So to give you a general idea of how empirical explanation can be a work-around for the inability of our color perceptual system or any other aspect of our visual system to get at reality. Let's consider these two scenes here, which you'll recognize as being very similar to the scenes that we used in luminance. So remember, in talking about luminance and the empirical resolution of the inverse problem in that context. What we did was to put a gray patch in the center of the same grey patch in the center of two different surrounds, a dark surround and a light surround. And to show that they look different. The one in the dark surround looked lighter than the one in the light surround. And now we're doing exactly the same thing. In the case of color, and we're also getting the same lightness contrast effect that we got before. So, let's consider what's happening here. So these are two spectrophotometrically identical patches and when you look at them in two similar surrounds, in two neutral surrounds, they look the same. When you put these same patches in a different color surround, so this being a reddish surround and this being a yellow surround, they now look different. And they look different in all the ways that we talked about in considering the diagram and the descriptors of the way we think about and talk about color perception. That is, this one is lighter than this one. The saturation between these two is now different and the hue is now different. Generated by the surround that's now not just a surround that's differently illuminate, but differently colored, different in hue. So again because of that, because of the context in the same simple color contrast example you're seeing all the perceptual qualities of color differently. Just by virtue of the surround, in these two different scenes, that are changing the color appearance of identical patches. So why is that happening? Well it's this, the answer, generally speaking, is the same as the argument that we went through, for illuminace. These two identical central targets look different because in this case their perceived color values are tracking empirical experience. Our experience is generating the colors that we see by trial and error over the course of evolutionary time not just by reading out the spectral qualities of those colors which the retina really doesn't have access to. And when we talk about evolution we of course are talking ultimately about reproductive success. So if you ask well how does seeing the luminance example or these simple examples lead to reproductive success? Well, I mean, it's a long way from here to there. But seeing colors in this way, just as seeing lightness and darkness in this way, is a way of empirically figuring out in our perceptual, visual apparatus. A way to work around the fact that we don't have the physical information to know what's out there in the world, but we do have the information gathered over millions of years and our lifetimes as well of experience with different spectral values and different spectral relationships just as we have with different luminance values. And different numerous relationships and by ranking, by laying out the perception of light and dark in grey scale or the perception of color in this case. According to that empirical experience rather than according to information that we just don't have, I mean there's really no alternative. We can succeed in the world [COUGH] and again I remind you that this information is instantiated over time in a visual apparatus by evolution, changed also in our life times but that's not heritable. And so the major work of making all this happen has to be done by evolution because what we experience in our lifetimes is valuable to us obviously but not heritable.
