WEBVTT - Ep 87 "How do we operate in the present when we perceive the past?" 0:00:05.000 --> 0:00:08.080 Have you ever seen those pictures of blobs on a 0:00:08.160 --> 0:00:11.399 page and it doesn't look like anything to you until 0:00:11.440 --> 0:00:14.640 you're told what it is, and then you suddenly see it. 0:00:15.400 --> 0:00:18.320 Why does that give us a great clue about the 0:00:18.360 --> 0:00:23.239 wiring of the brain. And why are neuroscientists so magnetically 0:00:23.239 --> 0:00:27.080 attracted to those visual illusions that you scroll through on 0:00:27.120 --> 0:00:31.040 social media? What is the deep trick about the way 0:00:31.120 --> 0:00:34.080 your visual system works that you were never taught in school? 0:00:34.520 --> 0:00:36.440 And what does any of this have to do with 0:00:36.640 --> 0:00:39.760 catching a baseball or zooming down the road in New 0:00:39.840 --> 0:00:46.640 York City or the warp drive in Star Trek. Welcome 0:00:46.680 --> 0:00:50.360 to Inner Cosmos with me David Eagleman. I'm a neuroscientist 0:00:50.360 --> 0:00:53.480 and author at Stanford and in these episodes we sail 0:00:53.600 --> 0:00:57.240 deeply into our three pound universe to understand why and 0:00:57.320 --> 0:00:59.840 how our lives look the way they do. 0:01:09.600 --> 0:01:09.880 Today. 0:01:09.920 --> 0:01:13.640 I'm going to start with the notion of visual illusions. 0:01:14.200 --> 0:01:18.120 Elementary school students love these and they stare at them 0:01:18.240 --> 0:01:20.200 for about a minute and then they're on to the 0:01:20.240 --> 0:01:23.600 next thing, because why not. The illusion is just an 0:01:23.640 --> 0:01:27.040 interesting trick. There's nothing further to do about it. It's 0:01:27.080 --> 0:01:30.080 only later when you grow up to be a neuroscientist 0:01:30.240 --> 0:01:33.520 or a fan of a neuroscience podcast, that you might 0:01:33.640 --> 0:01:37.360 even return to one of these illusions to ask, wait, 0:01:38.000 --> 0:01:41.280 why exactly does that happen? 0:01:41.440 --> 0:01:42.959 Does that tell us. 0:01:42.840 --> 0:01:48.800 Something deep and fundamental about the way my consciousness constructs 0:01:48.840 --> 0:01:49.880 the world for me? 0:01:50.560 --> 0:01:53.280 What does it reveal? So? 0:01:53.880 --> 0:01:56.840 Have you ever seen the illusion where you're looking at 0:01:57.200 --> 0:02:00.840 lines like bicycle spokes, and then there's some straight lines 0:02:00.920 --> 0:02:03.600 drawn on top of that, and they don't look straight, 0:02:03.720 --> 0:02:07.800 they look bent. Why does that happen? Seems like it 0:02:07.840 --> 0:02:10.280 shouldn't be hard to answer, but it's actually taken well 0:02:10.320 --> 0:02:13.359 over a century to figure this out, and the answer 0:02:13.720 --> 0:02:16.920 is gonna blow your mind. I promise you that. But 0:02:17.720 --> 0:02:20.600 in order to get us there, I'm gonna start with 0:02:20.680 --> 0:02:24.560 something completely different. I'm gonna start with those pictures that 0:02:24.639 --> 0:02:28.040 look like just a bunch of blobs. Probably you've seen 0:02:28.080 --> 0:02:30.880 one of these before. There's just a bunch of random 0:02:30.919 --> 0:02:34.840 looking splotches of black and white on a page. If 0:02:34.880 --> 0:02:40.240 your brain doesn't have a prior expectation about what it's seeing, 0:02:40.600 --> 0:02:44.360 about what the blobs mean, then you simply see black 0:02:44.400 --> 0:02:47.520 and white blobs and there's no particular meaning to the picture. 0:02:47.720 --> 0:02:49.720 I'm gonna link an example of this on the show 0:02:49.760 --> 0:02:52.720 notes at eagleman dot com slash podcast, and I want 0:02:52.760 --> 0:02:55.080 you to stare at it for a few moments and 0:02:55.120 --> 0:02:57.240 then scroll down to the very bottom of the page 0:02:57.360 --> 0:03:01.160 for the hint. What you'll see is that you can't 0:03:01.240 --> 0:03:04.320 make heads or tails of these blobs. But then I 0:03:04.360 --> 0:03:07.200 only change one thing, and it has nothing to do 0:03:07.280 --> 0:03:10.680 with what's on the screen. I give you a hint, 0:03:11.360 --> 0:03:14.520 and as soon as you have a notion about how 0:03:14.560 --> 0:03:18.119 to interpret what is on your retinas, then you say, 0:03:18.200 --> 0:03:21.720 oh yeah, I see it now. Now the exact same 0:03:21.840 --> 0:03:25.480 blobs that confused you a moment ago make perfect sense. 0:03:25.520 --> 0:03:29.079 But again, nothing changed out there in the world. The 0:03:29.120 --> 0:03:33.200 only thing that changed is something in your neural networks. 0:03:35.400 --> 0:03:38.000 So what's the lesson from this? There has to be 0:03:38.080 --> 0:03:43.040 a match between incoming data and your expectations for you 0:03:43.160 --> 0:03:48.480 to see anything. But wait, what, That's not how vision 0:03:48.560 --> 0:03:50.760 is supposed to work, is it. I mean, after all, 0:03:50.760 --> 0:03:54.280 you look at any basic biology textbook and it will 0:03:54.280 --> 0:03:57.400 tell you that photons hit the retina and the information 0:03:57.520 --> 0:04:00.440 is carried on back to the visual cortex, and then 0:04:00.480 --> 0:04:04.200 you just see what's out there. The visual cortex is 0:04:04.320 --> 0:04:07.720 like a television screen, So what's going on? 0:04:08.280 --> 0:04:10.080 Why can't you see. 0:04:09.720 --> 0:04:13.920 The image in the blobs until you've got the right expectation. 0:04:14.480 --> 0:04:15.760 This ties into. 0:04:15.520 --> 0:04:18.000 A concept that you hear me refer to all the 0:04:18.040 --> 0:04:20.599 time on this podcast, and that is the concept of 0:04:20.680 --> 0:04:25.760 the internal model. Remember that your brain is isolated in 0:04:26.160 --> 0:04:31.240 soundless and lightless solitude inside your skull, and its single 0:04:31.360 --> 0:04:37.039 mission is to construct a loud, colorful mental model of 0:04:37.080 --> 0:04:40.880 the outside world. In other words, it builds an inner 0:04:40.880 --> 0:04:46.520 reality that tries to accurately reflect the outside. The key 0:04:46.720 --> 0:04:51.200 is that you don't see by capturing television pixels from 0:04:51.279 --> 0:04:56.640 the world. Instead, all you ever see is your internal model, 0:04:57.160 --> 0:05:01.360 and your internal model only perceives some thing when its 0:05:01.360 --> 0:05:08.200 expectations are sufficiently supported by the sensory data coming in. Now, 0:05:08.480 --> 0:05:12.200 this isn't really a widely known idea. I think you'll 0:05:12.200 --> 0:05:15.240 find if you ask people about it on airplanes, as 0:05:15.360 --> 0:05:17.960 I often do. This isn't really the way that most 0:05:18.000 --> 0:05:21.280 people are used to thinking about the brain. So it's 0:05:21.279 --> 0:05:24.920 a bit surprising that the basic conceptualization of this idea 0:05:25.000 --> 0:05:28.520 is almost seventy years old. One of the earliest examples 0:05:28.600 --> 0:05:31.279 of this framework that I know of came from the 0:05:31.320 --> 0:05:35.640 neuroscientist Donald McKay, who in nineteen fifty six said, Look, 0:05:35.839 --> 0:05:39.120 the job of the visual cortex is to construct an 0:05:39.160 --> 0:05:43.159 internal model, and it's always trying to anticipate the data 0:05:43.560 --> 0:05:46.479 coming up from the retina. Now, just as a reminder, 0:05:46.520 --> 0:05:48.000 the retina is the part of the back of your 0:05:48.040 --> 0:05:51.080 eye that captures light, and the visual cortex is all 0:05:51.080 --> 0:05:52.560 the way at the back of your head, on the 0:05:52.560 --> 0:05:55.839 far side of the brain. But here's the surprise. The 0:05:55.920 --> 0:06:00.920 information doesn't just shoot from the retina to the visual cortex. Instead, 0:06:01.320 --> 0:06:05.880 there's a train station in between, a structure called the thalamus. 0:06:05.960 --> 0:06:09.279 The thalamus sits right in the middle. So information doesn't 0:06:09.320 --> 0:06:11.599 go straight from the eye to the visual cortex, but 0:06:11.680 --> 0:06:15.280 instead it makes a stop and changes trains halfway at 0:06:15.279 --> 0:06:15.960 the thalamus. 0:06:16.400 --> 0:06:18.920 Okay, well that's weird. Why is there the setup? 0:06:19.600 --> 0:06:22.640 Well, to understand this, we need to understand that the 0:06:22.760 --> 0:06:27.800 model of vision in introductory textbooks isn't just misleading, it's 0:06:27.920 --> 0:06:32.600 dead wrong. The brain isn't built on straight lines, it's 0:06:32.600 --> 0:06:37.040 built with loops. So what McKay suggested is that the 0:06:37.080 --> 0:06:40.599 retina sends its data to the thalamus. In other words, 0:06:40.600 --> 0:06:42.719 what the eye is capturing about the world out there, 0:06:43.160 --> 0:06:47.800 and the cortex sends its predictions to the thalamus what 0:06:47.839 --> 0:06:51.039 the cortex is expecting to see next, and all that 0:06:51.120 --> 0:06:54.000 ever comes out of the thalamus back to the cortex 0:06:54.720 --> 0:06:58.839 is the difference, the difference between what you expected and 0:06:58.880 --> 0:07:02.440 what you got. In other words, the information that goes 0:07:02.920 --> 0:07:06.080 from the thalamis to the visual cortex is just that 0:07:06.200 --> 0:07:10.920 little bit which was unanticipated, the difference between what's out 0:07:10.920 --> 0:07:14.440 there and what was already expected. The thalamus sends to 0:07:14.520 --> 0:07:19.040 the cortex only that difference signal, because that's the only 0:07:19.080 --> 0:07:23.000 part that wasn't predicted away. And then, by the way, 0:07:23.040 --> 0:07:27.640 this unpredicted information adjusts the internal model so there will 0:07:27.680 --> 0:07:30.720 be less of a mismatch in the future. That way, 0:07:30.720 --> 0:07:34.560 the brain refines its model of the world by paying 0:07:34.600 --> 0:07:39.360 attention to its mistakes. Okay, so the idea here is 0:07:39.360 --> 0:07:42.600 that the brain is always trying to anticipate what it's 0:07:42.640 --> 0:07:45.520 seeing out there, and McKay pointed out that this is 0:07:45.600 --> 0:07:50.560 consistent with the anatomical fact that there are ten times 0:07:50.640 --> 0:07:55.240 as many fibers projecting from the visual cortex back to 0:07:55.280 --> 0:07:59.440 the thalamis as there are going from thalamis to visual cortex, 0:07:59.720 --> 0:08:02.920 which no one would have guessed. But that's just what 0:08:03.000 --> 0:08:07.480 you'd expect if detailed predictions are going from the cortex 0:08:07.600 --> 0:08:11.320 to the thalamis, and the little signal from thalamis back 0:08:11.360 --> 0:08:15.280 to cortex is just carrying the difference. 0:08:15.160 --> 0:08:17.360 Between what was expected and what was seen. 0:08:18.080 --> 0:08:20.600 Okay, So why am I telling you this level of 0:08:20.680 --> 0:08:24.840 detail because it exposes a giant idea. It means that 0:08:24.920 --> 0:08:29.480 what you perceive about the world emerges from an active 0:08:29.600 --> 0:08:35.439 comparison of sensory data with your internal predictions. Again, think 0:08:35.480 --> 0:08:38.560 about those blobs. If you don't have a prediction of 0:08:38.600 --> 0:08:42.000 what you're seeing out there, there's really nothing there. As 0:08:42.080 --> 0:08:46.280 soon as you have a close enough expectation because you've 0:08:46.320 --> 0:08:49.320 been given a hint, then that lights up a forest 0:08:49.360 --> 0:08:53.040 fire in your brain and you see the thing because 0:08:53.080 --> 0:08:54.000 there's a match. 0:08:54.120 --> 0:08:56.280 Now. So what this. 0:08:56.280 --> 0:08:59.720 Means is that the brain is always trying to predict 0:09:00.240 --> 0:09:04.160 everything that is coming or expected. And here's one way 0:09:04.160 --> 0:09:07.480 that the brain helps itself along. Whenever it sends out 0:09:07.520 --> 0:09:10.720 a signal to your body, like move your head or 0:09:11.000 --> 0:09:15.960 move your arm, it also sends copies of that command 0:09:16.160 --> 0:09:20.239 internally all around the brain. These are called efference copies. 0:09:20.520 --> 0:09:24.240 So now your movement isn't just happening in the outside 0:09:24.320 --> 0:09:26.720 world and then you react to it, but there's also 0:09:26.760 --> 0:09:31.720 a simulation of that movement happening inside your internal model, 0:09:32.160 --> 0:09:35.840 so that you can predict the outcome of that action. 0:09:36.520 --> 0:09:39.760 And this, by the way, is the reason you can't 0:09:39.960 --> 0:09:45.120 tickle yourself. Other people can tickle you because they're tickling 0:09:45.200 --> 0:09:48.760 maneuvers are not predictable to you. But you can't tickle 0:09:48.840 --> 0:09:53.120 you because your brain moves your fingers into the tickle 0:09:53.200 --> 0:09:58.960 position and it already expects the resulting sensations, that already 0:09:59.000 --> 0:10:03.160 knows what's come. Now, by the way, if you'd really 0:10:03.240 --> 0:10:05.439 like to tickle yourself, there is a way to do it, 0:10:05.880 --> 0:10:11.440 and this just involves taking predictability away from your own actions. 0:10:12.000 --> 0:10:14.800 So what you do is you control the position of 0:10:14.840 --> 0:10:21.000 a feather with a joystick that inserts a random time delay, 0:10:21.080 --> 0:10:24.320 so when you move the joystick, at least a second 0:10:24.440 --> 0:10:29.000 passes before the feather moves accordingly, so that takes away 0:10:29.360 --> 0:10:33.560 the predictability and now you can self tickle. By the way, 0:10:33.720 --> 0:10:37.280 related to this, I described in episode forty four how 0:10:37.320 --> 0:10:42.440 people with schizophrenia can tickle themselves, and this is because 0:10:42.480 --> 0:10:46.720 of a problem with their internal timing that doesn't allow 0:10:46.840 --> 0:11:08.720 their motor actions and resulting sensations to be correctly sequenced. Okay, 0:11:08.760 --> 0:11:12.760 so back to this issue about having a brain that's 0:11:12.800 --> 0:11:16.720 not just moving signals down a one way assembly line, 0:11:16.720 --> 0:11:20.240 but instead has all these internal loops so that it 0:11:20.320 --> 0:11:24.760 can always be feeding its internal model and guessing what's 0:11:24.840 --> 0:11:28.960 going to happen next. What is the advantage of this, Well, 0:11:29.679 --> 0:11:35.160 it allows us to transcend stimulus response behavior. In other words, 0:11:35.440 --> 0:11:38.280 we don't have to just observe the world and then 0:11:38.520 --> 0:11:43.280 react to it. Instead, a brain with an internal model 0:11:43.800 --> 0:11:49.120 gives us the ability to make predictions ahead of actual 0:11:49.240 --> 0:11:52.440 sensory input, like predicting what your fingers will feel like 0:11:52.559 --> 0:11:57.840 in your underarm. So our brains build these predictive internal 0:11:57.920 --> 0:12:01.400 models that tell us how things are likely to go 0:12:01.880 --> 0:12:04.880 in the world. And this way our brains don't work 0:12:05.160 --> 0:12:09.199 solely from the latest sensory data, but instead they're always 0:12:09.440 --> 0:12:13.400 guessing ahead to the next moment. Now, why do we 0:12:13.480 --> 0:12:18.800 need a complicated brain like this because our perception is 0:12:19.040 --> 0:12:24.400 massively delayed from reality. Why is it delayed because signals 0:12:24.440 --> 0:12:27.600 from the world, like something you see or a touch 0:12:27.640 --> 0:12:32.319 on your toe. These signals have to travel along nerve cells, 0:12:32.840 --> 0:12:36.560 and they move about a meter per second in the cortex, 0:12:36.920 --> 0:12:41.839 which is, by the way, about three hundred million times 0:12:42.000 --> 0:12:47.480 slower than electricity moving through your laptop. We are giant 0:12:47.600 --> 0:12:51.640 systems of cells, and it takes time for impulses and 0:12:51.720 --> 0:12:55.440 cells to travel around. Yes, they use electricity, but it's 0:12:55.480 --> 0:12:59.360 not like a signal propagating along a wire. Instead, with 0:12:59.400 --> 0:13:04.000 a cell, you've got these long extensions called axons, and 0:13:04.080 --> 0:13:07.640 the signals travel by causing little channels to open in 0:13:07.640 --> 0:13:11.000 the membrane, which allows little charged particles to flow through 0:13:11.360 --> 0:13:15.480 and change the voltage locally, and that propagates down the axon. 0:13:15.920 --> 0:13:18.480 So this is a very cool way that mother nature 0:13:18.520 --> 0:13:21.960 discovered how to run a signal down a cell. But 0:13:22.040 --> 0:13:25.480 it ain't fast, and the consequence is that it just 0:13:25.520 --> 0:13:29.360 takes a long time for signals to propagate through the 0:13:29.400 --> 0:13:34.840 system and eventually come together and settle into a coherent pattern. 0:13:35.520 --> 0:13:39.600 So by the time you become consciously aware of something 0:13:39.640 --> 0:13:44.120 in the outside world, the event has already happened a 0:13:44.120 --> 0:13:48.840 while ago. We live in the past. For example, clap 0:13:48.880 --> 0:13:51.760 your hands in front of you. By the time you 0:13:51.920 --> 0:13:55.120 see and feel and hear that it's already happened a 0:13:55.160 --> 0:14:00.439 tiny little while ago. Whatever conscious movie you're seeing right now, now, 0:14:01.120 --> 0:14:02.640 that world is already gone. 0:14:03.600 --> 0:14:03.760 Now. 0:14:03.800 --> 0:14:07.760 We don't often think about this, but this delay from reality, 0:14:07.800 --> 0:14:10.040 the fact that we're living in the past, is a 0:14:10.080 --> 0:14:14.360 major problem because you need to operate in the present, 0:14:14.920 --> 0:14:19.480 but your brain is always working with old news. All 0:14:19.640 --> 0:14:23.120 your sensory inputs like vision and hearing and touch, these 0:14:23.280 --> 0:14:27.000 take time to travel to the brain to get processed, 0:14:27.000 --> 0:14:30.360 and finally the brain croaks out a response. And even 0:14:30.400 --> 0:14:33.760 though this delay is less than a second, that's plenty 0:14:33.800 --> 0:14:37.280 of time to create issues. So just think about trying 0:14:37.320 --> 0:14:40.880 to catch a baseball that someone throws to you. If 0:14:40.920 --> 0:14:45.160 you were merely an assembly line device, you couldn't do it. 0:14:45.280 --> 0:14:48.800 Why because there would be a delay of hundreds of 0:14:48.840 --> 0:14:52.440 milliseconds from the time the light strikes your eyes until 0:14:52.480 --> 0:14:54.840 you could put up your glove in the right spot. 0:14:55.400 --> 0:14:58.160 And the problem is that by the time the image 0:14:58.200 --> 0:15:01.240 of the ball reaches your brain and gets processed, the 0:15:01.280 --> 0:15:04.760 ball has moved. Your hand would always be reaching for 0:15:04.840 --> 0:15:07.160 a place where the ball used to be. 0:15:08.240 --> 0:15:09.560 So how do you catch a baseball? 0:15:09.600 --> 0:15:15.360 It's because of these deeply hardwired internal models. Your internal 0:15:15.400 --> 0:15:20.360 model generates expectations about when and where the ball's going 0:15:20.440 --> 0:15:24.080 to hit, given momentum and gravity and so on. Your 0:15:24.080 --> 0:15:29.560 brain is not just passively processing information. It's predicting. It's 0:15:29.560 --> 0:15:31.320 not reactive, it's. 0:15:31.200 --> 0:15:33.040 Constantly guessing ahead. 0:15:33.600 --> 0:15:36.360 It predicts where the ball is going to be based 0:15:36.400 --> 0:15:39.800 on clues about its trajectory and speed, and that's what 0:15:39.920 --> 0:15:42.800 allows you to catch it. By the way, as a 0:15:42.840 --> 0:15:46.480 side note, these predictive internal models you have are trained 0:15:46.560 --> 0:15:52.640 up by lifelong exposure in your normal experience. If your 0:15:52.920 --> 0:15:57.600 great grandkids grow up on Mars, their internal models will 0:15:57.600 --> 0:16:00.680 get trained up with different parameters of physics, and they'll 0:16:00.720 --> 0:16:04.000 put up their glove at a different time the moment 0:16:04.040 --> 0:16:07.720 that's right for a Martian pop fly. Okay, But the 0:16:07.760 --> 0:16:09.640 critical point I want to make here is that we 0:16:09.840 --> 0:16:14.080 have these predictive internal models, and these things tell us 0:16:14.120 --> 0:16:18.920 from experience how things are likely to move in the world. 0:16:19.080 --> 0:16:21.960 And this way our brains don't work solely from the 0:16:22.160 --> 0:16:26.440 latest sensory information, but instead they construct predictions about where 0:16:26.520 --> 0:16:29.360 the ball is going to be. The same idea is 0:16:29.360 --> 0:16:32.520 in play when you're walking through a busy airport, when 0:16:32.560 --> 0:16:35.119 you have a flow of people moving in all directions 0:16:35.160 --> 0:16:39.400 around you. If you had access to only outdated information 0:16:39.920 --> 0:16:43.840 from photons a few hundred milliseconds ago, you'd be constantly 0:16:43.880 --> 0:16:44.920 crashing into people. 0:16:44.960 --> 0:16:47.360 But you don't. Your brain solves this. 0:16:47.880 --> 0:16:51.480 Your brain is constantly forecasting where the people are going 0:16:51.520 --> 0:16:54.680 to be based on their speed and direction, and that's 0:16:54.720 --> 0:16:59.480 what allows you to smoothly navigate without crashing despite the 0:16:59.520 --> 0:17:04.400 neural the and processing the visual information. So I want 0:17:04.400 --> 0:17:07.400 to summarize where we are so far. The foundation we're 0:17:07.520 --> 0:17:10.400 establishing here is that the brain is not just reacting 0:17:10.400 --> 0:17:13.680 to the world. Instead, it's a machine that continuously makes 0:17:13.880 --> 0:17:19.679 educated guesses. Prediction is how we compensate for our signal 0:17:19.760 --> 0:17:23.919 processing delays, and from an evolutionary standpoint, this ability to 0:17:24.000 --> 0:17:28.520 predict was absolutely critical for survival because animals who wanted 0:17:28.600 --> 0:17:32.720 any chance of living how to anticipate the movements of 0:17:32.800 --> 0:17:36.439 predators or prey to react quickly enough. You have to 0:17:36.520 --> 0:17:40.200 somehow operate in real time if you want to evade 0:17:40.280 --> 0:17:43.840 a thread or catch the running animal. So whenever you 0:17:44.080 --> 0:17:46.480 are next catching a ball or moving through the airport, 0:17:47.000 --> 0:17:50.920 think about how much you rely on your brain's predictive 0:17:51.080 --> 0:17:55.399 abilities to act without having to wait for all the 0:17:55.480 --> 0:17:59.480 signals to dribble their way in there. Okay, now we're 0:17:59.560 --> 0:18:02.280 finally ready to return to the issue that I started with. 0:18:02.480 --> 0:18:05.320 The illusions where you have some lines that are straight 0:18:05.400 --> 0:18:08.560 but they look bent. These fall into the category of 0:18:09.160 --> 0:18:12.399 geometric illusions. So what in the world do they have 0:18:12.480 --> 0:18:17.000 to do with what we've been talking about so far. Well, 0:18:17.359 --> 0:18:19.560 what I told you is that the visual system has 0:18:19.600 --> 0:18:23.160 developed these predictive mechanisms to deal with the signal delays 0:18:23.560 --> 0:18:25.879 so that it can see something at this moment in 0:18:25.920 --> 0:18:28.560 time and make a really good guess where that thing 0:18:28.640 --> 0:18:32.439 is going to be in say, one hundred milliseconds. So 0:18:33.080 --> 0:18:38.160 some of my colleagues proposed a framework called perceiving the present, 0:18:38.720 --> 0:18:42.119 and the idea is that your brain sees what is 0:18:42.400 --> 0:18:46.199 likely to be the case, rather than to perceive the 0:18:46.280 --> 0:18:49.679 recent past. So the first examples of this framework came 0:18:49.720 --> 0:18:53.440 out in the early nineteen nineties. So imagine you're looking 0:18:53.480 --> 0:18:56.399 at a small horizontal line on a computer screen and 0:18:56.440 --> 0:18:59.439 you're trying to judge its exact position, but there's a 0:18:59.480 --> 0:19:03.640 field of dots drifting continuously in the upward direction behind 0:19:03.760 --> 0:19:06.720 that line. In this case, you'll judge the line to 0:19:06.760 --> 0:19:10.480 be higher up on the screen. This is called motion capture. 0:19:10.760 --> 0:19:13.400 So by the beginning of the two thousands, my colleague 0:19:13.480 --> 0:19:18.120 Mark Chengizi started proposing that the explanation for this motion 0:19:18.240 --> 0:19:23.080 capture was the perceiving the present framework, which is that 0:19:23.119 --> 0:19:26.480 your brain sees the line, and it sees the motion 0:19:27.040 --> 0:19:30.199 and decides that in the next moment the line is 0:19:30.320 --> 0:19:33.320 probably going to be pushed up by the motion. So 0:19:33.400 --> 0:19:37.679 it's actually perceiving the line in a different place where 0:19:37.920 --> 0:19:41.840 it expects the line to be in the next moment. 0:19:42.600 --> 0:19:46.880 And besides that, he argued, he could explain the classical 0:19:47.000 --> 0:20:08.560 geometric illusion. What are these classical geometric illusions. Well, let's 0:20:08.560 --> 0:20:12.560 take what's known as the Herring illusion. You almost certainly 0:20:12.600 --> 0:20:14.919 saw this as a kid. There are a bunch of 0:20:15.000 --> 0:20:17.440 lines coming out of the center, like the spokes on 0:20:17.480 --> 0:20:21.920 a bicycle wheel. Okay, now you put two parallel lines 0:20:22.359 --> 0:20:25.520 on that bicycle wheel, let's say, a vertical line to 0:20:25.560 --> 0:20:28.320 the right of center and one to the left. You 0:20:28.400 --> 0:20:32.240 could do this by taping two pencils on the bicycle spokes. 0:20:32.720 --> 0:20:36.560 Now here's the illusion to two pencils. Although they are straight, 0:20:36.960 --> 0:20:39.199 they don't look that way anymore. Instead, it looks like 0:20:39.240 --> 0:20:43.560 the pencils are curving, they look slightly bent. Their middles 0:20:43.680 --> 0:20:47.480 are bowing outwards slightly. So this is an illusion that 0:20:47.560 --> 0:20:52.080 was first described by the physiologist Ewald Herring in eighteen 0:20:52.160 --> 0:20:52.760 sixty one. 0:20:53.119 --> 0:20:55.040 But why in the world does it happen? 0:20:55.200 --> 0:20:57.320 Well, Herring proposed that this has to do with the 0:20:57.359 --> 0:21:01.520 brain overestimating angles where the lines are meeting. And then 0:21:01.600 --> 0:21:04.440 other people proposed different things in the brain that might 0:21:04.800 --> 0:21:10.200 explain that angle overestimation. But Changhizi proposed a new explanation 0:21:10.280 --> 0:21:14.520 when which was quite stunning. He said, look, when you're 0:21:14.560 --> 0:21:17.520 looking at these radial lines, in other words, the lines 0:21:17.640 --> 0:21:20.119 like the bicycle spokes coming out from a central point, 0:21:20.960 --> 0:21:24.280 your brain might think that it's just looking at the 0:21:24.400 --> 0:21:29.480 convergence of lines to the vanishing point, like imagine you're 0:21:29.520 --> 0:21:32.040 looking straight ahead on a street in New York City 0:21:32.280 --> 0:21:36.000 and everything converges in the middle. But equally, he said, 0:21:36.080 --> 0:21:38.879 these lines are what a scene looks like to your 0:21:38.960 --> 0:21:43.480 visual system when you are moving forward. For example, imagine 0:21:43.480 --> 0:21:46.040 that you're driving down the road in New York City 0:21:46.480 --> 0:21:48.680 and up ahead on the left there's a hot dog 0:21:48.800 --> 0:21:51.119 stand and that zips by you on your left side, 0:21:51.440 --> 0:21:53.840 and at the same time, there's a street juggler on 0:21:53.880 --> 0:21:54.520 your right. 0:21:54.320 --> 0:21:56.399 Side, and he gets bigger and he zips by you 0:21:56.520 --> 0:21:56.840 on that. 0:21:56.920 --> 0:21:59.840 Side, and there's an overhead street sign that at a 0:22:00.040 --> 0:22:03.240 distance starts essentially in the middle in front of you, 0:22:03.400 --> 0:22:05.919 but as you get closer and closer, it moves over 0:22:05.960 --> 0:22:11.919 your head. So everything is streaking past you like radial lines, 0:22:12.359 --> 0:22:15.399 and this is known as optic flow. So one place 0:22:15.480 --> 0:22:18.680 you've seen this before is on Star Trek, where they 0:22:19.080 --> 0:22:21.439 yank down the lever and put the ship into warp 0:22:21.560 --> 0:22:25.440 drive and all the stars suddenly shoot past them, all 0:22:25.600 --> 0:22:29.119 moving away from the center, like the radial spokes of 0:22:29.160 --> 0:22:32.720 the bicycle wheel. So Tanghizi said, when you see radial 0:22:32.840 --> 0:22:37.159 lines like that, it's typically a visual signature of you 0:22:37.840 --> 0:22:41.520 moving forward towards the vanishing point. And certainly when you're 0:22:41.520 --> 0:22:44.760 moving fast, there's a radial smear, like the way that 0:22:44.800 --> 0:22:49.200 the stars and Star Trek smear into lines. And he said, look, 0:22:49.600 --> 0:22:54.000 Herring's radial lines essentially mimic this. It's like you are 0:22:54.040 --> 0:22:58.119 in the spaceship moving directly ahead. Now here's the key. 0:22:58.880 --> 0:23:01.359 Let's come back to the de lays in the visual 0:23:01.400 --> 0:23:04.560 system and how they can be accounted for by the 0:23:04.720 --> 0:23:10.200 brain making projections where things are about to be. Imagine 0:23:10.200 --> 0:23:12.840 that you're in New York City and driving and there 0:23:12.840 --> 0:23:15.359 are two skyscrapers up ahead of you, one on the 0:23:15.440 --> 0:23:17.680 left and one on the right. Now, as you race 0:23:17.840 --> 0:23:22.160 forward in your car, those two buildings will loom closer. 0:23:22.200 --> 0:23:25.280 But now something interesting is happening. The parts of the 0:23:25.320 --> 0:23:30.000 buildings closer to you will seem farther apart, because if 0:23:30.080 --> 0:23:32.800 you look up, the tips of the buildings are coming 0:23:32.880 --> 0:23:34.639 closer together, way up in the sky. 0:23:35.160 --> 0:23:36.320 So the point is that. 0:23:36.520 --> 0:23:40.199 Even though you see essentially straight skyscrapers when they're at 0:23:40.200 --> 0:23:44.640 a distance, as you approach, they are bending away from you. 0:23:44.960 --> 0:23:50.720 Their centers are bowing out. And Shannghizi's idea was that 0:23:50.760 --> 0:23:54.000 when you look at the radial lines the bicycle spokes, 0:23:54.000 --> 0:23:56.240 your brain thinks this might be a clue that I'm 0:23:56.280 --> 0:23:59.240 moving forward, and I don't want there to be delays 0:23:59.280 --> 0:24:02.439 in my perception, so I'm going to see the world 0:24:02.680 --> 0:24:06.919 as it will be a moment later. And so you 0:24:07.000 --> 0:24:10.959 see the two parallel lines bowed outward from the center. 0:24:11.359 --> 0:24:13.439 In other words, when you look at the radio lines 0:24:13.440 --> 0:24:16.119 on the piece of paper, even though nothing is moving. 0:24:16.920 --> 0:24:20.040 Your brain thinks this is what movement looks like, and 0:24:20.119 --> 0:24:24.359 so it predicts the next moment, and that's what you perceived. 0:24:24.520 --> 0:24:26.600 In other words, you perceive the lines just as they 0:24:26.720 --> 0:24:29.320 would project in the next moment if you were moving 0:24:29.400 --> 0:24:33.480 forward toward the vanishing point. So, as Changhesi wrote in 0:24:33.520 --> 0:24:38.680 this paper, evolution has seen to it that geometric drawings 0:24:38.840 --> 0:24:45.400 like this elicit in us premonitions of the near future. Okay, 0:24:45.760 --> 0:24:50.000 so the framework by Changizi and colleagues suggests that several 0:24:50.040 --> 0:24:54.199 geometric illusions are caused by temporal delays with which the 0:24:54.280 --> 0:24:57.040 visual system must cope. The idea is that the visual 0:24:57.040 --> 0:25:03.000 system extrapolates its current information to perceive the present. Instead 0:25:03.000 --> 0:25:06.399 of providing a conscious image of how the world was 0:25:06.440 --> 0:25:09.600 a few hundred milliseconds ago when the signals first struck 0:25:09.640 --> 0:25:13.000 the retina, the visual system estimates how the world is 0:25:13.240 --> 0:25:16.399 likely to look in the next moment. But how would 0:25:16.400 --> 0:25:20.520 we get at clues to the possible neural basis? In 0:25:20.520 --> 0:25:23.240 other words, how does the brain actually pull this off? 0:25:23.960 --> 0:25:26.280 So in my laboratory we wanted to figure this out. 0:25:26.520 --> 0:25:29.840 So my student Don Vaughan and I had people look 0:25:29.920 --> 0:25:32.679 at the herring illusion on a screen. You've got a 0:25:32.720 --> 0:25:35.960