Are you refering to a real life situation or observing the wheel on tv.
There are two seperate reasons.
I can tell you what we've done on this. So, this is a debate that goes back over a century, where people have made the assumption that maybe the visual system acts like a movie camera and takes snapshots of the world. And this view of the brain was inspired by cinematography, and this happens commonly where a new technology gets invented, and people think that maybe the brain works that way, that's certainly what happens with digital computers as well, now everyone thinks that the brain works like a digital computer, which is also not true. People have been suggesting for a really long time that maybe the brain does this. And everyone had noticed the wagon wheel effect in movies, which comes about because you're taking discreet snapshots of a spinning wheel, and if the wheel spins most of the way where spoke one is now, almost of the way to where spoke two was, when you see the succession of photographs, you'll see that the shortest distance would be if spoke two had traveled in the backwards direction. So it looks like the wheel in movies spins the other way. What people noticed is that if you watch something in constant sunlight -- you're ruling out the stroboscopic effect of lights in you're building -- you can get something that's like a reversal. And people suggested that, ahh, this must mean that the visual system is like a camera. In fact, really, all of the big thinkers in the field have been citing this and claiming this for a little while. But a couple of things struck me. The way that reversals happen under constant sunlight is actually very different from the wagon wheel effect in movies. For one thing, you have to watch a spinning object for a little while before you see it go backwards, and the reversals are sporadic. So, in other words in the movies, if you have a wheel moving at a constant speed, it'll look as if it's going backwards. But in real life, if you watch something that's spinning at constant speed, let's say like a ceiling fan overhead, most of the time it's running in the normal direction, and only sporadically, just for a moment, you'll see it look like it's running backwards. So that's the first thing that's different. And also, in real life, you never see something come to a stop. Whereas in the movies, since you have a constant frame rate, you can spin a wheel at just the right speed and it'll look like it stops. So that can't happen in real life. So that's when I started suspecting that these are actually real phenomena. I did a very simple experiment to decide if in fact the visual system was taking something like snapshots or whether I could rule that out. Here's what I did. I had a spinning drum, and I had 16 dots evenly spaced around the drum. So as the drum spins, you see the dots coming past you, moving to the right. And you watch these dots for a while. I have subjects hold down one key when they're seeing the dots move in the normal direction, and then they hold down a different key when they see them appearing to reverse. And this way I'm able to quantify exactly when and how long they see these reversals. The first thing that I noticed is that the distribution of reversal time that people had, in other words, how long they see it reversing and how often. I noticed that that approximates a mathematical distribution that's the same type of distribution that you see in what's called rivalry. Rivalry is when you have two different interpretations that your brain can make of some sensory event, and they compete with each other. So, for example if you draw a cube, just the wire frame of a cube, 12 lines that make up the wire frame of a cube, you can see it coming out of the page in one direction, or in an other direction. You never see both at once, instead there's rivalry. Your brain has two, in this case, equally probably interpretations, and it will see one for a little while, and then it will switch and see the other interpretations. So that's called perceptual rivalry. I started wondering if maybe, when you're watching the drum, when you have dots moving to the right, you're activating the rightward motion detectors in your brain. But it just so happens that for technical reasons, that you're also tickling leftward motion detectors also. So I thought that maybe what this is just a rivalry between rightward and leftward motion. So most of the neurons in your brain that care about motion are telling you that you're seeing rightward motion, but some of them are telling you that you're seeing leftward, and once in a while they'll win out temporarily, and you'll end up seeing leftward motion. So to test this, I did a very simple experiment. I had people watch a drum, and I set up a mirror, so that they were now seeing two identical drums. So they're seeing two drums that have exactly the same characteristics -- they spinning at the same speed, same dots, but they're spinning in opposite directions, one clockwise and one counterclockwise. But the point is this. If we were taking snapshots of the world, then what you would expect is that when one drum reverses, they both reverse. By the way, the snapshot hypothesis requires that your frame rate changes. Sometimes it's going faster , sometimes it's going slower , that's a detail, but the frame rate hypothesis requires that to explain why you only see the reversal sporadically. But the point is that if you're really taking snapshots of the world, then both drums should always reverse at the same time, because that's what would happen in the movies. So, what I had people do is hold down one button if the right drum was reversing, one button is the left one was reversing, one button if both drums were reversing at the same time. And what I found was that, almost all of the time, the reversals of the two drums were independent. In other words, they don't reverse at the same time. So what appears to happen is that both drums independently have rivalry with the different interpretations. So the drum on the right is most of the time going clockwise, and every once in a while, because of rivalry there, it can appear to go the other way. The same with the drum on the left, it can appear to switch directions sometimes. But, it's independent. It's object based. So, it's just a very simple experiment to rule out this idea that we're taking snapshots of the world. Now one possibility, one remaining possibility was that since I had people fixate on a point right in between the two drums, one could still argue that each of the two hemispheres of the brain has their own frame rate, so to speak. So they're both taking snapshots independently. So what I did to address that was that I turned the whole contraption sideways, so that now the two drums, in other words the drum and its mirror image, are above and below each other, and in the same visual hemifield, the same side of the visual world. And I had people fixate on a point so that both of the drums were off to the right in the same visual field. And the same thing happens -- that is to say that the drums do not reverse together. And again that rules out the idea that we're even taking hemispheric snapshots. In other words, it's not the case that any part of the brain is taking snapshots. Instead it seems that the explanation of the effect is something that we're familiar with from other illusions, for example, the cube wire-frame that I mentioned, which is that you just have a competition between two competing interpretations that your brain can make. As I said, most of the cells in your brain that care about motion are streaming off that there's rightward motion, but because there's a periodic pattern that happens to tickle leftward motion detectors also, you just have a rivalry. And that's why you see the reversal under continuous sunlight. Now there's something else that happens under fluorescent lamps, even though most of us can't tell just with the naked eye, the fluorescent lights in a room, and the sodium lights in a room have a 60 hertz flicker, in other words, 60 times a second they're popping on and off. And for most of us, our eyes can't detect that. But, it turns out that when people see spinning hubcaps on a road, it's important to note whether that's during the night or the day. Because at nighttime, the flicker from the sodium lamps can induce this illusion, and that is a stroboscopic effect. That is, because there's a flicker from those lamps, and it really is more like the wagon wheel effect in the movies. But during the daytime, what you'll notice is that the reversal of the hubcaps is sporadic. It only happens once in a while. And you usually have to stare at a hubcap for a while to see it happen. And in that case, what you're seeing is the perceptual rivalry between the two interpretations.
It's exactly because the cells in your brain that care about detecting motion are wired up in a very specialized way. And when you see somebody walking towards the right, all of the signals will essentially get interpreted the right way by your brain. A periodic stimulus, in other words, one that's a pattern that repeats, there's, again for technical reasons, that plugs into your motion detectors in a way that fools them a little bit. They're not really specialized for detecting periodic patterns, and as a result they can get fooled by it.
Let's imagine you had a single dot moving towards the right. It would first activate a position towards the left , and then it would activate some spatial position on the right, because it's moving rightward. So it hits a position on the left, and then it hits a position on the left some short time later. So your brain interprets that as rightward motion. But, if you have a periodic pattern moving toward the left, you have one dot that hits the position on the left, and then you have a second dot that hits the position on the right, a short time later. And as a result, that activates the rightward motion detector. It can't tell that those were two separate dots. All it knows is that the left, and then the right, were activated in order. It doesn't know that it was activated by two separate dots in a periodic pattern. So, I hope that's clear. There's a figure of that in my paper. For that reason, the periodic pattern will tend to fool the system, whereas somebody moving to the right gives very clear signals to your brain.
Our brains come equipped with circuits of cells that are specialized for detecting motion. And, in fact these things are so specialized that their activity will tell you that you're seeing motion, whether or not you really are. So, for example, the motion-after effect can be demonstrated by watching a waterfall for a couple of minutes, and then looking away from the waterfall, everything appears to be crawling up. And that's because you're now activating these motion detectors in the opposite direction, and just their activity will tell you that you're seeing something moving, even though there's actually nothing moving in the real world. So you have these circuits that will tell you that you are seeing motion. So, a rightward motion detector looks for something that hits position (1) at time (1), and then looks for something that hits position (2) to the right of that at time (2). And if it sees , position (1) an (2), with some time delay between them, it is built to interpret that as something moving to the right. And almost all of the time that is correct, it's doing the right thing. It just so happens that if you have a repeating pattern that's moving to the left, it will activate that circuit. Because you'll have one of the dots of the pattern activate position (1) at time (1), and then a different dot on the pattern will activate position (2) at time (2), and your brain has no way of determining that these were different dots, it thinks that it's just seeing something moving to the right. And that's why a periodic, leftward moving pattern can tickle rightward moving detectors. And that's what's going on here. That's why, if you look at a pattern that's moving to the left, some of your rightward motion detectors are also getting activated, and now you have a competition, or rivalry between these. And most of the time the leftward motion detectors are going to win. But some of the time, the rightward motion detectors are going to win out temporarily, and what you see is an illusory reversal of motion. And of course it's the same thing as when you're watching a ceiling fan, except now it's clockwise versus counterclockwise detectors.
I think the right way to do it is just to say, that for technical reasons, these get tickled by a periodic, repeating pattern.
It's a lot longer that that actually, it's about 200 milliseconds. In other words, 1/5 of a second. It takes a lot longer to react. In fact, at the Olympics, if you launch off the blocks within 180 milliseconds of the gun, you get disqualified, because the Olympic Committee knows that you cannot possibly react faster than that. So in other words, it's a false start if you launch after the gun, but within 180 milliseconds after the gun, they know that it's a false start.
It's completely wrong that we take continuous snapshots of the world. There are many times when your brain tries to construct the answer to a question that it's asking, like, where is that red car right now. When your brain tries to construct and answer, it takes something like 100 milliseconds to construct that answer. So in other words, things don't happen instantaneously in the brain, there's a lot of signal processing time before an answer is reached. So, in a sense our world is chunky in time. Things aren't smooth. But that's the only sense in which there's even a relationship to cinematography, the way things are chunky in time. But other than that, it is not the case that our brains are anything like a movie camera.
Perceptual rivalry is a well-known phenomenon. The idea of perceptual rivalry is something that is well known in the field and totally excepted -- and that is, the idea that your brain is always making interpretations of things. If I hold up a coke can for you,, I might be holding up a regular coke can five feet from you, or it can be a coke can that's a mile away and 50 feet tall. And those will make the same pattern on your retina. It will make the exact same pattern on your retina. And so, your brain is always making interpretations of things that are inherently ambiguous. So, what happens is that if I set up a stimulus that has multiple interpretations, your brain then has these different interpretations competing. And, as I said, the wire frame cube is an example that everybody has seen in their own life. But there are many other examples of this as well, something that people haven't experimented on themselves, typically, but what you can do is something that's called binocular rivalry, where you put two different images under two different eyes. So one eye is seeing a house, and one eye is seeing a face, and it turns out that you don't see a fusion of the two stimuli, instead you have rivalry, where you see a face, and then it switches and you see a house, and then it switches and you see a face again. So that's an example of rivalry as well. There are many stimuli that have ambiguous interpretations, and what your brain does is typically land on the most probably one -- but if there are multiple interpretations, that are equally probable, or close to equally probable, then your brain will switch back and forth between them.
Yes, but I can't think of a good example. You can have auditory rivalry if there's a sound that can have multiple interpretations to it. Your brain could interpret the sound one way, and then suddenly hear it the other way.
Tell them that you have to keep the lights off, so this works best, of course, during the daytime, so if they have sunlight streaming into their house, and all of the lights are off, but they have the ceiling fan running at a slow speed, I mean they can try it at different speeds, but at a slow speed, yes, you can see it reversing. Now one thing this requires is patience, for myself, I usually have to watch for something like eight minutes before I see this happen. So it requires patience, but it's extraordinarily clear when it happens, you'll be absolutely certain when it happens because suddenly you'll see the fan, just for a second or so, running the other direction. And it's extraordinary when you see it, you can't miss it. So yes, you can demonstrate this easily to yourself at home. One thing that I've noticed is that some fraction of people, maybe 10% of people, don't seem to see it, and I don't have an explanation for that. But everyone else seems to, and maybe those 10% just give up too early, but yes, for just about everybody, with patience, you can easily demonstrate this for yourself with a ceiling fan.
Essentially, what we're trying to do is understand vision. We're trying to understand how you ever string ten billion neurons together and get something out of it like vision. And people have been trying to do artificial vision, or computational vision since the 60s, and it's been a total failure. We haven't yet figured out how vision works. And you've got about 1/3 of the human brain devoted to it, it's this enormous construction that the brain does, and we're still at the level of trying to figure out the basic, fundamental principals by which the visual system operates. So, understanding whether it's taking snapshots of the world, or whether there are other explanations for certain illusions is fundamental for understanding how the visual system operates. And illusion tends to be very powerful windows into how the brain operates. And these things that you can demonstrate in your own living room tend to be very deep clues into what's going on under the hood.