, let's projected back into space time and see if we can get something that looks like evolution by natural selection in quantum field theory. If so, that doesn't mean Oh, we've arrived, it just means Oh, we've taken the next baby step in our science.


Nick Jikomes 1:51:05

Okay, so there's a number of things I want to talk about. Basically, so maybe to help people think about some of this stuff, which which can get fairly abstract. So many people will have heard of Plato's Allegory of the Cave, right? So so these, you know, hypothetical people are in a cave, they're chained to the wall, all they can see are shadows being projected by a fire that's lit somewhere. And, you know, maybe there's people walking around in the cave and their stuff in the cave, but all they can really see are the shadows of those things. And I think someone like you or Bernardo would say our perception is sort of like that, we're only seeing the shadows, we're only seeing this sort of GUI graphical user interface, we're not seeing the stuff from which those shadows are coming. And I think what you're saying is, you know, the, the actual descriptor of reality is going to be some higher dimensional structure. And in order to define what that structure could be, we need to we need to just hypothesize, you need to make up some ideas what that structure could be, but no matter what we come up with, it should be dimensionally reducible to the shadows of perception, so to speak, we ought to be able to recover all the things that we observe, in our conscious perception with whatever it is the structure is that that you would use to replace the notion of space time is bad, I


Donald Hoffman 1:52:27

agree. So we're free to be as creative than irresponsible as we wish, as we're trying to come up with ideas. But eventually, we have to get responsible, and then say, Okay, I'm proposing that this is beyond space time, these conscious agents, and I've got to make a mathematically precise description of it. And its dynamics and a projection of, like you said, a projection into space time. And so we can measure and see if it matches with what we can observe here, the only difference, I would say, between what I'm saying and in Plato's Allegory of the Cave, so I agree with everything you said about that it's been, it's perfectly fine. I would just say one little difference. The shadows do resemble the objects, they have a similar shape. So we actually, whereas in a virtual reality example, that yellow fuzzy tennis ball does not resemble in any way anything inside the supercomputer. And that's why I mean, I think, Plato, if you were here would say, Yeah, that's a better allegory. I mean, I didn't have virtual reality. So I had to use what I could what I could use for to to make this. So I think, if you think about virtual reality, and I think what I'm saying well to the next generation is spends much of their life in the metaverse. What I'm what I'm saying is, this is going to be just playing obvious. I mean, our generation is not obvious. But the next generation, they're going to look back on these videos and go, what he was saying was news. That's silly naming. I take off my headset, of course, what I'm seeing around is probably another virtual reality. And this is just another heads up. I think it'll just be obvious. So I think our technology of the metaverse will really just have helped people see this and realize that there's the fuzziness of the tennis ball and the color has nothing to do with the reality behind it that that's causing me to see a tennis ball.


Nick Jikomes 1:54:11

You know, another thing I would love to ask you about to is, so you're saying you're saying that there's no requirement that our perceptions look like reality, and in fact, it's vanishingly unlikely that any one of them well, of course, it should still be a formal possibility that at least some tiny percentage of those perceptions might resemble as an icon does something about reality and against us language as an example. Our our language our words are mostly symbolic in the sense that the small mouth noises we make have no no logical correspondence to what they represent. And that's mostly true but it's not universally true, right? We have certain words that are iconic graphics, certain phonemes and phrases and things that resemble either the the The actual like physical sound wave patterns that the sounds come from, or they do have some kind of iconic graphic rep relationship to the reference in the world. But once you have just a small number of those, you can bootstrap language very quickly and start to use words that are truly arbitrary. And then that's most of what the language is. So, thinking roughly in those terms, I guess my question for you is, you know, even if we fully buy into your idea that our perceptions don't match up with what reality actually looks like, might there be some, some perceptual seeds, some some tiny percentage of our perceptions that actually do have that kind of correspondence?


Donald Hoffman 1:55:45

Well, the evolutionary theory says that the the answer is vanishingly small and precisely zero in the limit as we let the number of states go arbitrarily large. So it gets arbitrarily close to zero. Now, it is possible but but I would say another thing, evolutionary theory makes it very clear that that no connection with the truth is needed for the sensor systems to do their job. Just it's just not needed. The Jewel beetle doesn't need to know what a female does. And you know, if he happens to, you know, get hung up on bottles, the species may disappear. But but the chance of someone throwing out a bottle is very, very low, it only happened you didn't happen for a million years. And all of a sudden, some some Australian guys threw a mountain in. So bye bye to the Bielby. But in generally, the the sensory systems just don't need to correspond to the truth at all, except that probabilities and perception have to correspond to probabilities events in reality, but other than that, there needs to be no resemblance whatsoever.


Nick Jikomes 1:56:52

I see. So it's really the the only formal requirement here is this probabilistic correspondence, but But absolutely nothing else.


Donald Hoffman 1:57:02

That's right. So the probabilities of me seeing a green Fuzzy Ball are related to some really complicated probabilities of billions of interactions of electrons and diodes and resistors. Right. So it's that kind of thing. I just see this, this stupid fuzzy tennis ball, what's really going on, there are probabilistic events that are unbelievably complicated. So there is a correspondence but but it's like, billions to one. It was the correspondence in terms of probability. And the the language of the fuzziness and the color of the ball are just irrelevant to the description of what's going on in reality.


Nick Jikomes 1:57:39

And again, just maybe one last area in on the topic of language that I want to riff on for a minute is, you know, everyone will understand intuitively how I mean, language is powerful language is very useful. It's often fairly precise, but there's a lot of ambiguity in language. And there's a hell of a lot in our experience, that is nearly impossible for anyone to put words to just think about how hard it is to describe your emotions to someone else, who also seems to feel those emotions, let alone your example from earlier Donald of trying to describe a visual perceptive color to a blind person. And I'm wondering if the way to start to think about this is again, in terms of thinking of language in terms of fitness payoffs, rather than reality? So naively, if you ask someone, you know, what is language all about? So well, it's a communication strategy we use and, you know, if you think about all the different languages that exist, and that will exist, and that will evolve, you know, the ones you might say that the, you know, the ones that are going to do best are the ones that are the best at describing what's going on out in the world the best. But I wonder if that's not the way that you would think about language evolution?


Donald Hoffman 1:58:50

Well, there's been some beautiful work by some friends of mine on this. So Kimberly Jamison, Lewis neurons and Natalia carova, the University of California Irvine, they looked at colors on the on the color circle, and getting people to come to agreement on names for colors. And they, so they play so this is now sort of a game theoretic kind of thing. So using a game theory analysis of this, and you have people start making arbitrary carvings up of the color wheel, the color circle in and then they have to sort of converge on what they mean by the colors. And it turns out you can learn to groups people can learn to evolve toward the same shared boundaries on the colors in their language. What they found was one simulation, one study and not simulation actually experiment with with people when they've had trichromatic people who have you know, three color receptors red, blue, and green green. That we couldn't we shouldn't really call them Red when I grew up gluten


Nick Jikomes 2:00:02

red green, anyway, but the three the three receptors, yeah,


Donald Hoffman 2:00:05

yeah, that's right field. They're long medium short wavelength photoreceptors. So. So when they have tracker mats, just track them as doing this, they get one result. But when you have DICOM, as people who are colorblind, they're missing, say the blue cones or something like that, it turns out that the dichromatic have a bigger influence than the rest on the final color boundaries that are received. And that makes sense, because you have to play the language game such that the to include the people that have the the worst perceptual understanding. So really that intro,


Nick Jikomes 2:00:39

that is very interesting, for a number of reasons, including that I am red green colorblind. And so I, I know from direct experience, it's, it's just the best gets back to the difficulty and the ambiguity, language. It is so hard for me to describe for people what this is like, and for me to imagine what it's like for them. But I can sort of tell her surmise. And I also just know intellectually, this is this is what's going on that, you know, oftentimes when I see something and someone asked me what color it is, or the colors needs to come up in my speech, I probably can't tell if it's more red or more green. But I literally just kind of guess I think, and I want you to repeat what you said. So you said that the people with the perceptual deficit someone like me, in this case, who's colorblind have an outsized influence on the boundaries, and the concepts that people construct about where things are, is right. In other words, the people with the deficits are leading the people who actually see see things more cleanly, I guess,


Donald Hoffman 2:01:39

that's right, so that they can communicate, so that communication was possible, because if it otherwise would leave that person out, if we forced them to see all the color boundaries that we see in the language, they couldn't use them, as well. So it was it's an interesting again, that's just 111 simulation. But I'll just mention another interesting thing in the same direction. Kimberly Jamison is one of the people who's really taken lead on this, studying people, women who have four color receptors. So they have the long medium and short wavelength photoreceptors. But they also have another one, typically another version of the long wavelength. And so they have four color receptors. And so they're called Tetra combats, that instead of TrackMaster Tetra commands, and she's done careful experiments that strongly suggest that they have a richer color experience than the rest of us programmatic and certainly dichromatic. So it's not just they have one new color, they have a whole new dimension of color that the rest of us can't even imagine if I asked you to imagine a new color you've never seen before. Right? What? try real hard imagine the color you've never seen before. Nothing happens. These women, and I think your various estimates, it might be 1% of the population that might have this or other women have this, these women are seeing colors on a daily basis that you and I can't even imagine. So So again, we have to we play language game, they have to play the language game so that they can talk with us. So language works for us. But the tetra, tetra cannot see things that transcend the language that you and I have.


Nick Jikomes 2:03:18

I see. So they have an extra dimension of colors. It's not like they see a few more reds or a few more blues, it's these, it's these things that we can't even imagine. So on. This reminds me of another topic I wanted to ask you about, which is psychedelic experiences, which which have become, you know, fairly widely discussed recently for different reasons. Now, when people have a high dose psychedelic experience on something like psilocybin or DMT, they say a number of things that are very interesting. And I'm wondering if you can comment on these for us. One, they might say things almost like what you're just describing, like I saw colors that I didn't even know or colors, I saw I perceived things that not only I hadn't perceived before, but that I didn't know, were even available kinds of things to be perceived. They'll say things like, well, it's also very hard to talk about, right? It's, they'll say things like I can't describe it, I can't put it in words, it seems to be very hard to encode in memory, it's sort of slips slips away very quickly based on the perceptual weirdness of the experience. And finally, the you know, the other thing that I think they say that's interesting to get your take on is they'll say things like, there was a sense that things were more real than real. And so in the context of your ideas around perceptions, not reflecting reality, when we think about something like a high dose psychedelic experience, where people have these completely, sort of new and alien perceptual experiences, what how would you as a cognitive scientists think about what's actually going on there to create those experiences.


Donald Hoffman 2:04:47

Well, first, I'll say what my physical is colleagues would say, okay, so if I'm a hard nosed physicalist, and I'm a neurobiological reductionist, so I'm saying that the brain is fundamental reality and It's only a brain activity that causes my experiences, then, then these people that are on psychedelic drugs, and getting these different kinds of experiences, they're not perceiving any new fundamental nature of reality, their brains are just being out. But there's, there's just nothing more to say about the mean in terms of depth of reality, you might get personal insights out of emotional insights, and so forth. But the very the visual objects that you're seeing in the new colors you're seeing are not insights into a new, deeper reality, your brain is just being fooled with now, in the framework that I'm working on, in which space time is doomed. Physical objects in space time are not fundamental. And therefore, in fact, they only exist when they're perceived. Right? So So I'm saying, I don't have a brain right now. If you looked inside my skull, you would see a brain because you're creating the brain. Just like if I into VR with a tennis ball. There is no tennis ball until I look, when I look, there's a tennis ball. It exists in the act of looking, as soon as I turn away, that tennis ball is gone. It's nowhere, there is a reality, but it's not a tennis ball, right? That's the supercomputer. So there is no. So I want to be very clear about what I'm saying. I have no brain right now. There are no neurons in me. If you looked at me, you would, I would absolutely predict that you would see neurons. Absolutely. But that doesn't mean the neurons are there when you don't look, that just means like the tennis ball, I create the tennis ball when I need it. I delete it when I don't. When you look inside brains, you create inside heads, you create the brains, that's the way we create what we're seeing. Just like we create the VR tennis ball, we create neurons, we create that whole story. And as soon as we look away, that is gone. Our story, there's some deeper reality. And we have to recognize we just got this little tiny projection of a little tiny story we mistake it for the reality. It's it's a rookie freshman, sophomore mistake. Right? We have to get over that. That's that's why I'm saying this is the biggest obstacle to the advancement of science now, is this physicalist sophomore mistake. Our perceptions are just our perceptions are not an insight into reality, once we get used to that we can make the next step beyond. So I guess I'll stop on that.


Nick Jikomes 2:07:20

Yeah, so I mean, so how would you start to think about so so when someone has a high dose, psychedelic experience, say they they based on their reports, it would appear that they encounter qualia, that they have percepts that they've never encountered before and that are potentially like like your Tetra chromatic female example that some women have this extra dimension of color perception that people claim that there's these other dimensions of perception, whether it's colors or shapes, or topologies, or whatever, that simply never enter into entered into their conscious awareness prior to having that experience. So how would you think about what what those percepts actually are and what they're what they're reflecting?


Donald Hoffman 2:08:05

Well, so what my colleagues, as I said, have one answer. And this is fixed, there's nothing beyond space time. So if they, if they think they're seeing dimensions beyond space time, they're wrong. But in this other framework that I'm talking about, in which space time is doomed, and in which I won proposing a theory in which conscious agents are fundamental, there's an infinite variety of experiences that are far beyond anything we've experienced, it opens the possibility that the experiences that people have on drugs, could be new portals into this realm of conscious agents and new portals into new kinds of experiences. It doesn't in tail that that's the case, it allows that that's the case. So what we have to do, as responsible scientists is to develop this theory of conscious agents. And this projection into what we call our brain and the nervous system, we have to see how the default mode network and area v4 and all these different structures of the brain in their activity, are a projection of this deeper dynamics of conscious agents, we then have to model what what five Meo DMT is doing to neurotransmitter systems in the in the, in the nervous system of the brain, do a pull back from that into this theory of conscious agents just say what we think of that corresponds to the theory of conscious agents. And then we can try it in that that context to say, that is possibly a real new insight into a new dimension of conscious agents that we never had before, or no, that was really just like, the drug did a hardware error. And you got a bug from it, right? So we don't know which is, but at least the theory of conscious agents leaves open the possibility that drug experiences in some cases could be insights. It also leaves open the possibility that they're just errors, because they're causing malfunctions and so forth. So we'll see whereas the physicalist framework, there is nothing beyond space time. And so there's no insights that you could be having to go beyond space time.


Nick Jikomes 2:10:10

So, you know, when we use some of these philosophical terms like physicalism, or idealism and things like this, or naturalism or other things, how would you so so your way of thinking about things is not physicalist. But most scientists would be considered physicalist? In a presumption that many people would make or concern they might have as well, if, if Donald is not a physicalist, there must be some kind of airy fairy or quasi religious metaphysics that's being smuggled into this. Is that accurate? Or can you? Is it fair to say that your viewpoint is completely naturalistic, even though it's not physicalist?


Donald Hoffman 2:10:50

I view this is completely naturalistic, I'm just saying, our best theories tell us that space time is doomed. As a scientist, I have to accept what my best theories tell me and try to come up with a deeper theory I have to come up with it's in the best interests of science as a natural scientist, to say okay, space time was in fundamental, I was wrong. I thought it was fundamental. I thought that electrons and protons and quarks and gluons were fundamental, they're not. Okay, well, I can cry over for a few weeks now. Nellis nom over. Now let's go on and find out what a deeper theory, and it's going to still be the normal science in the sense that whatever I propose out there, you better project back into space and time and give me back the scientific theories that we know and love, and better give me back, you know, empirically verifiable experiments to test to test my theory. So it's entirely naturalistic in the sense that I for every entity that I propose on proposing a mathematical dynamics, and proposing experiments that we can go to test it. And also, even before we get there, showing how I can get all of our current scientific theories as a projection of this deeper theory, right, so I'll just give you just a one can't about the kind of thing that we'll have to do. So time. Right, our universe, our physical space time, universe seems to have a very finite time started 13 point 8 billion years ago, and the Big Bang. And there's this entropic time and entropy keeps increasing and time, time increases. And it seems like they're very intimately linked that in some sense, our time is a measure of entropy, and sometimes, in some sense. And in his theory of conscious agents, now we have a mathematical model in which you have, I haven't gotten into the mathematical model at all, but I'll just say that, that we have these spaces of possible experiences. So the probability spaces that have possible experiences, and agents can have experiences and then can affect the experiences of other agents.


As it's referred to go all the way there, so I'm holding here, but maybe I'll want to just say that, that we can then take that, and we can have a notion of a dynamics in this assault, I'll just say briefly that the dynamics of conscious agents, but it did not have an increasing entropy, we can have, for example, something called a recombinant dynamics, that's the stationary. So we got this dynamical system of conscious agents. It's a big dynamics really complicated, but it's stationary. So there's, there the entropy does not increase with the time parameter, the step parameter of that dynamics. But it's, it's easy to show trivial proof that if you have this Marco into that mix that has no entropic time, if you take any projection of it, like by conditional probability, so you look at it from a perspective, any projection from a particular perspective, you will induce entropy, increasing entropy.


Nick Jikomes 2:14:05

I see. So what you're saying is time I see. So what you're saying is, if I'm understanding correctly, there are certain dynamical systems that one can construct with properties such that there's no entropy, there's no disorder though. No increasing entropy, no increasing entropy, no increasing entropy, entropy over time, like, like what we see in our, the physical universe that we're perceiving right now. But if you take any lower dimensional slice of that, if you look at a shadow of that system, it will in all cases give you just that.


Donald Hoffman 2:14:40

It'll give you all of a sudden now the entropy is increasing with your with your so you get this entropic time. So so who knows in this realm of conscious agents, there may not be any entropic time. And so, like time is one of our most important limiting resources right now. Right right time One of the most valuable resources that we have, we compete because of time, time limits. So this this world of conscious agents, the whole notion of of competition may not be there, because the resource of time isn't, isn't a problem. So but when we look at the dynamics and projected into space time, all of a sudden we're gonna get this artificial time. This entropic time is an artifact of the projection. And so maybe evolution by natural selection will be the artifact, way of looking at our dynamics that you and I are, appear to be competing for finite resources. And that's just a limited projection perspective of a deeper dynamics in which it's not true that we're competing for resources, and there is no entropic time. So So as much as I've touted the the glories of evolutionary psychology and evolution of natural selection, it may be that that whole theory is is really revealed as an artifact of the projection of a deeper dynamics in which there is no competition. We're not competing for resources in the hole that will say,


Nick Jikomes 2:16:06

interesting, well, it sounds like Donald that you're, you're still actively doing research or running a lab, it sounds like maybe you're not.


Donald Hoffman 2:16:14

I'm actively collaborating. I no longer am working with graduate students, but I'm working with with postdoc, and several professors. So we're continuing to work. And we're interacting with, for example, there's a lab at Yale University that recently sent us paper, some simulations are doing this trying to contradict our work on evolution by natural selection and not seeing the truth. So we're, we're writing a paper and response and so forth. So so a lot of good work is going on. But but the work I'm planning to really focus on, I was waylaid for a little bit with COVID. So for the last year, basically, I had long COVID. And it took out my heart and I had to have surgery. And so this last year was pretty much a loss because of COVID. But but now I'm getting back into working on this dynamics of conscious agents and trying to build a model of it and show how what I'm hoping to do is to show how the the so called asymptotic behavior of the long term behavior of these conscious agents gives rise to structures that the physicists are finding like the cosmological polytopes and the amplitude hedron. If I can do that, then they take me from the cosmological polytopes. And after to hidden into space time, and so I can go all the way from the dynamics of conscious agents through one projection, namely, the the long term behavior, the asymptotic behavior as a kind of projection of the dynamics, instead of seeing the whole dynamics, you only see it in the long term behavior. So that's a projection. So that projection gives rise to the structures that the physicists find behind spacetime, like the cosmological Paulito. That's my hope. And then we can go from there and just based on, so it's a big project, but the idea would eventually start with dynamics of conscious agents go through its asymptotic through cosmological poly took Napoli to hedron into space time, and then make predictions that we can test. Ethical, and then I'll have lunch.


Nick Jikomes 2:18:12

With Donald, this is all very interesting stuff. I enjoyed reading your book, even though didn't get all the way through by the time that we spoke today. Do you want to tell everyone, like the name of your book or name of any other books that you have and where they can find you?


Donald Hoffman 2:18:27

Sure, you can. My most recent book on this is called the case against reality, the case against reality. And then the subtitle is why evolution hid the truth from our eyes. And so that came out in 2019. I have a book called Visual Intelligence, how we create what we see. So this goes through in great detail about how you create the colors and objects and shapes that you see. It's great. It's been used as a textbook for photography and art classes, and so forth, but also for for Neuro psych cognitive neuroscience classes. And then, for those who want some real nasty math, I have a book with Bruce Bennett and shaytaan. Prakash called observer mechanics, which goes into an earlier version of this conscious agent during dynamic selectors. So it's called observer mechanics, a formal theory of perception. And I've got a TED talk, if people are interested, I've got a TED Talk. If you just Google my name, and TED talk, you can see where I've talked about these ideas for a very broad audience and sort of compacted them all down into only 18 minutes. So


Nick Jikomes 2:19:34

all right, well, Professor Donald Hoffman, thank you for joining me.


Donald Hoffman 2:19:37

My pleasure. Thank you very much.


Transcribed by https://otter.ai


22 views0 comments