WEBVTT - Ep83 "Why Do Your 30 Trillion Cells Feel Like a Self?" Part 2 0:00:05.200 --> 0:00:09.080 Why do you think of your thirty trillion cells as 0:00:09.080 --> 0:00:13.160 a self? Does an ant colony have a sense of self? 0:00:13.600 --> 0:00:16.040 And could you think of all those ants as a 0:00:16.160 --> 0:00:19.120 liquid brain? What does any of this have to do 0:00:19.640 --> 0:00:23.520 with how the brain of a caterpillar transitions to the 0:00:23.560 --> 0:00:26.840 brain of a butterfly, or how we might think of 0:00:26.920 --> 0:00:31.520 a memory as a pattern that stays alive and has 0:00:31.560 --> 0:00:37.680 its own life. Welcome to Inner Cosmos with me David Eagleman. 0:00:38.240 --> 0:00:41.559 I'm a neuroscientist and an author at Stanford and in 0:00:41.640 --> 0:00:45.320 these episodes we dive deeply into our three pound universe 0:00:45.680 --> 0:00:53.640 to uncover some of the most surprising aspects of our lives. 0:00:59.240 --> 0:01:02.120 In the last piso, I talked about who we are 0:01:02.240 --> 0:01:06.920 and how we change through time. The pieces and parts 0:01:06.959 --> 0:01:10.840 of every single cell in your body degrade and get 0:01:10.920 --> 0:01:16.679 replaced continuously, such that you are physically speaking, a totally 0:01:16.760 --> 0:01:21.360 new person every few years, and yet we experience the 0:01:21.480 --> 0:01:24.840 illusion that we are the same person we've always been. 0:01:25.040 --> 0:01:29.399 We have this illusion of constancy. So last week we 0:01:29.520 --> 0:01:34.199 explore this by considering the thought experiment of the Ship 0:01:34.319 --> 0:01:35.319 of Theseus. 0:01:35.360 --> 0:01:37.080 The story here is that. 0:01:37.200 --> 0:01:41.240 Each plank of a famous ship is replaced one by 0:01:41.360 --> 0:01:45.920 one over time, which raises the question, is it's still 0:01:46.040 --> 0:01:50.720 the same ship even after every plank has been replaced 0:01:50.920 --> 0:01:55.320 and nothing of the original remains. We consider this question, 0:01:55.400 --> 0:01:59.440 of course, because like the ship, we too exist with 0:01:59.520 --> 0:02:04.680 physical changes that never stop, and yet we perceive ourselves 0:02:04.720 --> 0:02:09.720 as constant through time. The planks of theseus's ship map 0:02:09.800 --> 0:02:12.800 onto the cells and molecules in our bodies. So what 0:02:13.000 --> 0:02:17.360 maintains your sense of self over time? Okay, so maybe 0:02:17.360 --> 0:02:21.520 the thing that links our different selves through time is 0:02:21.639 --> 0:02:25.639 our memory. But there's a problem here as well, which 0:02:25.680 --> 0:02:31.600 is that memory is notoriously unreliable. It's constantly being reshaped 0:02:31.600 --> 0:02:35.320 and revised. So this is all very strange. And there's 0:02:35.360 --> 0:02:39.040 an added difficulty, which is that we're constantly changing, and 0:02:39.120 --> 0:02:42.200 even when we recognize that we have changed, we always 0:02:42.240 --> 0:02:45.919 assume that we're not going to change very much into 0:02:45.919 --> 0:02:49.480 the future, and that's always incorrect. This is a cognitive 0:02:49.520 --> 0:02:53.400 bias called the end of history illusion. We tend to 0:02:53.480 --> 0:02:59.200 believe that our current preferences and personalities are fixed, even 0:02:59.280 --> 0:03:03.480 though we will in fact continue to evolve. So that's 0:03:03.520 --> 0:03:07.839 our situation. We constantly change, but we're reaching back into 0:03:07.960 --> 0:03:11.200 our not so good memory to try to understand who 0:03:11.240 --> 0:03:12.120 we are, or. 0:03:12.120 --> 0:03:13.080 At least who we were. 0:03:13.600 --> 0:03:17.200 All this encourages us to start looking for new frameworks 0:03:17.240 --> 0:03:20.160 when we think about the self. For example, what if 0:03:20.160 --> 0:03:23.919 we thought of the memories themselves like their own little 0:03:24.000 --> 0:03:28.240 creatures and they are competing to stay alive. So in 0:03:28.280 --> 0:03:31.600 this episode, I want to dive into new ways of 0:03:31.639 --> 0:03:33.960 thinking about all of this, and there is no one 0:03:34.000 --> 0:03:37.760 better to ring up for that than my colleague Michael Levins. 0:03:37.840 --> 0:03:40.480 He's one of the most energetic and original thinkers in 0:03:40.520 --> 0:03:45.960 the field. Michael is a developmental biologist and a synthetic biologist, 0:03:46.080 --> 0:03:49.560 meaning he makes new kinds of organisms. He's a Toughts 0:03:49.680 --> 0:03:53.160 University and what I love about Michael is his extremely 0:03:53.320 --> 0:03:59.280 broad interests like bioelectrical signals by which cells communicate, and 0:04:00.080 --> 0:04:04.320 what cognition looks like across totally different body plans, and 0:04:04.920 --> 0:04:08.600 how you get similar forms and functions across all kinds 0:04:08.600 --> 0:04:11.800 of different scales and biology. So I'm going to have 0:04:11.880 --> 0:04:14.400 Michael back to talk about some of these other topics 0:04:14.400 --> 0:04:17.640 on a future episode, but today I want to zoom 0:04:17.680 --> 0:04:21.160 in with him on his recent thinking about the self 0:04:21.640 --> 0:04:29.200 and memory. So Mike, tell us how you think about 0:04:29.480 --> 0:04:30.440 the self. 0:04:31.160 --> 0:04:33.599 I work in diverse intelligence. I'm interested in all different 0:04:33.680 --> 0:04:37.640 kinds of implementations of minds, and I think a self 0:04:37.880 --> 0:04:41.240 is a useful way to think about selves is as 0:04:41.480 --> 0:04:46.000 a system that has goals of some particular size. It's 0:04:46.160 --> 0:04:48.640 what the dark hop Statu would call a strange loop. 0:04:48.680 --> 0:04:51.839 It's an observer of the outside world, but it's also 0:04:51.880 --> 0:04:55.040 an observer of itself. It's something that can loop back 0:04:55.080 --> 0:04:58.520 in and interpret what its own memories mean and what 0:04:58.880 --> 0:05:01.000 it is doing, and make decision and going forward. Those 0:05:01.000 --> 0:05:03.279 are the components of being a self. And so do 0:05:03.560 --> 0:05:07.520 selves change? Yeah? Part of the paradox of being a 0:05:07.560 --> 0:05:10.000 self is that you have to change in order to 0:05:10.000 --> 0:05:12.039 stay alive. You have to change in order to persist. 0:05:12.279 --> 0:05:14.279 Every time you learn something, every time you make a 0:05:14.320 --> 0:05:17.320 decision that then feeds back and alters the inputs that 0:05:17.360 --> 0:05:20.000 you receive and thus changes your own cognitive system and 0:05:20.040 --> 0:05:23.120 your own behavior going forward. You've now changed. I you know, 0:05:23.400 --> 0:05:27.200 we selves face this this paradox that in order to 0:05:27.200 --> 0:05:30.960 to to to persist, they must change and transform over time. 0:05:31.320 --> 0:05:34.320 Why do you suppose we have the illusion of ourselves 0:05:34.360 --> 0:05:37.640 when we think about you know yourself and your life, 0:05:37.800 --> 0:05:40.320 you have the illusion that it's unchanging. 0:05:40.800 --> 0:05:43.440 What do you suppose that's about. Yeah, I think that, 0:05:43.760 --> 0:05:48.320 Uh well, let's let's just think about very primitive, basic, fundamental, 0:05:48.400 --> 0:05:51.120 basic agents at the very beginning of of of of 0:05:51.160 --> 0:05:53.839 that of that spectrum. If if you're any kind of 0:05:53.839 --> 0:05:56.279 a system that's going to survive in the real world, 0:05:56.560 --> 0:05:58.279 what you can't afford to do is to be a 0:05:58.360 --> 0:06:01.599 kind of placium demon where where you're tracking the micro 0:06:01.640 --> 0:06:05.080 states of you know, every single stimulus, every molecular impact 0:06:05.160 --> 0:06:07.120 on your membrane or whatever you can. You can't afford 0:06:07.160 --> 0:06:09.160 to track all that. If you try to do that, 0:06:09.400 --> 0:06:11.480 you will run out of time, You'll run out of energy, 0:06:11.480 --> 0:06:13.159 You'll be eaten and dead in no time. And so 0:06:13.960 --> 0:06:17.360 I think that what we have in at least at 0:06:17.400 --> 0:06:21.520 least in biological evolution, is an incredible pressure to coarse grain, 0:06:21.800 --> 0:06:23.800 that is, to look out into the world and to 0:06:24.000 --> 0:06:30.160 group inputs and sensations and stimuli and observations that you make, 0:06:30.400 --> 0:06:33.000 to group them into categories, and to try to understand 0:06:33.000 --> 0:06:36.279 your world by reducing its complexity, and by trying to 0:06:37.120 --> 0:06:39.640 make models of the world that have these large things 0:06:39.680 --> 0:06:42.400 in it that are themselves agents. You know that do things, 0:06:42.440 --> 0:06:44.840 and that you can then make decisions about what they 0:06:44.839 --> 0:06:46.600 do and you don't have to pay attention to all 0:06:46.600 --> 0:06:49.360 the micro states. If you're going to do that, you 0:06:49.440 --> 0:06:54.000 fundamentally need a way to make models of persistent things. 0:06:54.320 --> 0:06:57.560 You have to have some way of saying that this 0:06:57.680 --> 0:06:59.800 is an object I need to stay away from. More conversely, 0:06:59.800 --> 0:07:01.800 this something I need to approach, or this is something 0:07:01.839 --> 0:07:05.400 I need to recognize meanwhile, of course, so let's just 0:07:05.400 --> 0:07:07.120 look at vision, you know. For example, So you have 0:07:07.160 --> 0:07:11.680 these these pixels and impacting onto your retina, these these 0:07:11.880 --> 0:07:14.680 visual stimuli, and depending on how you're looking at something, 0:07:14.720 --> 0:07:17.560 the details will be completely different, the lighting will be different, 0:07:17.600 --> 0:07:20.440 the motion, the way that it's angled. But your job 0:07:20.600 --> 0:07:25.360 as a successful cognitive system is to recognize as well 0:07:25.400 --> 0:07:27.640 as possible that this is always the same thing, right 0:07:27.680 --> 0:07:30.360 that you can, despite the fact that it's now just shifted, 0:07:30.400 --> 0:07:32.520 tinted green or whatever it is, that that hey, wait 0:07:32.520 --> 0:07:34.520 a minute, I recognize it. I know what this is. 0:07:34.840 --> 0:07:37.800 And so I think from the very beginning to be 0:07:37.880 --> 0:07:42.000 a successful being that can survive in a world war, 0:07:42.080 --> 0:07:45.400 time and energy are extremely precious. These are precious resources. 0:07:45.600 --> 0:07:48.280 You have to get good at noticing and in fact 0:07:48.360 --> 0:07:51.440 magnifying invariance. You know, things that stay the same. And 0:07:51.480 --> 0:07:53.400 then I think you turn that on yourself and you say, 0:07:53.400 --> 0:07:55.360 wait a minute, I am also an agent that does things. 0:07:55.400 --> 0:07:58.040 And I don't mean consciously, and we don't do this consciously. 0:07:58.200 --> 0:08:01.720 But every system, every living system, I think, has an 0:08:01.760 --> 0:08:04.760 internal self model, and part of that model is to 0:08:04.800 --> 0:08:08.160 recognize you as a persistent thing even though things change. 0:08:08.440 --> 0:08:10.680 And then and then you know, those those of us 0:08:10.680 --> 0:08:13.520 that have self consciousness then have a have a story 0:08:13.560 --> 0:08:16.760 that we tell ourselves about being a persistent thing as 0:08:16.760 --> 0:08:19.760 opposed to I think what we much more close to 0:08:19.800 --> 0:08:22.000 what we really are, which is a flowing process. 0:08:23.600 --> 0:08:26.160 So so we all walk through life and we have 0:08:26.280 --> 0:08:29.680 our notion of self, and we observe the selves in 0:08:29.840 --> 0:08:32.439 others that we love, and we try to figure out 0:08:32.440 --> 0:08:35.520 what this is all about. You have a very cool approach, 0:08:35.559 --> 0:08:40.080 which is that you think about diverse intelligences. Tell us 0:08:40.120 --> 0:08:41.200 what that term means. 0:08:42.000 --> 0:08:45.000 Yeah, diverse intelligence is the name of a of a 0:08:45.160 --> 0:08:47.160 of a feel that on the one hand is the 0:08:47.200 --> 0:08:50.120 emerging now and it's a very exciting kind of feel. 0:08:50.360 --> 0:08:52.560 On the other hand, this is a collection of ideas 0:08:52.600 --> 0:08:54.560 that have really been around for a very long time. 0:08:54.840 --> 0:08:58.640 And it goes back to the efforts of trying to 0:08:58.720 --> 0:09:03.400 understand what what do we actually mean by intelligence, by minds, 0:09:03.600 --> 0:09:07.679 by cognition, by all all these kinds of all these 0:09:07.720 --> 0:09:10.800 kinds of terms. And we have familiar examples of them 0:09:10.800 --> 0:09:13.920 in brainy animals and and and you know, in ourselves. 0:09:14.200 --> 0:09:17.280 But actually the key thing that that we need to 0:09:17.320 --> 0:09:20.280 understand is that all of us, at one point, both 0:09:20.400 --> 0:09:23.080 on an evolutionary scale and on a developmental scale, we 0:09:23.080 --> 0:09:25.520 were all a single sell once. You know, at one 0:09:25.520 --> 0:09:28.280 point we were a little blob of quies and cytoplasm 0:09:28.320 --> 0:09:30.800 and unfertilized noocyte. And we would look at that and 0:09:30.840 --> 0:09:33.720 we would say, well, there's there's a little a little 0:09:33.720 --> 0:09:36.880 blob of chemicals that obeys physics and chemistry. And then 0:09:36.920 --> 0:09:38.840 at some point we have something that we would say 0:09:38.840 --> 0:09:41.400 has an inner perspective, It has a mind, it has preferences, 0:09:41.400 --> 0:09:43.920 it has goals, it has memories, and and you know 0:09:44.040 --> 0:09:46.719 it and it has a full blown self conscious kind 0:09:46.720 --> 0:09:49.520 of meta cognition and so on. Now, how did we 0:09:49.559 --> 0:09:51.560 get there? Right? Because we started out you know, this 0:09:51.679 --> 0:09:53.920 this amazing journey from physics to mind. You know, you 0:09:54.000 --> 0:09:56.120 start off with with a little blob of chemistry, and 0:09:56.160 --> 0:10:00.240 then eventually here we are, so diverse. Intelligence is is 0:10:00.480 --> 0:10:04.280 I think, is predicated on the on the idea that 0:10:04.800 --> 0:10:07.720 we need to understand how it is that minds are 0:10:07.760 --> 0:10:10.720 embodied in the physical world, and that by understanding that 0:10:10.760 --> 0:10:13.360 we emerge slowly and gradually, this is we are not. 0:10:13.640 --> 0:10:15.920 You know, there's no there's no magical category called human 0:10:15.960 --> 0:10:18.920 that suddenly snaps into existence at some particular moment of 0:10:18.960 --> 0:10:22.840 embryonic development. But this is a slow, gradual process. It's 0:10:22.880 --> 0:10:25.800 it's an effort to understand what those processes are that 0:10:25.920 --> 0:10:29.400 give rise to collective intelligence. And by the way, all intelligence, 0:10:29.440 --> 0:10:31.920 I think is collective intelligence because all of us are 0:10:31.920 --> 0:10:33.880 made of parts, and we have to understand how the 0:10:33.880 --> 0:10:36.760 competencies of those parts give rise to whatever the larger 0:10:36.800 --> 0:10:39.800 scale system is capable of doing. And as soon as 0:10:39.840 --> 0:10:41.520 as soon as you frame the problem that way, to 0:10:41.640 --> 0:10:44.880 understand what is it that the components are doing to 0:10:44.960 --> 0:10:48.000 scale that cognition from the competencies of a single cell 0:10:48.040 --> 0:10:50.480 to that of an animal or a human, then then 0:10:50.480 --> 0:10:52.680 immediately you start to ask yourself, okay, so not only 0:10:52.760 --> 0:10:56.640 is there a history going backwards where we see increasingly 0:10:56.720 --> 0:11:00.440 simpler forms and ask what is their intelligence like? But 0:11:00.480 --> 0:11:04.520 then we can really try to shed some of the 0:11:04.960 --> 0:11:08.200 limitations that we're given to us by our evolutionary firmware 0:11:08.520 --> 0:11:10.800 that make it easy for us to recognize, you know, 0:11:11.040 --> 0:11:16.559 large intelligence, large large, medium sized objects moving at medium 0:11:17.520 --> 0:11:19.760 speeds through three dimensional space, and we so, okay, that's 0:11:19.800 --> 0:11:22.480 an intelligent animal doing whatever it's doing. But to learn 0:11:22.520 --> 0:11:26.040 to recognize intelligence and unfamiliar guises, you know, what other 0:11:26.160 --> 0:11:28.720 spaces can intelligence be operating and what what else can 0:11:28.720 --> 0:11:32.200 it be made of? What other processes give rise to 0:11:32.360 --> 0:11:35.439 the scale up of intelligence. That's diverse intelligence. It's the 0:11:36.040 --> 0:11:39.400 broad attempt to understand intelligence as it can be. What's 0:11:39.440 --> 0:11:40.160 an example of that? 0:11:40.280 --> 0:11:44.720 For example, ants in a colony understanding that as an 0:11:44.760 --> 0:11:48.319 intelligence system where they're laying down memories in particular ways. 0:11:48.360 --> 0:11:49.920 It's very different than the way we think about the 0:11:49.960 --> 0:11:51.560 human brain. Is that an example? 0:11:52.400 --> 0:11:54.319 Yes, I'll give you a few examples and get the 0:11:54.400 --> 0:11:59.680 sort of progressively progressively weirder with them. Traditional collective intelligence 0:11:59.720 --> 0:12:03.280 is like ants and beehives and termite colonies and blocks 0:12:03.280 --> 0:12:07.280 of birds and things like that. Those are typical collective intelligence. 0:12:07.559 --> 0:12:10.520 They are what our Carsole calls liquid brains because they're 0:12:10.559 --> 0:12:13.280 moving around relative to each other. Our neurons tend to 0:12:13.320 --> 0:12:16.400 stand still relative to each other. And yet we too 0:12:16.440 --> 0:12:18.840 are a collective intelligence, right, We are a pile of 0:12:18.880 --> 0:12:21.839 neurons and other cells that work together to give rise 0:12:21.880 --> 0:12:25.120 to an emergent being with memories and preferences and goals 0:12:25.120 --> 0:12:27.599 that don't belong to any of the individual cells. So, 0:12:28.040 --> 0:12:29.800 no matter what you are, whether you're a collection of 0:12:29.800 --> 0:12:32.360 neurons or your collection of birds or of termites or 0:12:32.360 --> 0:12:34.800 anything like that, what you need is a kind of 0:12:34.800 --> 0:12:38.240 cognitive glue. You need these policies that are implemented by 0:12:38.280 --> 0:12:40.760 the pieces that allow the collective to be more than 0:12:40.880 --> 0:12:43.560 the sum of its parts in relevant ways. And so 0:12:43.600 --> 0:12:48.679 if we define intelligence as problem solving competencies in some space, 0:12:48.880 --> 0:12:51.120 so the ability to and so this is kind of 0:12:51.120 --> 0:12:54.560 William James's definition of intelligence, same goal by different meats, right, 0:12:54.600 --> 0:12:57.240 the ability to navigate some space to reach your objective 0:12:57.640 --> 0:13:02.000 and do so despite novel despite perturbations, and and so on. 0:13:02.440 --> 0:13:06.319 Then we can get we can get progressively more inclusive 0:13:06.320 --> 0:13:08.800 with this. We can say, okay, so we have we 0:13:08.840 --> 0:13:13.240 have animals that solve problems. We have unfamiliar things like 0:13:13.280 --> 0:13:16.040 slime molds, right, so so so there are people including us, 0:13:16.040 --> 0:13:20.080 who study slime mold behavior, in slime mold learning and 0:13:20.120 --> 0:13:22.960 things like that. So slime molds are a very unusual organism. 0:13:23.040 --> 0:13:25.040 It's a single cell. It can be quite large, and 0:13:25.120 --> 0:13:27.960 yet it can it can solve all kinds of problems. 0:13:28.040 --> 0:13:32.280 People have studied memory and learning in in bacteria, in 0:13:32.600 --> 0:13:36.000 unicellular organisms and plants, and and then you can get 0:13:36.040 --> 0:13:39.520 then you can get you know, even even more sort 0:13:39.559 --> 0:13:41.400 of broad with this, and you can ask, well, what 0:13:41.480 --> 0:13:44.160 happens beyond three dimensional space? For example, the cells that 0:13:44.200 --> 0:13:46.360 make up our bodies and our and our tissues and 0:13:46.400 --> 0:13:50.600 our organs, they operate in physiological state space, so the 0:13:50.640 --> 0:13:54.080 space of all possible physiological conditions. They operate in gene 0:13:54.120 --> 0:13:59.360 expression space. Embryonic and regenerative cells operate in anatomical space. 0:13:59.480 --> 0:14:01.800 They have to navigate from whatever shape they have now 0:14:01.880 --> 0:14:05.360 to some kind of complex organ structure. And in all 0:14:05.360 --> 0:14:07.679 of these spaces you can see if you look for 0:14:07.720 --> 0:14:11.000 them using the techniques of behavioral science, you can find 0:14:11.000 --> 0:14:13.439 them solving problems. You can find them getting to their 0:14:13.440 --> 0:14:16.640 goal in very creative ways despite various problems So that's 0:14:16.720 --> 0:14:19.040 kind of the range of the diverse intelligences. And then 0:14:19.040 --> 0:14:22.400 beyond that you have embodied robotics, and you have maybe 0:14:22.400 --> 0:14:27.480 software intelligent agents, and maybe someday exobiological beings and so on. 0:14:28.760 --> 0:14:32.280 Okay, so the challenge is that everything's changing all the 0:14:32.360 --> 0:14:36.640 time in our brains and all these other biological systems 0:14:36.640 --> 0:14:40.120 and colonies whatever, things are changing all the time. Cells 0:14:40.160 --> 0:14:45.840 are dying or aging or getting mutations. And yet somehow, 0:14:45.920 --> 0:14:50.480 when it comes to memory of oneself, we tend to 0:14:50.520 --> 0:14:53.120 have some sort of memory. It's not perfect, it only 0:14:53.160 --> 0:14:56.120 has a few details, things like that, but we retain 0:14:56.240 --> 0:15:00.720 this memory through time. Now, what is your framework for 0:15:00.840 --> 0:15:01.640 thinking about that? 0:15:02.600 --> 0:15:06.400 Yeah, I'm not sure we retain memory through time as 0:15:06.440 --> 0:15:10.440 much as we reconstruct memory through time. And of course 0:15:10.600 --> 0:15:12.960 there's a lot of human neuroscience being done about this. 0:15:12.960 --> 0:15:14.280 But I want to take a step back and give 0:15:14.320 --> 0:15:17.480 you another example that kind of drives my thinking on this. 0:15:17.720 --> 0:15:20.880 Consider what happens from a butterfly to a caterpillar. So 0:15:21.040 --> 0:15:23.520 in the caterpillar, you have the soft bodied creature that 0:15:23.560 --> 0:15:26.160 lives in basically a two dimensional world where it crawls 0:15:26.160 --> 0:15:29.600 around and eats leaves. Now, one thing that it has 0:15:29.640 --> 0:15:31.280 to do is that has to turn into a butterfly. 0:15:31.440 --> 0:15:36.240 During that process of metamorphosis, its brain is massively rebuilt. 0:15:36.320 --> 0:15:38.920 Many of the cells die, all the connections are broken, 0:15:38.960 --> 0:15:42.080 it's completely refactored, and you get a new brain suitable 0:15:42.080 --> 0:15:45.160 for driving a now hard bodied kind of vehicle, completely 0:15:45.160 --> 0:15:48.640 different controller you need for that. And now it flies 0:15:48.760 --> 0:15:51.440 and it lives in the three dimensional world and so on. 0:15:51.960 --> 0:15:55.800 Now it turns out that if you train the caterpillar 0:15:57.040 --> 0:16:00.000 the let's say you train it to associate a specific 0:16:00.080 --> 0:16:02.640 color to a specific leaf that it wants to eat, 0:16:02.840 --> 0:16:05.840 what you will find is that the butterfly, despite having 0:16:05.880 --> 0:16:09.240 a totally refactored brain, will remember that information. And for 0:16:09.360 --> 0:16:11.520 years I thought the amazing thing here was how do 0:16:11.560 --> 0:16:14.520 you hold on to information when your information medium is 0:16:14.520 --> 0:16:17.040 being totally ripped apart and refactored. Right, we don't, you know, 0:16:17.040 --> 0:16:19.520 our computer technology doesn't doesn't like that. We don't have 0:16:19.560 --> 0:16:23.360 anything that has that that robustness to it. But if 0:16:23.360 --> 0:16:26.680 you think about it more, what turns out is that 0:16:27.120 --> 0:16:29.640 the amazing thing is not holding onto the memory because 0:16:29.640 --> 0:16:33.440 actually the memories, the exact memories of the caterpillar, are 0:16:33.440 --> 0:16:36.080 of no use to the butterfly whatsoever, because it doesn't 0:16:36.080 --> 0:16:38.560 move the same way. It doesn't want the leaves a 0:16:38.640 --> 0:16:42.400 drinks nectar, right, And so just holding onto that memory 0:16:42.520 --> 0:16:44.920 is of no use. What you need is to actually 0:16:45.680 --> 0:16:49.680 transform and remap that memory into a completely novel context 0:16:49.920 --> 0:16:53.360 where it's like not leaves but food for example. Right, 0:16:53.400 --> 0:16:55.240 So you have to generalize a little bit, and you 0:16:55.280 --> 0:16:57.320 have to now remap it on a different on a 0:16:57.320 --> 0:17:01.080 different controller. And you know, numerous people who have done 0:17:01.120 --> 0:17:04.280 experiments in memory transfer, most recently David Landsman and others 0:17:04.480 --> 0:17:07.400 show that that it's it's really wild how you can 0:17:07.440 --> 0:17:10.879 move either tissues or extra extracts let's say, RNA extracts 0:17:10.960 --> 0:17:14.480 or whatever, and introduce them into a new into a 0:17:14.520 --> 0:17:16.959 new creature and and have that creature like take on 0:17:17.040 --> 0:17:18.639 that that initial learning. 0:17:18.960 --> 0:17:20.480 So let me interrupt for a second. So tell us 0:17:20.480 --> 0:17:24.800 about David Landsman's experiments in these sea slugs. Yeah, Well, 0:17:24.840 --> 0:17:30.120 specifically in David's work, what he was doing is training 0:17:30.119 --> 0:17:33.320 these c slugs to a particular to a particular task. 0:17:33.359 --> 0:17:36.160 I mean it was a simple, you know, simple reflex 0:17:36.560 --> 0:17:40.800 and extracting RNA from from the brain. Because the hypothesis 0:17:40.880 --> 0:17:42.840 was that memory is stored in some way in this 0:17:42.960 --> 0:17:46.600 in this medium of RNA, and then he was injecting 0:17:46.600 --> 0:17:48.760 it into a naive animal and showing that they now 0:17:48.880 --> 0:17:51.880 have you know, they show a recall of that information 0:17:52.119 --> 0:17:54.160 I mean to me, I mean, it's it's a fabulous 0:17:54.200 --> 0:17:54.720 body of work. 0:17:55.240 --> 0:17:57.640 To me. One of the most interesting things about it 0:17:57.680 --> 0:18:00.280 is that when you've got this RNA, you don't have 0:18:00.359 --> 0:18:03.520 to go put it exactly in the right neuron where 0:18:03.600 --> 0:18:05.679 it was supposed to be, you know, and that you 0:18:05.840 --> 0:18:09.000 just kind of inject it somewhere in the brain of 0:18:09.080 --> 0:18:12.080 a second of a second sea slug exactly of the 0:18:12.119 --> 0:18:14.560 recipient right of the host seasug, and it somehow gets 0:18:14.560 --> 0:18:17.639 taken out. And that's that's a that's a theme that 0:18:17.640 --> 0:18:20.679 that is like central to all this because what I 0:18:20.680 --> 0:18:22.920 think this is, this is telling us is so now, 0:18:22.960 --> 0:18:25.040 so now let's let's walk into this from from the 0:18:25.040 --> 0:18:28.600 other end. Whatever you are, human or anything else, you 0:18:28.640 --> 0:18:31.080 don't have access to the past. What you have access 0:18:31.119 --> 0:18:33.439 to is the end grams, the memory traces that the 0:18:33.480 --> 0:18:36.240 past has left in your brain or body or wherever 0:18:36.480 --> 0:18:39.000 that were formed by past experience. And so what you 0:18:39.080 --> 0:18:41.399 now have to do is to at any given moment 0:18:41.840 --> 0:18:43.399 and it's you know, these moments. I don't know how 0:18:43.400 --> 0:18:46.080 many milliseconds they are, but but some some number of milliseconds. 0:18:46.320 --> 0:18:50.520 You have to reconstruct that memory to be meaningful to 0:18:50.600 --> 0:18:53.840 you now because you don't know what it used to mean. 0:18:53.880 --> 0:18:55.399 This is you have to do this on the on 0:18:55.480 --> 0:18:57.760 the fly. And I think there's a lot of good 0:18:57.760 --> 0:19:00.280 neuroscience showing how plastic these memories are and how there 0:19:00.400 --> 0:19:03.480 how you know, even even recall of the memories means 0:19:03.480 --> 0:19:07.240 they're getting changed. There's no non destructive read. Accessing these 0:19:07.240 --> 0:19:09.880 memories changes them and and and so so the way 0:19:09.920 --> 0:19:12.680 I think about this is as an architecture in the 0:19:12.720 --> 0:19:14.679 shape of a kind of bow tie. And this is 0:19:14.920 --> 0:19:17.280 for the computer science listeners. You might imagine like a 0:19:17.400 --> 0:19:21.639 like a an auto encoder architecture where there's a funnel 0:19:21.680 --> 0:19:25.280 on the left side which receives the primary experiences, the 0:19:25.359 --> 0:19:28.560 raw data that come in the sense impressions, and that 0:19:28.840 --> 0:19:31.080 the process of learning has to and and and this 0:19:31.160 --> 0:19:36.040 is fundamental to intelligence, is abstracting from individual instances of 0:19:36.080 --> 0:19:38.800 things you experience to a rule, to some kind of 0:19:39.560 --> 0:19:41.480 some kind of pattern, and it's the pattern that you 0:19:41.520 --> 0:19:45.200 remember it, so you compress all those experiences, you throw 0:19:45.200 --> 0:19:47.720 away all the irrelevant details and you form a memory. 0:19:47.760 --> 0:19:50.280 You store that in some sort of enngram. Uh. You know. 0:19:50.560 --> 0:19:53.800 Sometimes people think of this as synaptic modifications. Sometimes other 0:19:53.840 --> 0:19:55.560 people think it's in the RNA or in a cyber 0:19:55.600 --> 0:19:59.000 skeleton wherever. So so you store this but now but 0:19:59.080 --> 0:20:02.160 now here comes here comes the really interesting part. When 0:20:02.200 --> 0:20:04.840 you need to recall this. You have to know, here 0:20:04.880 --> 0:20:06.640 comes the right side of that, on the right side 0:20:06.680 --> 0:20:09.040 of that bow tie. You have to sort of reinflate 0:20:09.080 --> 0:20:13.320 that compressed memory. And because you've lost all kinds of information, 0:20:13.800 --> 0:20:18.679 this process, this, this recall process is creative because you 0:20:18.720 --> 0:20:22.240 don't have all the details that were there, and nor 0:20:22.320 --> 0:20:24.960 can you really be sure at a later time what 0:20:25.000 --> 0:20:27.639 the meaning was to your past self. In other words, 0:20:27.800 --> 0:20:31.200 I also think of memories as literally messages from your 0:20:31.240 --> 0:20:33.439 past self. So I tend to think of memory and 0:20:33.480 --> 0:20:36.479 recall as communication events. And it's just you know, it's 0:20:36.480 --> 0:20:38.440 a communication with a past version of you, but it's 0:20:38.440 --> 0:20:40.880 the same. They sent you a message. It was encrypted 0:20:40.960 --> 0:20:44.960 and compressed in these n grams, and then you try 0:20:44.960 --> 0:20:47.360 to reinflate it. And your goal at any given time 0:20:47.760 --> 0:20:51.200 is not to have any kind of allegiance to what 0:20:51.280 --> 0:20:54.880 the memory meant before. Your goal now is to reinterpret 0:20:54.880 --> 0:20:57.040 it the way the butterfly does in whatever is the 0:20:57.080 --> 0:21:01.760 most optimal adaptive way that makes sense. Now, so this really, 0:21:01.800 --> 0:21:03.480 you know, the first part is kind of algorithmic, which 0:21:03.520 --> 0:21:06.160 is the compression. But now here comes a creative process. 0:21:06.200 --> 0:21:09.000 It's not really deductive. It's a creative process where you 0:21:09.040 --> 0:21:11.920 take that prompt. It's more of a prompt than anything else, 0:21:11.920 --> 0:21:13.880 and you say, okay, what does this mean to me? Now? 0:21:13.880 --> 0:21:17.680 How can I incorporate this into my current constantly evolving 0:21:17.720 --> 0:21:19.600 model of the self, of the outside world and on. 0:21:19.760 --> 0:21:22.000 So it's very much And this goes on, goes on 0:21:22.080 --> 0:21:26.080 all the time. And this is consistent with the plasticity 0:21:26.119 --> 0:21:30.160 of memories. It's consistent with confabulation, which has been seen 0:21:30.200 --> 0:21:32.879 in all kinds of experiments with human subjects, you know, 0:21:33.000 --> 0:21:33.879 split brain and so on. 0:21:33.960 --> 0:21:36.800 Wait, let's take a second to talk about confabulation. So 0:21:37.640 --> 0:21:41.280 this is where brains seem to make something up. You know, 0:21:41.320 --> 0:21:44.240