 Hi, and welcome back to this video course on biological psychology in this video video 3.3 We're going to take a look at the vision or visual perception Now let's start at the very basics. What is vision? Well vision is sensation of light as you probably know Now and what is light light is electromagnetic radiation and visible light so the part of the electromagnetic spectrum that we can see is only a very small part of the entire electromagnetic spectrum and that's what you see here in this figure. So basically this area right here these wavelengths so that basically the property of electromagnetic magnetic radiation is the wavelength and if the wavelength becomes longer we perceive it as being more Reddish if the wavelengths becomes shorter we perceive it as being more bluish purplish and you see that if things become more blue than blue We start to perceive ultra or we don't actually perceive the their electromagnetic Magnetic radiation becomes ultraviolet so sort of purple that is too purple for us to see And if the wavelength becomes longer than we are able to see it becomes infrared So more red in a sense than we are able to see right so and you see there is no qualitative difference really between other forms of electromagnetic radiation such as gamma rays or x-rays or radio waves or microwaves All of these are simply electromagnetic magnetic radiation just like visible light They just have a different wavelength and you also see here You see the honeybee and you see that the honeybee like us can see visually But it sees in a different part of the of the electromagnetic spectrum So the honeybee can see things that to us are infrared Now so how does light lay lead to brain activity? So light falls through the lens in our eye So here you see a picture of the eye Viewed from above and here you have the lens now light comes from outside Passes through the lens and then focuses on the retina here in the back of our eye And the retina is a part of our eye that contains photoreceptors So those are cells nerve cells that are responsive to light. We will look at the photoreceptors in more detail in a bit and And photoreceptors transfer nerve impulses to the brain So here we have the retina and in the retina there are photoreceptors now these photoreceptors Trigger action potentials and these go through axons here through the optic nerve towards the brain And you could say that our eyes and our photoreceptors in a sense are part of the brain Sometimes people say that obviously it's a bit ambiguous because what does it mean to be part of the brain? Right, but at least our eyes are so closely and directly connected to the brain that it makes in a way sense to talk about them as though They are part of the brain Now so let's trace back Let's trace what happens to these nerve impulses in a bit more detail. So here we see again a Schematic of the eyes, but now with the whole brain included So when light falls through the lens onto the retina the nerve impulses leave our Our eyes through these optic nerve and then they arrive they are transferred to the back of the brain to the occipital cortex So this area right here This is the occipital lobe and it contains what is called the primary visual cortex or v1 So every sense has a primary cortex, right? We have a primary auditory cortex where auditory information is processed first We will see that in a later video and also for vision. We have a primary visual cortex So that is essentially the first input station that the first part of the cortex that receive receives visual information Now So what happens to visual information in the in the primary visual cortex? Well, this was demonstrated in groundbreaking work in the 1950s by Hubelem Wiesel who studied neurons in v1 and They demonstrated so-called simple cells. Now. What does it mean to study neurons in v1? So they Hubelem Wiesel literally opened up the skull of cats and anesthetized cats and they inserted electrodes in the brain of these cats in the occipital cortex and they were able to isolate individual neurons and Then what they did is essentially measure the action potentials of these individual neurons. So they just Figured out through trial and error what caused these neurons in the visual cortex to become active And then what they found were particular properties that they labeled simple cells Now simple cells are cells that respond only to a specific orientation and a specific location I will demonstrate that in a bit more detail So what if you want to take a look at a very cool video? I recommend that you follow this link is on YouTube where you can actually see the experiments that you Hubelem Wiesel did in in With actual footage, which I believe that has been recorded by Hubelem Wiesel themselves I cannot include it here in this this presentation because of copyright limitations But I recommend that you that you watch that video because it's really quite cool And if you're worried that it's going to be gruesome because it has to do with electrodes in the brain It's not you will not see any of that. So you can safely watch it Now so what does it mean? What does what is a simple cell? So here we have a hypothetical simple cell and This simple cell has a so-called receptive field which I've indicated here by this this thought cloud For better or worse and what this means a receptive field is that this simple cell becomes active When there is visual stimulation in a particular part of the visual field So you could say for example whenever there's something appears a little bit in the top left Then this neuron becomes active, but not any kind of visual information causes this neurons to become active It becomes specifically active when there is a Rightward tilted bar in its receptive field So you could say colloquially this that this neuron looks at a small part of the world And that part of the world that it looks at is its receptive field And it monitors that small part of the world for the presence of these Tilted gray bars Now neuron now there is these tilted bar present in the receptive field. So this neuron becomes active Now, but if the same tilted bar would be presented somewhere else the neuron would not become active And if a tilted bar would be presented in its receptive field, but it would be tilted in a different way This neuron would also not become active. So the neuron is very simple very selective for Bars that are tilted like this in its receptive field. And those are the properties of a simple cell that Hubel and Wiesel demonstrated Now a very important concept in brain Function in general is the concept of topography So what that means is that the world in front of us is organized in a particular way, right? So in terms of vision the Main way that the world is organized is in space, right? Something can be left to something from the left to something else or on the right of something else For example for sound the main way in which the world is organized is in terms of pitch Something can be high or low pitched etc. But there was always some kind of organization in the world. The world has some kind of topography Now and the way that our brain represents the world Generally speaking preserves some aspect of this topography of the world And that sounds very abstract But I can make it more more concrete by talking about the specific kind of topography that is important in vision And there the topography that's important is retinal topography or retinotopy What this means is that adjacent parts of the retina, so parts of your of your retina in your eye that are next to each other Are processed by neurons that are also next to each other in your primary visual cortex So you see how that preserves the spatial layout of the world in front of you, right? Things that are near each other in the world outside are processed by brain areas that are near each other in your brain And that is the concept of retinotopy. It's a very important concept and we will see different forms of topography for different senses in later videos Now we've looked at simple cells And simple cells are a very early step in visual processing you could say it it it reflects some some form of processing of visual information Right? There's some selectivity, but it is it is very basic And this is a property of this shows that the visual system processes visual information in a hierarchical way What this means is that progressively Cells that are connected to each other start processing progressively more complicated Properties of the visual input So the first essentially the most simple cells that That process vision are mostly responsive to contrast, right? So differences in luminosity Then we have the cells that are also Sensitive to orientation. So whether something is tilted clockwise or counterclockwise like the simple cells that we've looked at Now any information from the simple cells then feeds into another group of cells that processes even more complicated forms of Visual properties such as shape for example, whether something is Rectangular or whether there's a particular angle or corner in the shape, etc And in this way in a hierarchical way with neurons that feed into other neurons that feed into other neurons We progressively analyze more complicated aspects of the visual input And then finally Our visual processing becomes so complicated that we have for example object recognition, right? So what does object recognition now? It means simply that if I here I have a mouse And I can look at this mouse and I can tell you that it's a mouse. I've recognized the object As a mouse, right? And that requires a very hierarchical very complicated chain of processing of visual information Of which these simple cells are only one small step Then we have another property of vision namely visually guided action And in our brain visually guided action and object recognition are represented in quite different ways So visually guided action would mean simply that if I look at this mouse Not only do I recognize that it's a mouse, but I'm also able to to to interact with it and to pick it up In a way that's appropriate for this mouse, right? I know how to interact with it and how to put my hand on it, etc And both of course object recognition and visually guided action both rely on visual input, right? But they are different ways in they they are different pathways essentially in our brain Different ways in which visual input is processed and it's also possible for people to for example be unable to recognize what things are A deficit called agnosia that also Oliver sex talks about While they are able to interact with things and conversely it's possible for people to be unable to interact with things A deficit called aprexia While they are still able to recognize those things, right? So keep in mind that object recognition and visually guided action are reasonably distinct capacities of the human brain although both rely on vision Now here below in this picture you see this concept of Of a hierarchy of visual processing, right? So we have an input and that input feeds into one layer of cells and yeah These this layer of cells feeds into another layer, etc, etc, etc This picture is actually not a picture that reflects a human brain, but it reflects an artificial neural network that does object recognition And artificial neural networks are For object recognition are very interesting because they work really well and they kind of mimic the way that our visual brain Visual cortex also works Now while we're On the topic of artificial neural networks Let me show you this picture, which I personally find very fascinating So what is this? This is essentially a neural network simulation of our brain hallucinating you could say well that sounds very generic So I'll try to explain it a little bit more concrete terms what is actually happening here We have a neural network that is able to recognize things, right? So you show a picture to this neural network and a neural network will turn it into say, okay This is a dog or it is a cat, etc Now the neural network has been trained to do so with a particular set of training images And it will recognize whatever is in those training images Now in this is the case of this particular neural network A lot of the training images contain noses of dogs and cats and eyes, etc Things that are popular on the internet and that are therefore in the training set which is taken from the internet, right? Now the trick here and this is very cool. It sounds abstract Try to follow me Is that what they do is they present a picture to this neural network and the neural network processes it and essentially interprets it, right? It turns the sensation of the picture into a perception of the picture With a certain meaning attached to it What they then do they take this and this perception of the picture is of course slightly biased by the way that the network has been trained Because the network will be inclined to see things in the picture that it has been trained to recognize Just like we are inclined to see things that we know And then what they do is they feed the perception of that network So they create a picture out of the perception of the network and they feed it in as the sensation of the network again So they take the output of the network and feed it back in as input Then the neural network will start processing it again And again, there will be this tendency to recognize things in the picture that it knows that they expect and by this Essentially connecting the output of the network to the input of the network Then the neural network will add progressively more Features to the image that it knows about that it expects. You could say it's a kind of confirmation bias We we make the network see what it wants to see Now what does the network want to see in this case? It wants to see these kinds of weird doggy like noses and faces, etc Right kind of weird esoteric stimuli that kind of reflect what it has been trained with The training set of images that it has been trained with that it has been brought up with in a sense I think this is very cool. This particular algorithm is called deep dreams developed by google If you're interested in that and you want to learn more just google it You will find a lot of very beautiful pictures and a lot of fascinating information It's really a revolution in computer vision Now let's go back to the biology. This is of course biological psychology not artificial psychology So let's get back to the concept of photoreceptors So photoreceptors are the things that are these light sensitive cells that are in our retina the back of our eye And we have three kinds of photoreceptors We have the cones of which there are three subtypes. I will explain them in more detail in a bit We have the rods of which there is only one and we have the so-called intrinsically photosensitive retinal ganglion cells Now in most textbooks and including the open stacks textbooks They will only talk about the cones in the rods because they are more most familiar But the intrinsically photosensitive retinal ganglion cells have been discovered quite recently And they are actually a particular kind of photoreceptor that responds more slowly And that is presumably involved in maintaining our day night with So there are three really three different kinds of photoreceptors and not two as you will often hear Now let's first talk about cones in a bit more detail Cones mostly serve central foveal vision. What does that mean? Well, it means say that you have this later you have your retina right with photoreceptors in it The photoreceptors are not distributed equally across the retina and more specifically These photoreceptors these cones are mostly focused in one part of the retina that is called the fovea And when we look at things right like now, for example, I'm looking at the camera What that means basically is that I'm pointing my eyes toward the camera in such a way That the light from the camera falls onto the fovea In other words, I look at things to make sure that light is processed by the cones And that's what it means to The statement that cones serve central or foveal vision Means that that exact thing right that we point our eyes if we point our eyes towards something What we're doing is effectively making sure that light from the thing that we're looking at falls on the cones in our fovea Now cones are not terribly sensitive in the sense that they require quite a lot of light to become active Which means that they dominate in brightness. So if you're out in daylight, the cones will dominate your vision But if you're out at night, the cones will not dominate your vision And cones can distinguish colors in a very specific way Cones can distinguish colors because we have three subtypes and that's what you see here on the right So you have cones that are mostly sensitive to blue light meaning that it will start to fire mostly if light is kind of blue We have cones that are mostly sensitive to green light and cones that are mostly sensitive to red light Now an individual cone is only responsive to one of these things. So it cannot really see color But what happens because we have different cones if for example, all our blue cones start to fire Whereas our red and green cones are mostly silent Then we experience that as blue whereas if the red cones start to fire and the green and blue ones are mostly silent We experience that as red. So the mix of activation of these three different cone types That's what underlies color vision not the fact that an individual cone can distinguish colors Because there are three subtypes Now let's move on to rods Roads mostly serve peripheral vision, right? So right now i'm looking here at this camera Uh and everything that's not the camera that i'm looking at is processed to some extent mostly by rods, right? There are also cones in our peripheral vision, but there are mostly rods in our peripheral vision Now rods are very sensitive meaning that they don't need a lot of light in order to become active So they dominate in darkness, right? So if you're out in darkness your rod your vision will be mostly rod based and not cone based And they cannot see color because we there's only one subtype of rod And it is responsive to kind of bluish color So rods by themselves don't allow us to distinguish colors And this also is the reason why if it is very dark and your vision relies mostly on rods The color seems to disappear from the world, right? The world kind of appears Maybe a little bit bluish or devoid of color But essentially our sensation of color disappears because rod vision does simply not allow us to distinguish color And to the extent that you still see color that is actually them because your cone vision is also Still to some extent present Right because there's only some one subtype so rods don't see color Yes, now now I want to talk about color oponency, but before before I do let's take a look at this very cool optical illusion So what I would like you to do is look at this central fixation cross And then keep your eyes on the central dot and see what happens to the Missing sir missing dots in the circles, right? So you see Let me you see for example here. There's a dot missing You will see this dot will start to rotate and this missing dot will also rotate and this missing dot will also rotate And you will see illusory colors appearing. So look at the central fixation cross And see the illusory colors appear. It will take a few seconds And then you will see in the central ring. You will see a pinkish color appear in the middle ring it will be kind of uh Aqua well green aquaish and in the outer color it will kind of bluish Right and it will be the same for you because everyone has color vision in that same sense All right Now you can do that for a little while. It's a very very compelling optical illusion Let's switch back What's going on here? Um, well the main thing to notice is that we don't perceive colors in an absolute sense Rather we perceive colors relative to other colors, right? So we perceive something as being red in a sense that is more red than it is green and not an absolute number of absolute quantity of red Uh, and this is because colors sensitive neurons inhibit each other. This sounds kind of complicated, but it's not that complicated really So red inhibits green and vice versa And that's what you see here in this this picture on the right So if we have a neuron in our brain, right some cell in our brain that becomes active when you see something that is red And we have another neuron that becomes active when you see something that is green then these neurons inhibit each other So if the red neuron becomes active, it will make the green neuron less active and the other way around In other words There is no greenish red because these neurons are in a sense mutually exclusive, right? Red and green are mutually exclusive colors. If we see one it will extinguish the other color Now and of course you can mix you can for example get a red in a green light and mix it Or have red and green paint to mix it, but we will not subjectively experience that as a color that is somehow intermediate between red and green And the same is true for yellow and blue So if we have a neuron in our brain that responds to yellow now, what is yellow? Actually yellow is for our brain a mixture of red and green, right? Even though for us it doesn't subjectively feel like a mixture of red and green. That's exactly what it is Now and if we have a neuron in our brain that is sensitive to yellow It will reduce activity in neurons that are sensitive to blue and the other way around And that is why we do not see such a thing as a yellowish blue if we mix yellow and blue We will get kind of a greenish color, but that to us is something that is qualitatively different from yellow and blue We don't subjectively experience that as being an intermediate between yellow and blue in the way that for example We most people would subscribe subscribe to the notion that orange is an intermediate between yellow and red Because there is no mutual inhibition between yellow and red. So we can see yellow and red Sort of we can see intermediate forms between yellow and red and we call that orange This is the principle of color oponency Now how does that explain the color adaptation after effect the illusion that you just saw? Well, if we if there is a dot a green dot for a little while say or a red dot say that we're looking at a red dot for a little while Then these neurons that become active because of this red color after a long after some time start to adapt or they become tired They start to fire less Now this will lead to less inhibition of green And therefore when you remove the red we perceive an illusory green You see how this works because we adapt the red neurons we reduce the inhibition of green We create an illusory green if we then remove the actual red color Because simply put green is simply an absence of red Right, so adapting adapting to red will cause things to become more green And that is what causes the color adaptation after effect And I think it's really a victory of vision science that we can explain the color adaptation after effect in such a very concrete way Knowing really how this relates to the the activation of photoreceptors and these neurons these mutually inhibitive Inhibitory neurons in our brain, right? So I think that's very cool There are not many things in our brain that we understand in that way that we can explain it at that level of detail Okay, so that explains the color adaptation after effect Now with that let's move on to the next video video 3.4 in which we're going to take a look at gestalt theory