1 00:00:05,000 --> 00:00:08,080 Speaker 1: Have you ever seen those pictures of blobs on a 2 00:00:08,160 --> 00:00:11,399 Speaker 1: page and it doesn't look like anything to you until 3 00:00:11,440 --> 00:00:14,640 Speaker 1: you're told what it is, and then you suddenly see it. 4 00:00:15,400 --> 00:00:18,320 Speaker 1: Why does that give us a great clue about the 5 00:00:18,360 --> 00:00:23,239 Speaker 1: wiring of the brain. And why are neuroscientists so magnetically 6 00:00:23,239 --> 00:00:27,080 Speaker 1: attracted to those visual illusions that you scroll through on 7 00:00:27,120 --> 00:00:31,040 Speaker 1: social media? What is the deep trick about the way 8 00:00:31,120 --> 00:00:34,080 Speaker 1: your visual system works that you were never taught in school? 9 00:00:34,520 --> 00:00:36,440 Speaker 1: And what does any of this have to do with 10 00:00:36,640 --> 00:00:39,760 Speaker 1: catching a baseball or zooming down the road in New 11 00:00:39,840 --> 00:00:46,640 Speaker 1: York City or the warp drive in Star Trek. Welcome 12 00:00:46,680 --> 00:00:50,360 Speaker 1: to Inner Cosmos with me David Eagleman. I'm a neuroscientist 13 00:00:50,360 --> 00:00:53,480 Speaker 1: and author at Stanford and in these episodes we sail 14 00:00:53,600 --> 00:00:57,240 Speaker 1: deeply into our three pound universe to understand why and 15 00:00:57,320 --> 00:00:59,840 Speaker 1: how our lives look the way they do. 16 00:01:09,600 --> 00:01:09,880 Speaker 2: Today. 17 00:01:09,920 --> 00:01:13,640 Speaker 1: I'm going to start with the notion of visual illusions. 18 00:01:14,200 --> 00:01:18,120 Speaker 1: Elementary school students love these and they stare at them 19 00:01:18,240 --> 00:01:20,200 Speaker 1: for about a minute and then they're on to the 20 00:01:20,240 --> 00:01:23,600 Speaker 1: next thing, because why not. The illusion is just an 21 00:01:23,640 --> 00:01:27,040 Speaker 1: interesting trick. There's nothing further to do about it. It's 22 00:01:27,080 --> 00:01:30,080 Speaker 1: only later when you grow up to be a neuroscientist 23 00:01:30,240 --> 00:01:33,520 Speaker 1: or a fan of a neuroscience podcast, that you might 24 00:01:33,640 --> 00:01:37,360 Speaker 1: even return to one of these illusions to ask, wait, 25 00:01:38,000 --> 00:01:41,280 Speaker 1: why exactly does that happen? 26 00:01:41,440 --> 00:01:42,959 Speaker 2: Does that tell us. 27 00:01:42,840 --> 00:01:48,800 Speaker 1: Something deep and fundamental about the way my consciousness constructs 28 00:01:48,840 --> 00:01:49,880 Speaker 1: the world for me? 29 00:01:50,560 --> 00:01:53,280 Speaker 2: What does it reveal? So? 30 00:01:53,880 --> 00:01:56,840 Speaker 1: Have you ever seen the illusion where you're looking at 31 00:01:57,200 --> 00:02:00,840 Speaker 1: lines like bicycle spokes, and then there's some straight lines 32 00:02:00,920 --> 00:02:03,600 Speaker 1: drawn on top of that, and they don't look straight, 33 00:02:03,720 --> 00:02:07,800 Speaker 1: they look bent. Why does that happen? Seems like it 34 00:02:07,840 --> 00:02:10,280 Speaker 1: shouldn't be hard to answer, but it's actually taken well 35 00:02:10,320 --> 00:02:13,359 Speaker 1: over a century to figure this out, and the answer 36 00:02:13,720 --> 00:02:16,920 Speaker 1: is gonna blow your mind. I promise you that. But 37 00:02:17,720 --> 00:02:20,600 Speaker 1: in order to get us there, I'm gonna start with 38 00:02:20,680 --> 00:02:24,560 Speaker 1: something completely different. I'm gonna start with those pictures that 39 00:02:24,639 --> 00:02:28,040 Speaker 1: look like just a bunch of blobs. Probably you've seen 40 00:02:28,080 --> 00:02:30,880 Speaker 1: one of these before. There's just a bunch of random 41 00:02:30,919 --> 00:02:34,840 Speaker 1: looking splotches of black and white on a page. If 42 00:02:34,880 --> 00:02:40,240 Speaker 1: your brain doesn't have a prior expectation about what it's seeing, 43 00:02:40,600 --> 00:02:44,360 Speaker 1: about what the blobs mean, then you simply see black 44 00:02:44,400 --> 00:02:47,520 Speaker 1: and white blobs and there's no particular meaning to the picture. 45 00:02:47,720 --> 00:02:49,720 Speaker 1: I'm gonna link an example of this on the show 46 00:02:49,760 --> 00:02:52,720 Speaker 1: notes at eagleman dot com slash podcast, and I want 47 00:02:52,760 --> 00:02:55,080 Speaker 1: you to stare at it for a few moments and 48 00:02:55,120 --> 00:02:57,240 Speaker 1: then scroll down to the very bottom of the page 49 00:02:57,360 --> 00:03:01,160 Speaker 1: for the hint. What you'll see is that you can't 50 00:03:01,240 --> 00:03:04,320 Speaker 1: make heads or tails of these blobs. But then I 51 00:03:04,360 --> 00:03:07,200 Speaker 1: only change one thing, and it has nothing to do 52 00:03:07,280 --> 00:03:10,680 Speaker 1: with what's on the screen. I give you a hint, 53 00:03:11,360 --> 00:03:14,520 Speaker 1: and as soon as you have a notion about how 54 00:03:14,560 --> 00:03:18,119 Speaker 1: to interpret what is on your retinas, then you say, 55 00:03:18,200 --> 00:03:21,720 Speaker 1: oh yeah, I see it now. Now the exact same 56 00:03:21,840 --> 00:03:25,480 Speaker 1: blobs that confused you a moment ago make perfect sense. 57 00:03:25,520 --> 00:03:29,079 Speaker 1: But again, nothing changed out there in the world. The 58 00:03:29,120 --> 00:03:33,200 Speaker 1: only thing that changed is something in your neural networks. 59 00:03:35,400 --> 00:03:38,000 Speaker 1: So what's the lesson from this? There has to be 60 00:03:38,080 --> 00:03:43,040 Speaker 1: a match between incoming data and your expectations for you 61 00:03:43,160 --> 00:03:48,480 Speaker 1: to see anything. But wait, what, That's not how vision 62 00:03:48,560 --> 00:03:50,760 Speaker 1: is supposed to work, is it. I mean, after all, 63 00:03:50,760 --> 00:03:54,280 Speaker 1: you look at any basic biology textbook and it will 64 00:03:54,280 --> 00:03:57,400 Speaker 1: tell you that photons hit the retina and the information 65 00:03:57,520 --> 00:04:00,440 Speaker 1: is carried on back to the visual cortex, and then 66 00:04:00,480 --> 00:04:04,200 Speaker 1: you just see what's out there. The visual cortex is 67 00:04:04,320 --> 00:04:07,720 Speaker 1: like a television screen, So what's going on? 68 00:04:08,280 --> 00:04:10,080 Speaker 2: Why can't you see. 69 00:04:09,720 --> 00:04:13,920 Speaker 1: The image in the blobs until you've got the right expectation. 70 00:04:14,480 --> 00:04:15,760 Speaker 2: This ties into. 71 00:04:15,520 --> 00:04:18,000 Speaker 1: A concept that you hear me refer to all the 72 00:04:18,040 --> 00:04:20,599 Speaker 1: time on this podcast, and that is the concept of 73 00:04:20,680 --> 00:04:25,760 Speaker 1: the internal model. Remember that your brain is isolated in 74 00:04:26,160 --> 00:04:31,240 Speaker 1: soundless and lightless solitude inside your skull, and its single 75 00:04:31,360 --> 00:04:37,039 Speaker 1: mission is to construct a loud, colorful mental model of 76 00:04:37,080 --> 00:04:40,880 Speaker 1: the outside world. In other words, it builds an inner 77 00:04:40,880 --> 00:04:46,520 Speaker 1: reality that tries to accurately reflect the outside. The key 78 00:04:46,720 --> 00:04:51,200 Speaker 1: is that you don't see by capturing television pixels from 79 00:04:51,279 --> 00:04:56,640 Speaker 1: the world. Instead, all you ever see is your internal model, 80 00:04:57,160 --> 00:05:01,360 Speaker 1: and your internal model only perceives some thing when its 81 00:05:01,360 --> 00:05:08,200 Speaker 1: expectations are sufficiently supported by the sensory data coming in. Now, 82 00:05:08,480 --> 00:05:12,200 Speaker 1: this isn't really a widely known idea. I think you'll 83 00:05:12,200 --> 00:05:15,240 Speaker 1: find if you ask people about it on airplanes, as 84 00:05:15,360 --> 00:05:17,960 Speaker 1: I often do. This isn't really the way that most 85 00:05:18,000 --> 00:05:21,280 Speaker 1: people are used to thinking about the brain. So it's 86 00:05:21,279 --> 00:05:24,920 Speaker 1: a bit surprising that the basic conceptualization of this idea 87 00:05:25,000 --> 00:05:28,520 Speaker 1: is almost seventy years old. One of the earliest examples 88 00:05:28,600 --> 00:05:31,279 Speaker 1: of this framework that I know of came from the 89 00:05:31,320 --> 00:05:35,640 Speaker 1: neuroscientist Donald McKay, who in nineteen fifty six said, Look, 90 00:05:35,839 --> 00:05:39,120 Speaker 1: the job of the visual cortex is to construct an 91 00:05:39,160 --> 00:05:43,159 Speaker 1: internal model, and it's always trying to anticipate the data 92 00:05:43,560 --> 00:05:46,479 Speaker 1: coming up from the retina. Now, just as a reminder, 93 00:05:46,520 --> 00:05:48,000 Speaker 1: the retina is the part of the back of your 94 00:05:48,040 --> 00:05:51,080 Speaker 1: eye that captures light, and the visual cortex is all 95 00:05:51,080 --> 00:05:52,560 Speaker 1: the way at the back of your head, on the 96 00:05:52,560 --> 00:05:55,839 Speaker 1: far side of the brain. But here's the surprise. The 97 00:05:55,920 --> 00:06:00,920 Speaker 1: information doesn't just shoot from the retina to the visual cortex. Instead, 98 00:06:01,320 --> 00:06:05,880 Speaker 1: there's a train station in between, a structure called the thalamus. 99 00:06:05,960 --> 00:06:09,279 Speaker 1: The thalamus sits right in the middle. So information doesn't 100 00:06:09,320 --> 00:06:11,599 Speaker 1: go straight from the eye to the visual cortex, but 101 00:06:11,680 --> 00:06:15,280 Speaker 1: instead it makes a stop and changes trains halfway at 102 00:06:15,279 --> 00:06:15,960 Speaker 1: the thalamus. 103 00:06:16,400 --> 00:06:18,920 Speaker 2: Okay, well that's weird. Why is there the setup? 104 00:06:19,600 --> 00:06:22,640 Speaker 1: Well, to understand this, we need to understand that the 105 00:06:22,760 --> 00:06:27,800 Speaker 1: model of vision in introductory textbooks isn't just misleading, it's 106 00:06:27,920 --> 00:06:32,600 Speaker 1: dead wrong. The brain isn't built on straight lines, it's 107 00:06:32,600 --> 00:06:37,040 Speaker 1: built with loops. So what McKay suggested is that the 108 00:06:37,080 --> 00:06:40,599 Speaker 1: retina sends its data to the thalamus. In other words, 109 00:06:40,600 --> 00:06:42,719 Speaker 1: what the eye is capturing about the world out there, 110 00:06:43,160 --> 00:06:47,800 Speaker 1: and the cortex sends its predictions to the thalamus what 111 00:06:47,839 --> 00:06:51,039 Speaker 1: the cortex is expecting to see next, and all that 112 00:06:51,120 --> 00:06:54,000 Speaker 1: ever comes out of the thalamus back to the cortex 113 00:06:54,720 --> 00:06:58,839 Speaker 1: is the difference, the difference between what you expected and 114 00:06:58,880 --> 00:07:02,440 Speaker 1: what you got. In other words, the information that goes 115 00:07:02,920 --> 00:07:06,080 Speaker 1: from the thalamis to the visual cortex is just that 116 00:07:06,200 --> 00:07:10,920 Speaker 1: little bit which was unanticipated, the difference between what's out 117 00:07:10,920 --> 00:07:14,440 Speaker 1: there and what was already expected. The thalamus sends to 118 00:07:14,520 --> 00:07:19,040 Speaker 1: the cortex only that difference signal, because that's the only 119 00:07:19,080 --> 00:07:23,000 Speaker 1: part that wasn't predicted away. And then, by the way, 120 00:07:23,040 --> 00:07:27,640 Speaker 1: this unpredicted information adjusts the internal model so there will 121 00:07:27,680 --> 00:07:30,720 Speaker 1: be less of a mismatch in the future. That way, 122 00:07:30,720 --> 00:07:34,560 Speaker 1: the brain refines its model of the world by paying 123 00:07:34,600 --> 00:07:39,360 Speaker 1: attention to its mistakes. Okay, so the idea here is 124 00:07:39,360 --> 00:07:42,600 Speaker 1: that the brain is always trying to anticipate what it's 125 00:07:42,640 --> 00:07:45,520 Speaker 1: seeing out there, and McKay pointed out that this is 126 00:07:45,600 --> 00:07:50,560 Speaker 1: consistent with the anatomical fact that there are ten times 127 00:07:50,640 --> 00:07:55,240 Speaker 1: as many fibers projecting from the visual cortex back to 128 00:07:55,280 --> 00:07:59,440 Speaker 1: the thalamis as there are going from thalamis to visual cortex, 129 00:07:59,720 --> 00:08:02,920 Speaker 1: which no one would have guessed. But that's just what 130 00:08:03,000 --> 00:08:07,480 Speaker 1: you'd expect if detailed predictions are going from the cortex 131 00:08:07,600 --> 00:08:11,320 Speaker 1: to the thalamis, and the little signal from thalamis back 132 00:08:11,360 --> 00:08:15,280 Speaker 1: to cortex is just carrying the difference. 133 00:08:15,160 --> 00:08:17,360 Speaker 2: Between what was expected and what was seen. 134 00:08:18,080 --> 00:08:20,600 Speaker 1: Okay, So why am I telling you this level of 135 00:08:20,680 --> 00:08:24,840 Speaker 1: detail because it exposes a giant idea. It means that 136 00:08:24,920 --> 00:08:29,480 Speaker 1: what you perceive about the world emerges from an active 137 00:08:29,600 --> 00:08:35,439 Speaker 1: comparison of sensory data with your internal predictions. Again, think 138 00:08:35,480 --> 00:08:38,560 Speaker 1: about those blobs. If you don't have a prediction of 139 00:08:38,600 --> 00:08:42,000 Speaker 1: what you're seeing out there, there's really nothing there. As 140 00:08:42,080 --> 00:08:46,280 Speaker 1: soon as you have a close enough expectation because you've 141 00:08:46,320 --> 00:08:49,320 Speaker 1: been given a hint, then that lights up a forest 142 00:08:49,360 --> 00:08:53,040 Speaker 1: fire in your brain and you see the thing because 143 00:08:53,080 --> 00:08:54,000 Speaker 1: there's a match. 144 00:08:54,120 --> 00:08:56,280 Speaker 2: Now. So what this. 145 00:08:56,280 --> 00:08:59,720 Speaker 1: Means is that the brain is always trying to predict 146 00:09:00,240 --> 00:09:04,160 Speaker 1: everything that is coming or expected. And here's one way 147 00:09:04,160 --> 00:09:07,480 Speaker 1: that the brain helps itself along. Whenever it sends out 148 00:09:07,520 --> 00:09:10,720 Speaker 1: a signal to your body, like move your head or 149 00:09:11,000 --> 00:09:15,960 Speaker 1: move your arm, it also sends copies of that command 150 00:09:16,160 --> 00:09:20,239 Speaker 1: internally all around the brain. These are called efference copies. 151 00:09:20,520 --> 00:09:24,240 Speaker 1: So now your movement isn't just happening in the outside 152 00:09:24,320 --> 00:09:26,720 Speaker 1: world and then you react to it, but there's also 153 00:09:26,760 --> 00:09:31,720 Speaker 1: a simulation of that movement happening inside your internal model, 154 00:09:32,160 --> 00:09:35,840 Speaker 1: so that you can predict the outcome of that action. 155 00:09:36,520 --> 00:09:39,760 Speaker 1: And this, by the way, is the reason you can't 156 00:09:39,960 --> 00:09:45,120 Speaker 1: tickle yourself. Other people can tickle you because they're tickling 157 00:09:45,200 --> 00:09:48,760 Speaker 1: maneuvers are not predictable to you. But you can't tickle 158 00:09:48,840 --> 00:09:53,120 Speaker 1: you because your brain moves your fingers into the tickle 159 00:09:53,200 --> 00:09:58,960 Speaker 1: position and it already expects the resulting sensations, that already 160 00:09:59,000 --> 00:10:03,160 Speaker 1: knows what's come. Now, by the way, if you'd really 161 00:10:03,240 --> 00:10:05,439 Speaker 1: like to tickle yourself, there is a way to do it, 162 00:10:05,880 --> 00:10:11,440 Speaker 1: and this just involves taking predictability away from your own actions. 163 00:10:12,000 --> 00:10:14,800 Speaker 1: So what you do is you control the position of 164 00:10:14,840 --> 00:10:21,000 Speaker 1: a feather with a joystick that inserts a random time delay, 165 00:10:21,080 --> 00:10:24,320 Speaker 1: so when you move the joystick, at least a second 166 00:10:24,440 --> 00:10:29,000 Speaker 1: passes before the feather moves accordingly, so that takes away 167 00:10:29,360 --> 00:10:33,560 Speaker 1: the predictability and now you can self tickle. By the way, 168 00:10:33,720 --> 00:10:37,280 Speaker 1: related to this, I described in episode forty four how 169 00:10:37,320 --> 00:10:42,440 Speaker 1: people with schizophrenia can tickle themselves, and this is because 170 00:10:42,480 --> 00:10:46,720 Speaker 1: of a problem with their internal timing that doesn't allow 171 00:10:46,840 --> 00:11:08,720 Speaker 1: their motor actions and resulting sensations to be correctly sequenced. Okay, 172 00:11:08,760 --> 00:11:12,760 Speaker 1: so back to this issue about having a brain that's 173 00:11:12,800 --> 00:11:16,720 Speaker 1: not just moving signals down a one way assembly line, 174 00:11:16,720 --> 00:11:20,240 Speaker 1: but instead has all these internal loops so that it 175 00:11:20,320 --> 00:11:24,760 Speaker 1: can always be feeding its internal model and guessing what's 176 00:11:24,840 --> 00:11:28,960 Speaker 1: going to happen next. What is the advantage of this, Well, 177 00:11:29,679 --> 00:11:35,160 Speaker 1: it allows us to transcend stimulus response behavior. In other words, 178 00:11:35,440 --> 00:11:38,280 Speaker 1: we don't have to just observe the world and then 179 00:11:38,520 --> 00:11:43,280 Speaker 1: react to it. Instead, a brain with an internal model 180 00:11:43,800 --> 00:11:49,120 Speaker 1: gives us the ability to make predictions ahead of actual 181 00:11:49,240 --> 00:11:52,440 Speaker 1: sensory input, like predicting what your fingers will feel like 182 00:11:52,559 --> 00:11:57,840 Speaker 1: in your underarm. So our brains build these predictive internal 183 00:11:57,920 --> 00:12:01,400 Speaker 1: models that tell us how things are likely to go 184 00:12:01,880 --> 00:12:04,880 Speaker 1: in the world. And this way our brains don't work 185 00:12:05,160 --> 00:12:09,199 Speaker 1: solely from the latest sensory data, but instead they're always 186 00:12:09,440 --> 00:12:13,400 Speaker 1: guessing ahead to the next moment. Now, why do we 187 00:12:13,480 --> 00:12:18,800 Speaker 1: need a complicated brain like this because our perception is 188 00:12:19,040 --> 00:12:24,400 Speaker 1: massively delayed from reality. Why is it delayed because signals 189 00:12:24,440 --> 00:12:27,600 Speaker 1: from the world, like something you see or a touch 190 00:12:27,640 --> 00:12:32,319 Speaker 1: on your toe. These signals have to travel along nerve cells, 191 00:12:32,840 --> 00:12:36,560 Speaker 1: and they move about a meter per second in the cortex, 192 00:12:36,920 --> 00:12:41,839 Speaker 1: which is, by the way, about three hundred million times 193 00:12:42,000 --> 00:12:47,480 Speaker 1: slower than electricity moving through your laptop. We are giant 194 00:12:47,600 --> 00:12:51,640 Speaker 1: systems of cells, and it takes time for impulses and 195 00:12:51,720 --> 00:12:55,440 Speaker 1: cells to travel around. Yes, they use electricity, but it's 196 00:12:55,480 --> 00:12:59,360 Speaker 1: not like a signal propagating along a wire. Instead, with 197 00:12:59,400 --> 00:13:04,000 Speaker 1: a cell, you've got these long extensions called axons, and 198 00:13:04,080 --> 00:13:07,640 Speaker 1: the signals travel by causing little channels to open in 199 00:13:07,640 --> 00:13:11,000 Speaker 1: the membrane, which allows little charged particles to flow through 200 00:13:11,360 --> 00:13:15,480 Speaker 1: and change the voltage locally, and that propagates down the axon. 201 00:13:15,920 --> 00:13:18,480 Speaker 1: So this is a very cool way that mother nature 202 00:13:18,520 --> 00:13:21,960 Speaker 1: discovered how to run a signal down a cell. But 203 00:13:22,040 --> 00:13:25,480 Speaker 1: it ain't fast, and the consequence is that it just 204 00:13:25,520 --> 00:13:29,360 Speaker 1: takes a long time for signals to propagate through the 205 00:13:29,400 --> 00:13:34,840 Speaker 1: system and eventually come together and settle into a coherent pattern. 206 00:13:35,520 --> 00:13:39,600 Speaker 1: So by the time you become consciously aware of something 207 00:13:39,640 --> 00:13:44,120 Speaker 1: in the outside world, the event has already happened a 208 00:13:44,120 --> 00:13:48,840 Speaker 1: while ago. We live in the past. For example, clap 209 00:13:48,880 --> 00:13:51,760 Speaker 1: your hands in front of you. By the time you 210 00:13:51,920 --> 00:13:55,120 Speaker 1: see and feel and hear that it's already happened a 211 00:13:55,160 --> 00:14:00,439 Speaker 1: tiny little while ago. Whatever conscious movie you're seeing right now, now, 212 00:14:01,120 --> 00:14:02,640 Speaker 1: that world is already gone. 213 00:14:03,600 --> 00:14:03,760 Speaker 2: Now. 214 00:14:03,800 --> 00:14:07,760 Speaker 1: We don't often think about this, but this delay from reality, 215 00:14:07,800 --> 00:14:10,040 Speaker 1: the fact that we're living in the past, is a 216 00:14:10,080 --> 00:14:14,360 Speaker 1: major problem because you need to operate in the present, 217 00:14:14,920 --> 00:14:19,480 Speaker 1: but your brain is always working with old news. All 218 00:14:19,640 --> 00:14:23,120 Speaker 1: your sensory inputs like vision and hearing and touch, these 219 00:14:23,280 --> 00:14:27,000 Speaker 1: take time to travel to the brain to get processed, 220 00:14:27,000 --> 00:14:30,360 Speaker 1: and finally the brain croaks out a response. And even 221 00:14:30,400 --> 00:14:33,760 Speaker 1: though this delay is less than a second, that's plenty 222 00:14:33,800 --> 00:14:37,280 Speaker 1: of time to create issues. So just think about trying 223 00:14:37,320 --> 00:14:40,880 Speaker 1: to catch a baseball that someone throws to you. If 224 00:14:40,920 --> 00:14:45,160 Speaker 1: you were merely an assembly line device, you couldn't do it. 225 00:14:45,280 --> 00:14:48,800 Speaker 1: Why because there would be a delay of hundreds of 226 00:14:48,840 --> 00:14:52,440 Speaker 1: milliseconds from the time the light strikes your eyes until 227 00:14:52,480 --> 00:14:54,840 Speaker 1: you could put up your glove in the right spot. 228 00:14:55,400 --> 00:14:58,160 Speaker 1: And the problem is that by the time the image 229 00:14:58,200 --> 00:15:01,240 Speaker 1: of the ball reaches your brain and gets processed, the 230 00:15:01,280 --> 00:15:04,760 Speaker 1: ball has moved. Your hand would always be reaching for 231 00:15:04,840 --> 00:15:07,160 Speaker 1: a place where the ball used to be. 232 00:15:08,240 --> 00:15:09,560 Speaker 2: So how do you catch a baseball? 233 00:15:09,600 --> 00:15:15,360 Speaker 1: It's because of these deeply hardwired internal models. Your internal 234 00:15:15,400 --> 00:15:20,360 Speaker 1: model generates expectations about when and where the ball's going 235 00:15:20,440 --> 00:15:24,080 Speaker 1: to hit, given momentum and gravity and so on. Your 236 00:15:24,080 --> 00:15:29,560 Speaker 1: brain is not just passively processing information. It's predicting. It's 237 00:15:29,560 --> 00:15:31,320 Speaker 1: not reactive, it's. 238 00:15:31,200 --> 00:15:33,040 Speaker 2: Constantly guessing ahead. 239 00:15:33,600 --> 00:15:36,360 Speaker 1: It predicts where the ball is going to be based 240 00:15:36,400 --> 00:15:39,800 Speaker 1: on clues about its trajectory and speed, and that's what 241 00:15:39,920 --> 00:15:42,800 Speaker 1: allows you to catch it. By the way, as a 242 00:15:42,840 --> 00:15:46,480 Speaker 1: side note, these predictive internal models you have are trained 243 00:15:46,560 --> 00:15:52,640 Speaker 1: up by lifelong exposure in your normal experience. If your 244 00:15:52,920 --> 00:15:57,600 Speaker 1: great grandkids grow up on Mars, their internal models will 245 00:15:57,600 --> 00:16:00,680 Speaker 1: get trained up with different parameters of physics, and they'll 246 00:16:00,720 --> 00:16:04,000 Speaker 1: put up their glove at a different time the moment 247 00:16:04,040 --> 00:16:07,720 Speaker 1: that's right for a Martian pop fly. Okay, But the 248 00:16:07,760 --> 00:16:09,640 Speaker 1: critical point I want to make here is that we 249 00:16:09,840 --> 00:16:14,080 Speaker 1: have these predictive internal models, and these things tell us 250 00:16:14,120 --> 00:16:18,920 Speaker 1: from experience how things are likely to move in the world. 251 00:16:19,080 --> 00:16:21,960 Speaker 1: And this way our brains don't work solely from the 252 00:16:22,160 --> 00:16:26,440 Speaker 1: latest sensory information, but instead they construct predictions about where 253 00:16:26,520 --> 00:16:29,360 Speaker 1: the ball is going to be. The same idea is 254 00:16:29,360 --> 00:16:32,520 Speaker 1: in play when you're walking through a busy airport, when 255 00:16:32,560 --> 00:16:35,119 Speaker 1: you have a flow of people moving in all directions 256 00:16:35,160 --> 00:16:39,400 Speaker 1: around you. If you had access to only outdated information 257 00:16:39,920 --> 00:16:43,840 Speaker 1: from photons a few hundred milliseconds ago, you'd be constantly 258 00:16:43,880 --> 00:16:44,920 Speaker 1: crashing into people. 259 00:16:44,960 --> 00:16:47,360 Speaker 2: But you don't. Your brain solves this. 260 00:16:47,880 --> 00:16:51,480 Speaker 1: Your brain is constantly forecasting where the people are going 261 00:16:51,520 --> 00:16:54,680 Speaker 1: to be based on their speed and direction, and that's 262 00:16:54,720 --> 00:16:59,480 Speaker 1: what allows you to smoothly navigate without crashing despite the 263 00:16:59,520 --> 00:17:04,400 Speaker 1: neural the and processing the visual information. So I want 264 00:17:04,400 --> 00:17:07,400 Speaker 1: to summarize where we are so far. The foundation we're 265 00:17:07,520 --> 00:17:10,400 Speaker 1: establishing here is that the brain is not just reacting 266 00:17:10,400 --> 00:17:13,680 Speaker 1: to the world. Instead, it's a machine that continuously makes 267 00:17:13,880 --> 00:17:19,679 Speaker 1: educated guesses. Prediction is how we compensate for our signal 268 00:17:19,760 --> 00:17:23,919 Speaker 1: processing delays, and from an evolutionary standpoint, this ability to 269 00:17:24,000 --> 00:17:28,520 Speaker 1: predict was absolutely critical for survival because animals who wanted 270 00:17:28,600 --> 00:17:32,720 Speaker 1: any chance of living how to anticipate the movements of 271 00:17:32,800 --> 00:17:36,439 Speaker 1: predators or prey to react quickly enough. You have to 272 00:17:36,520 --> 00:17:40,200 Speaker 1: somehow operate in real time if you want to evade 273 00:17:40,280 --> 00:17:43,840 Speaker 1: a thread or catch the running animal. So whenever you 274 00:17:44,080 --> 00:17:46,480 Speaker 1: are next catching a ball or moving through the airport, 275 00:17:47,000 --> 00:17:50,920 Speaker 1: think about how much you rely on your brain's predictive 276 00:17:51,080 --> 00:17:55,399 Speaker 1: abilities to act without having to wait for all the 277 00:17:55,480 --> 00:17:59,480 Speaker 1: signals to dribble their way in there. Okay, now we're 278 00:17:59,560 --> 00:18:02,280 Speaker 1: finally ready to return to the issue that I started with. 279 00:18:02,480 --> 00:18:05,320 Speaker 1: The illusions where you have some lines that are straight 280 00:18:05,400 --> 00:18:08,560 Speaker 1: but they look bent. These fall into the category of 281 00:18:09,160 --> 00:18:12,399 Speaker 1: geometric illusions. So what in the world do they have 282 00:18:12,480 --> 00:18:17,000 Speaker 1: to do with what we've been talking about so far. Well, 283 00:18:17,359 --> 00:18:19,560 Speaker 1: what I told you is that the visual system has 284 00:18:19,600 --> 00:18:23,160 Speaker 1: developed these predictive mechanisms to deal with the signal delays 285 00:18:23,560 --> 00:18:25,879 Speaker 1: so that it can see something at this moment in 286 00:18:25,920 --> 00:18:28,560 Speaker 1: time and make a really good guess where that thing 287 00:18:28,640 --> 00:18:32,439 Speaker 1: is going to be in say, one hundred milliseconds. So 288 00:18:33,080 --> 00:18:38,160 Speaker 1: some of my colleagues proposed a framework called perceiving the present, 289 00:18:38,720 --> 00:18:42,119 Speaker 1: and the idea is that your brain sees what is 290 00:18:42,400 --> 00:18:46,199 Speaker 1: likely to be the case, rather than to perceive the 291 00:18:46,280 --> 00:18:49,679 Speaker 1: recent past. So the first examples of this framework came 292 00:18:49,720 --> 00:18:53,440 Speaker 1: out in the early nineteen nineties. So imagine you're looking 293 00:18:53,480 --> 00:18:56,399 Speaker 1: at a small horizontal line on a computer screen and 294 00:18:56,440 --> 00:18:59,439 Speaker 1: you're trying to judge its exact position, but there's a 295 00:18:59,480 --> 00:19:03,640 Speaker 1: field of dots drifting continuously in the upward direction behind 296 00:19:03,760 --> 00:19:06,720 Speaker 1: that line. In this case, you'll judge the line to 297 00:19:06,760 --> 00:19:10,480 Speaker 1: be higher up on the screen. This is called motion capture. 298 00:19:10,760 --> 00:19:13,400 Speaker 1: So by the beginning of the two thousands, my colleague 299 00:19:13,480 --> 00:19:18,120 Speaker 1: Mark Chengizi started proposing that the explanation for this motion 300 00:19:18,240 --> 00:19:23,080 Speaker 1: capture was the perceiving the present framework, which is that 301 00:19:23,119 --> 00:19:26,480 Speaker 1: your brain sees the line, and it sees the motion 302 00:19:27,040 --> 00:19:30,199 Speaker 1: and decides that in the next moment the line is 303 00:19:30,320 --> 00:19:33,320 Speaker 1: probably going to be pushed up by the motion. So 304 00:19:33,400 --> 00:19:37,679 Speaker 1: it's actually perceiving the line in a different place where 305 00:19:37,920 --> 00:19:41,840 Speaker 1: it expects the line to be in the next moment. 306 00:19:42,600 --> 00:19:46,880 Speaker 1: And besides that, he argued, he could explain the classical 307 00:19:47,000 --> 00:20:08,560 Speaker 1: geometric illusion. What are these classical geometric illusions. Well, let's 308 00:20:08,560 --> 00:20:12,560 Speaker 1: take what's known as the Herring illusion. You almost certainly 309 00:20:12,600 --> 00:20:14,919 Speaker 1: saw this as a kid. There are a bunch of 310 00:20:15,000 --> 00:20:17,440 Speaker 1: lines coming out of the center, like the spokes on 311 00:20:17,480 --> 00:20:21,920 Speaker 1: a bicycle wheel. Okay, now you put two parallel lines 312 00:20:22,359 --> 00:20:25,520 Speaker 1: on that bicycle wheel, let's say, a vertical line to 313 00:20:25,560 --> 00:20:28,320 Speaker 1: the right of center and one to the left. You 314 00:20:28,400 --> 00:20:32,240 Speaker 1: could do this by taping two pencils on the bicycle spokes. 315 00:20:32,720 --> 00:20:36,560 Speaker 1: Now here's the illusion to two pencils. Although they are straight, 316 00:20:36,960 --> 00:20:39,199 Speaker 1: they don't look that way anymore. Instead, it looks like 317 00:20:39,240 --> 00:20:43,560 Speaker 1: the pencils are curving, they look slightly bent. Their middles 318 00:20:43,680 --> 00:20:47,480 Speaker 1: are bowing outwards slightly. So this is an illusion that 319 00:20:47,560 --> 00:20:52,080 Speaker 1: was first described by the physiologist Ewald Herring in eighteen 320 00:20:52,160 --> 00:20:52,760 Speaker 1: sixty one. 321 00:20:53,119 --> 00:20:55,040 Speaker 2: But why in the world does it happen? 322 00:20:55,200 --> 00:20:57,320 Speaker 1: Well, Herring proposed that this has to do with the 323 00:20:57,359 --> 00:21:01,520 Speaker 1: brain overestimating angles where the lines are meeting. And then 324 00:21:01,600 --> 00:21:04,440 Speaker 1: other people proposed different things in the brain that might 325 00:21:04,800 --> 00:21:10,200 Speaker 1: explain that angle overestimation. But Changhizi proposed a new explanation 326 00:21:10,280 --> 00:21:14,520 Speaker 1: when which was quite stunning. He said, look, when you're 327 00:21:14,560 --> 00:21:17,520 Speaker 1: looking at these radial lines, in other words, the lines 328 00:21:17,640 --> 00:21:20,119 Speaker 1: like the bicycle spokes coming out from a central point, 329 00:21:20,960 --> 00:21:24,280 Speaker 1: your brain might think that it's just looking at the 330 00:21:24,400 --> 00:21:29,480 Speaker 1: convergence of lines to the vanishing point, like imagine you're 331 00:21:29,520 --> 00:21:32,040 Speaker 1: looking straight ahead on a street in New York City 332 00:21:32,280 --> 00:21:36,000 Speaker 1: and everything converges in the middle. But equally, he said, 333 00:21:36,080 --> 00:21:38,879 Speaker 1: these lines are what a scene looks like to your 334 00:21:38,960 --> 00:21:43,480 Speaker 1: visual system when you are moving forward. For example, imagine 335 00:21:43,480 --> 00:21:46,040 Speaker 1: that you're driving down the road in New York City 336 00:21:46,480 --> 00:21:48,680 Speaker 1: and up ahead on the left there's a hot dog 337 00:21:48,800 --> 00:21:51,119 Speaker 1: stand and that zips by you on your left side, 338 00:21:51,440 --> 00:21:53,840 Speaker 1: and at the same time, there's a street juggler on 339 00:21:53,880 --> 00:21:54,520 Speaker 1: your right. 340 00:21:54,320 --> 00:21:56,399 Speaker 2: Side, and he gets bigger and he zips by you 341 00:21:56,520 --> 00:21:56,840 Speaker 2: on that. 342 00:21:56,920 --> 00:21:59,840 Speaker 1: Side, and there's an overhead street sign that at a 343 00:22:00,040 --> 00:22:03,240 Speaker 1: distance starts essentially in the middle in front of you, 344 00:22:03,400 --> 00:22:05,919 Speaker 1: but as you get closer and closer, it moves over 345 00:22:05,960 --> 00:22:11,919 Speaker 1: your head. So everything is streaking past you like radial lines, 346 00:22:12,359 --> 00:22:15,399 Speaker 1: and this is known as optic flow. So one place 347 00:22:15,480 --> 00:22:18,680 Speaker 1: you've seen this before is on Star Trek, where they 348 00:22:19,080 --> 00:22:21,439 Speaker 1: yank down the lever and put the ship into warp 349 00:22:21,560 --> 00:22:25,440 Speaker 1: drive and all the stars suddenly shoot past them, all 350 00:22:25,600 --> 00:22:29,119 Speaker 1: moving away from the center, like the radial spokes of 351 00:22:29,160 --> 00:22:32,720 Speaker 1: the bicycle wheel. So Tanghizi said, when you see radial 352 00:22:32,840 --> 00:22:37,159 Speaker 1: lines like that, it's typically a visual signature of you 353 00:22:37,840 --> 00:22:41,520 Speaker 1: moving forward towards the vanishing point. And certainly when you're 354 00:22:41,520 --> 00:22:44,760 Speaker 1: moving fast, there's a radial smear, like the way that 355 00:22:44,800 --> 00:22:49,200 Speaker 1: the stars and Star Trek smear into lines. And he said, look, 356 00:22:49,600 --> 00:22:54,000 Speaker 1: Herring's radial lines essentially mimic this. It's like you are 357 00:22:54,040 --> 00:22:58,119 Speaker 1: in the spaceship moving directly ahead. Now here's the key. 358 00:22:58,880 --> 00:23:01,359 Speaker 1: Let's come back to the de lays in the visual 359 00:23:01,400 --> 00:23:04,560 Speaker 1: system and how they can be accounted for by the 360 00:23:04,720 --> 00:23:10,200 Speaker 1: brain making projections where things are about to be. Imagine 361 00:23:10,200 --> 00:23:12,840 Speaker 1: that you're in New York City and driving and there 362 00:23:12,840 --> 00:23:15,359 Speaker 1: are two skyscrapers up ahead of you, one on the 363 00:23:15,440 --> 00:23:17,680 Speaker 1: left and one on the right. Now, as you race 364 00:23:17,840 --> 00:23:22,160 Speaker 1: forward in your car, those two buildings will loom closer. 365 00:23:22,200 --> 00:23:25,280 Speaker 1: But now something interesting is happening. The parts of the 366 00:23:25,320 --> 00:23:30,000 Speaker 1: buildings closer to you will seem farther apart, because if 367 00:23:30,080 --> 00:23:32,800 Speaker 1: you look up, the tips of the buildings are coming 368 00:23:32,880 --> 00:23:34,639 Speaker 1: closer together, way up in the sky. 369 00:23:35,160 --> 00:23:36,320 Speaker 2: So the point is that. 370 00:23:36,520 --> 00:23:40,199 Speaker 1: Even though you see essentially straight skyscrapers when they're at 371 00:23:40,200 --> 00:23:44,640 Speaker 1: a distance, as you approach, they are bending away from you. 372 00:23:44,960 --> 00:23:50,720 Speaker 1: Their centers are bowing out. And Shannghizi's idea was that 373 00:23:50,760 --> 00:23:54,000 Speaker 1: when you look at the radial lines the bicycle spokes, 374 00:23:54,000 --> 00:23:56,240 Speaker 1: your brain thinks this might be a clue that I'm 375 00:23:56,280 --> 00:23:59,240 Speaker 1: moving forward, and I don't want there to be delays 376 00:23:59,280 --> 00:24:02,439 Speaker 1: in my perception, so I'm going to see the world 377 00:24:02,680 --> 00:24:06,919 Speaker 1: as it will be a moment later. And so you 378 00:24:07,000 --> 00:24:10,959 Speaker 1: see the two parallel lines bowed outward from the center. 379 00:24:11,359 --> 00:24:13,439 Speaker 1: In other words, when you look at the radio lines 380 00:24:13,440 --> 00:24:16,119 Speaker 1: on the piece of paper, even though nothing is moving. 381 00:24:16,920 --> 00:24:20,040 Speaker 1: Your brain thinks this is what movement looks like, and 382 00:24:20,119 --> 00:24:24,359 Speaker 1: so it predicts the next moment, and that's what you perceived. 383 00:24:24,520 --> 00:24:26,600 Speaker 1: In other words, you perceive the lines just as they 384 00:24:26,720 --> 00:24:29,320 Speaker 1: would project in the next moment if you were moving 385 00:24:29,400 --> 00:24:33,480 Speaker 1: forward toward the vanishing point. So, as Changhesi wrote in 386 00:24:33,520 --> 00:24:38,680 Speaker 1: this paper, evolution has seen to it that geometric drawings 387 00:24:38,840 --> 00:24:45,400 Speaker 1: like this elicit in us premonitions of the near future. Okay, 388 00:24:45,760 --> 00:24:50,000 Speaker 1: so the framework by Changizi and colleagues suggests that several 389 00:24:50,040 --> 00:24:54,199 Speaker 1: geometric illusions are caused by temporal delays with which the 390 00:24:54,280 --> 00:24:57,040 Speaker 1: visual system must cope. The idea is that the visual 391 00:24:57,040 --> 00:25:03,000 Speaker 1: system extrapolates its current information to perceive the present. Instead 392 00:25:03,000 --> 00:25:06,399 Speaker 1: of providing a conscious image of how the world was 393 00:25:06,440 --> 00:25:09,600 Speaker 1: a few hundred milliseconds ago when the signals first struck 394 00:25:09,640 --> 00:25:13,000 Speaker 1: the retina, the visual system estimates how the world is 395 00:25:13,240 --> 00:25:16,399 Speaker 1: likely to look in the next moment. But how would 396 00:25:16,400 --> 00:25:20,520 Speaker 1: we get at clues to the possible neural basis? In 397 00:25:20,520 --> 00:25:23,240 Speaker 1: other words, how does the brain actually pull this off? 398 00:25:23,960 --> 00:25:26,280 Speaker 1: So in my laboratory we wanted to figure this out. 399 00:25:26,520 --> 00:25:29,840 Speaker 1: So my student Don Vaughan and I had people look 400 00:25:29,920 --> 00:25:32,679 Speaker 1: at the herring illusion on a screen. You've got a 401 00:25:32,720 --> 00:25:35,960 Speaker 1: background of radial lines like the bicycle spokes, and we 402 00:25:36,040 --> 00:25:39,320 Speaker 1: flashed two vertical lines on top of this. And I'll 403 00:25:39,320 --> 00:25:40,960 Speaker 1: just take a quick second to give you a sense 404 00:25:41,000 --> 00:25:46,119 Speaker 1: of how we quantify illusions. In the laboratory, A person 405 00:25:46,160 --> 00:25:48,440 Speaker 1: sits in front of the computer and they use let's 406 00:25:48,440 --> 00:25:51,239 Speaker 1: say the right and left arrows on the keyboard to 407 00:25:51,440 --> 00:25:55,400 Speaker 1: change the curvature of those two lines. So at one 408 00:25:55,520 --> 00:25:58,920 Speaker 1: end of the range, they're actually physically bending the lines outward, 409 00:25:59,119 --> 00:26:01,760 Speaker 1: and at the other extreme they're bending them inward, and 410 00:26:01,840 --> 00:26:04,600 Speaker 1: somewhere in the middle they're physically straight. And what the 411 00:26:04,600 --> 00:26:08,560 Speaker 1: person does is adjust the curvature of the line until 412 00:26:08,600 --> 00:26:11,560 Speaker 1: it looks straight to them. But of course with the 413 00:26:11,600 --> 00:26:15,199 Speaker 1: hairing illusion, you need to actually adjust the lines so 414 00:26:15,240 --> 00:26:18,159 Speaker 1: the middles are curving inward in order for it to 415 00:26:18,240 --> 00:26:22,440 Speaker 1: look straight. In other words, we see how much curvature 416 00:26:22,520 --> 00:26:25,760 Speaker 1: it takes to cancel out the illusion, and that's the 417 00:26:25,800 --> 00:26:29,480 Speaker 1: way we can quantify the size of the illusion. Okay, 418 00:26:29,520 --> 00:26:32,240 Speaker 1: so we measure the heiring illusion, and no surprises there. 419 00:26:32,520 --> 00:26:36,080 Speaker 1: But now what we do is we replaced the radio 420 00:26:36,200 --> 00:26:40,800 Speaker 1: lines with an actual star field. We have dots moving 421 00:26:40,960 --> 00:26:44,639 Speaker 1: in an expanding pattern like the stars and Star Trek, 422 00:26:45,359 --> 00:26:48,240 Speaker 1: And now people are judging the size of the hering 423 00:26:48,280 --> 00:26:52,120 Speaker 1: illusion against the background of expanding dots. 424 00:26:52,800 --> 00:26:54,080 Speaker 2: And what we find is that. 425 00:26:54,000 --> 00:26:58,920 Speaker 1: The illusion still happens. The lines still appear bent, which 426 00:26:58,960 --> 00:27:04,040 Speaker 1: is just what you'd expect from the perceiving the present hypothesis. Okay, 427 00:27:04,040 --> 00:27:07,360 Speaker 1: but here's the really wacky part, the part that uncovers 428 00:27:07,480 --> 00:27:11,080 Speaker 1: an unexpected secret in the brain. We now measure the 429 00:27:11,080 --> 00:27:15,520 Speaker 1: size of the hering illusion over a field of contracting dots. 430 00:27:16,000 --> 00:27:19,520 Speaker 1: So picture that Star Trek footage running backwards. Now everything 431 00:27:19,600 --> 00:27:21,840 Speaker 1: is moving from the outside to the inside. 432 00:27:22,359 --> 00:27:23,440 Speaker 2: And here's the surprise. 433 00:27:23,560 --> 00:27:26,760 Speaker 1: We find that the size of the herring illusion was 434 00:27:26,840 --> 00:27:30,399 Speaker 1: exactly the same here. In other words, the lines still 435 00:27:30,520 --> 00:27:33,440 Speaker 1: curve outward, just like in the other two cases. 436 00:27:34,080 --> 00:27:35,920 Speaker 2: So what does that mean. We can get this. 437 00:27:35,920 --> 00:27:42,480 Speaker 1: Illusion by having radial lines or dots expanding or dots contracting, 438 00:27:43,000 --> 00:27:45,760 Speaker 1: and you find that the bars bend out in the 439 00:27:45,760 --> 00:27:51,119 Speaker 1: same direction. Now, at first glance, the bending of the 440 00:27:51,160 --> 00:27:56,560 Speaker 1: bars during contracting motion would seem to refute the perceiving 441 00:27:56,600 --> 00:27:59,959 Speaker 1: the present framework. If your brain is doing an active 442 00:28:00,119 --> 00:28:04,159 Speaker 1: of temporal extrapolation of the scene, it should make the 443 00:28:04,200 --> 00:28:08,200 Speaker 1: bars bend in the other direction. But the key thing 444 00:28:08,240 --> 00:28:12,800 Speaker 1: to note is that backward motion is ecologically quite rare. 445 00:28:13,000 --> 00:28:16,000 Speaker 1: Unless you're a dog looking out the back window of 446 00:28:16,040 --> 00:28:21,280 Speaker 1: a car. Most animals probably never see backward optic flow 447 00:28:21,320 --> 00:28:23,520 Speaker 1: in their lives. Okay, so we did a lot of 448 00:28:23,560 --> 00:28:27,080 Speaker 1: other experiments in this paper, but just this first result 449 00:28:27,119 --> 00:28:31,240 Speaker 1: that the Herring illusion happens the same with expanding or 450 00:28:31,400 --> 00:28:35,919 Speaker 1: contracting optic flow tells us some critical things. First, it 451 00:28:36,000 --> 00:28:39,760 Speaker 1: tells us that this spatial warping we see isn't a 452 00:28:40,120 --> 00:28:46,160 Speaker 1: sophisticated online computation. Instead, it's a basic mechanism that just 453 00:28:46,360 --> 00:28:49,400 Speaker 1: acts to get it right in the most common scenario 454 00:28:49,720 --> 00:28:53,760 Speaker 1: of forward motion. And because backward motion essentially never happens, 455 00:28:54,120 --> 00:28:58,120 Speaker 1: it doesn't matter that the mechanism is so unrefined, and 456 00:28:58,160 --> 00:29:00,960 Speaker 1: it gives us a big clue about the the underlying 457 00:29:01,120 --> 00:29:05,560 Speaker 1: neural mechanisms. The most parsimonious explanation. In other words, the 458 00:29:05,600 --> 00:29:09,160 Speaker 1: simplest idea would look for something in the brain that's 459 00:29:09,240 --> 00:29:13,640 Speaker 1: equally sensitive to lines like the bicycle spoke and also 460 00:29:13,760 --> 00:29:17,680 Speaker 1: to motion in either direction along that line. And it 461 00:29:17,720 --> 00:29:20,840 Speaker 1: turns out there are very simple neurons in primary visual 462 00:29:20,840 --> 00:29:25,640 Speaker 1: cortex that do exactly this. They're called orientation selective neurons, 463 00:29:25,920 --> 00:29:28,680 Speaker 1: and they respond to lines, and they also respond to 464 00:29:29,160 --> 00:29:32,479 Speaker 1: dots moving along the same trajectory as the line in 465 00:29:32,600 --> 00:29:35,960 Speaker 1: either direction. So what this means is that our findings 466 00:29:36,000 --> 00:29:40,360 Speaker 1: are consistent with the perceiving the present hypothesis, with the 467 00:29:40,400 --> 00:29:45,520 Speaker 1: caveat that the spatial warping to counteract neural delays is 468 00:29:45,560 --> 00:29:49,680 Speaker 1: not a smart active neural process, but instead it's just 469 00:29:49,720 --> 00:29:54,920 Speaker 1: a simple mechanism that succeeds only under forward moving circumstances, 470 00:29:54,920 --> 00:29:57,520 Speaker 1: which is most of the time. Now, as I said, 471 00:29:57,520 --> 00:29:59,320 Speaker 1: we did a lot of other experiments, and I'm gonna 472 00:29:59,320 --> 00:30:01,720 Speaker 1: skip the detail because they're not as important as this 473 00:30:01,960 --> 00:30:06,240 Speaker 1: main point. But all the other experiments supported this hypothesis 474 00:30:06,520 --> 00:30:10,680 Speaker 1: that the bars look bent because the brain is extrapolating 475 00:30:10,720 --> 00:30:14,440 Speaker 1: ahead what things would look like in the next moment. 476 00:30:14,920 --> 00:30:17,920 Speaker 1: And by the way, our findings weren't consistent with older 477 00:30:18,000 --> 00:30:22,320 Speaker 1: theories like angle over estimation, as Herring had suggested. So 478 00:30:22,640 --> 00:30:24,360 Speaker 1: if you want to dive deeper into the paper, I'm 479 00:30:24,400 --> 00:30:28,520 Speaker 1: linking it to the show notes. So what we saw 480 00:30:28,600 --> 00:30:31,800 Speaker 1: today is the issue that it always takes time for 481 00:30:31,960 --> 00:30:35,520 Speaker 1: signals to move through the brain. And although people got 482 00:30:35,520 --> 00:30:39,280 Speaker 1: excited when they discovered that the signals were carried by electricity, 483 00:30:39,320 --> 00:30:43,520 Speaker 1: it's nothing like the way that electricity runs on wires, 484 00:30:43,520 --> 00:30:46,400 Speaker 1: which is close to the speed of light. In the brain, 485 00:30:46,840 --> 00:30:50,480 Speaker 1: we're talking hundreds of millions of times slower than that. 486 00:30:51,040 --> 00:30:54,280 Speaker 1: So you have these signals limping along in the brain, 487 00:30:54,800 --> 00:30:58,120 Speaker 1: and that means we are always living in the past. 488 00:30:58,520 --> 00:31:01,920 Speaker 1: Our brains sometimes see fake news about the world out there, 489 00:31:01,920 --> 00:31:05,560 Speaker 1: but we always see old news. Your brain is always 490 00:31:05,600 --> 00:31:10,280 Speaker 1: getting information about an outdated version of the world, one 491 00:31:10,320 --> 00:31:13,600 Speaker 1: that no longer exists. And so one of the many 492 00:31:13,800 --> 00:31:18,200 Speaker 1: incredible things your brain does is make predictions so that 493 00:31:18,280 --> 00:31:21,880 Speaker 1: you can perceive things as they probably are right now, 494 00:31:22,440 --> 00:31:26,560 Speaker 1: rather than perceiving the stale version of the data, which 495 00:31:26,600 --> 00:31:28,560 Speaker 1: is much less useful by the time you see it. 496 00:31:29,120 --> 00:31:32,720 Speaker 1: This is the brain's very bold move to compensate for 497 00:31:32,800 --> 00:31:36,600 Speaker 1: its own delays, and that wacky fact explains a lot 498 00:31:36,680 --> 00:31:39,520 Speaker 1: about how we catch balls and walk through crowded airports, 499 00:31:39,800 --> 00:31:43,959 Speaker 1: but it also seems to explain surprisingly why we see 500 00:31:44,000 --> 00:31:49,080 Speaker 1: this basic geometric illusion of straight lines not looking straight, 501 00:31:49,640 --> 00:31:53,480 Speaker 1: discovered by Heiring almost one hundred and sixty five years ago. Now, 502 00:31:53,520 --> 00:31:56,480 Speaker 1: I wrote an article in Nature Reviews Neuroscience in which 503 00:31:56,520 --> 00:31:59,240 Speaker 1: I covered all kinds of visual illusions, and that's in 504 00:31:59,280 --> 00:32:02,960 Speaker 1: the show notes. And each one of these illusions opens 505 00:32:02,960 --> 00:32:07,760 Speaker 1: a new mind shaft into the brain, teaching us why 506 00:32:07,840 --> 00:32:11,240 Speaker 1: it's happening. And once we understand that, we can usually 507 00:32:11,560 --> 00:32:15,920 Speaker 1: construct new illusions, which is why amazing new illusions are 508 00:32:15,920 --> 00:32:16,920 Speaker 1: coming out each year. 509 00:32:17,600 --> 00:32:17,800 Speaker 2: Now. 510 00:32:17,840 --> 00:32:21,040 Speaker 1: Sometimes people see all these illusions on their social media 511 00:32:21,120 --> 00:32:24,720 Speaker 1: feeds and they might become tempted to say that everything 512 00:32:24,800 --> 00:32:28,040 Speaker 1: is an illusion. We don't accurately see what's really out there. 513 00:32:28,760 --> 00:32:31,680 Speaker 1: But that's probably not the right conclusion, because often we 514 00:32:31,840 --> 00:32:35,720 Speaker 1: do see what's out there accurately, as verified by our 515 00:32:35,840 --> 00:32:40,280 Speaker 1: other senses and by our objective machines. For example, if 516 00:32:40,320 --> 00:32:42,840 Speaker 1: I put a cup of steaming coffee on the table 517 00:32:42,880 --> 00:32:45,480 Speaker 1: in front of you, you see that, and you heard 518 00:32:45,560 --> 00:32:47,760 Speaker 1: the clunk when I set it down, and you can 519 00:32:47,880 --> 00:32:51,040 Speaker 1: verify the heat with your fingers and smell the coffee 520 00:32:51,080 --> 00:32:53,400 Speaker 1: and pick it up and taste it, and you can 521 00:32:53,480 --> 00:32:56,280 Speaker 1: measure the presence of the cup with video, or the 522 00:32:56,360 --> 00:32:59,960 Speaker 1: heat with an infrared camera, and the coffee with spectroscopy 523 00:33:00,080 --> 00:33:02,640 Speaker 1: and so on. So it's not that everything you see 524 00:33:02,760 --> 00:33:09,040 Speaker 1: is illusory. Instead, we see what is maximally useful to us. 525 00:33:09,720 --> 00:33:12,200 Speaker 1: And every once in a while we can throw an 526 00:33:12,320 --> 00:33:16,160 Speaker 1: unexpected wrench in the system by showing the brain something 527 00:33:16,520 --> 00:33:20,200 Speaker 1: that happens to tickle the right receptors. And in those 528 00:33:20,400 --> 00:33:24,760 Speaker 1: very special cases we can catch the system coming to 529 00:33:24,800 --> 00:33:28,880 Speaker 1: the wrong conclusion. But it's precisely because the brain has 530 00:33:28,920 --> 00:33:32,440 Speaker 1: evolved to do the optimal thing in all the other cases. 531 00:33:32,760 --> 00:33:37,280 Speaker 1: And traditionally, when we find these special cases, we just 532 00:33:37,520 --> 00:33:40,800 Speaker 1: laugh and enjoy and keep scrolling through our feed But 533 00:33:40,920 --> 00:33:46,760 Speaker 1: the endeavor of figuring ourselves out encourages us to pause, 534 00:33:47,400 --> 00:33:51,480 Speaker 1: to ask why, to dig deeper, and although these things 535 00:33:51,520 --> 00:33:57,160 Speaker 1: sometimes take centuries to answer, they typically yield deep insights 536 00:33:57,560 --> 00:34:01,560 Speaker 1: into who we are and what is actually going on. 537 00:34:07,280 --> 00:34:10,280 Speaker 1: Go to Eagleman dot com slash podcast for more information 538 00:34:10,440 --> 00:34:13,399 Speaker 1: and to find further reading. Send me an email at 539 00:34:13,440 --> 00:34:16,640 Speaker 1: podcasts at eagleman dot com with questions or discussion, and 540 00:34:16,760 --> 00:34:20,120 Speaker 1: check out and subscribe to Inner Cosmos on YouTube for 541 00:34:20,239 --> 00:34:24,479 Speaker 1: videos of each episode and to leave comments until next time. 542 00:34:24,600 --> 00:34:27,879 Speaker 1: I'm David Eagleman, and this is Inner Cosmos.