Now in this next lesson, let's talk about the strange way that we see colors. I introduced you at the outset to a little bit of this, and I'm going to go over it now in a more pointed way to show you that the way we see colors is really strange. I introduced this picture of a cube before just to kind of get you hooked on the idea that the color of this piece of a surface, the orange piece here and the dark brown piece on the surface of the cube, were seen differently when you took away all the information in the scene. So you remember, I masked out all the other color information on the cube and you saw that these two surfaces were actually coming as they are shown here. And I just showed you that to get you interested in the idea, whoa, color is kind of doing the same weird thing in having a discrepancy between what we are physically presented with and what we see in the same way that black and white is obviously discrepant in terms of the perception that we see and the physical reality that instruments measure. So here, I want to extend this a little bit more by describing the phenomena of color contrast and color constancy. So, this phenomenon of seeing the same spectral inputs as differently colored in a natural scene like this. This one looking orange and this one looking brown, when in fact both of them are sort of a different shade of dark tan. That's called color contrast. And people have known this for a long time. There are lots of demonstrations going back to the 19th century that you can see that make this point. There is an opposite effect called a color constancy. Color constancy is taking two different spectra, in this case, two spectra that look different when they're presented without a color context. And putting them into, again, exactly the same color context, but now having these physically different spectra look much more similar. So these now, they don't look identical, of course, but these you would describe both as some kind of shade of orange, whereas here they are very different. This one being orange and this one being brown, again perceptually. These are different, in a sense, opposite effects, that are called contrast, color contrast and color constancy. And what's puzzling about this is that contrast and constancy are coming from exactly the same natural scene. It's not like you had to put these in different color scenes. You just put any two spectra that are the same in this position or any position in this color scene and they will look different. And you take any two spectra that are physically different and put them in this color scene and they will look more similar. So that's constancy and that's contrast, and what's that all about? Why is that happening? It's even more dramatic, or you can make these effects even more dramatic, by changing the scene in other ways. So, let me ask you to take a look at the colors that you see that are elicited by these four service patches, on again, using the same cube but now represented as if it were being seen in a yellow light. And I think you would describe the colors that you see, these four colors that you see here as blue. Those surface patches are blue. But in fact, they are coming from patches that are not colored at all. They are coming from neutral gray patches that take on a blue perceptual appearance when they are shown in this scene. But when they are shown outside of the color information that's in the scene then you see them as neutral gray. So let's take a look at the patches that you see here. As if in a blue light, there are seven of them, and I think you see these seven patches, as I do. Again, if you have normal color vision you see these seven patches as yellowish. And again I don't think you will be surprised at this point to know that they are not really yellow patches, whatever the yellow patch means. They are coming from patches that are grey, that don't have any spectral energy distribution in them. And it's the color surrounds that are giving you this impression of these patches being blue, and these patches being yellow. And that happens, again, as I said before, instantly. It's not some long-term adaptation. Let's look at this again. As soon as I take off these surrounds, these patches that are blue turn to gray. And thee patches that are yellow turn to gray because they are in fact coming from the same gray patches. This is the physical reality that's generating, on the one hand, the perception of blue here, and on the other hand, the perception of yellow here. Now you remember from the description of the circular perception of color, of hue, that around that color circle called the Newton color circles. Newton was really the first one to try and describe colors. Not the first one, but he described colors in the circular fashion as we did. These are color opposites in the Hering sense. Blue and yellow are as far apart as you can get on the Newton color circle. And red and green, and you can do this making the same demonstration in red and green, they're as far apart as you can get on the Newton color circle. So you're taking colors that are diametrically opposed in perceptual color space and making the same gray patch look either blue or yellow just depending on the characteristics that are in the rest of the scene. Again, this is not anything cooked up. This is the way we see the world of color. And it raises a deep question. I asked the question before, but let's ask it again. What's going on here? What's the explanation for why the same colors, whether it's in a situation of color contrast or color constancy, why this is happening? Well, one way in which people have tried to explain this is to sort of pin the logical burden for why this is happening on color constancy. Let's go back and look at color contrast and constancy. Contrast is making the same spectral patches look look different, and constancy is making different spectral patches look more similar. And people who've sought to explain this really want to put the burden on constancy in a kind of a common sense way, by saying well, look, you need to see a yellow object, let's take Banana. You need to see that banana as yellow because you want to identify it as a banana whether the illuminative of that banana is blue or yellow or anything else. So you want to have something in your perceptual apparatus that allows you to see the same things as being yellow, or at least yellowish, when they are illuminated in different kinds of lights as you might find in early morning, middle of the day, evening, when the light is either bluish in the morning, yellowish in the middle of the day, or reddish at sunset. You want to see the color as constant. But I don't think that really, [SOUND] it's a sensible enough, but I don't think that you can pin the burden on constancy, because contrast and constancy as you've seen in these demonstrations, comes from exactly the same qualities of the scene. So we haven't changed anything here to generate contrast in the one hand and constancy in the other. These, in some sense are different sides of the same coin and you can't just say, well, it's all about constancy. Contrast and constancy are really the same thing. It's just whether you put different colors in the scene, or the same colors in the scene which effect you get. So I think it's much deeper than just these phenomenal, this phenomenology of contrast and constancy. It's much deeper than just saying, well, we need to see a banana in different lights as yellow. That won't suffice to explain the depth and the puzzle that underlies the fact that contrast and constancy come from the same scenes, depending on what the colors are that are placed in these positions in the scene, or any other scene that you might want to talk about in these terms.
