This is a satisfactory document on color vision that is strong in considering the psychological aspects of color but weaker on the physiological ones. I have several suggestions to make to the authors.
1. Indeed color vision depends on wavelength contrasts but color vision combines both wavelength (chromatic) and energy (achromatic) contrasts to create “color”. I often cite the statement of Maxwell that “all vision is color vision”.
2. The weakest aspect of the paper is its consideration of the physiology. This comes up in Figure 3, for example. The S cone + is combined with an L and M - signal. But in reality there also seems to be S cone – combined with and L and M + signal in this opponent channel. Again they combine a L+ against a M- channel in the next example of color channels in Figure 3. But there is also a L- combined with a M+ signal. There is a purpose to these additional channels.
3. I laud them for accuracy in citing Gunnar Svaetichin for initially discovering opponency between cone mechanisms in fish retina and indeed Russ DeValois deserves credit for detecting the first cone opponent cells in the primate lateral geniculate nucleus just a bit later. I think Russ had not seen the Svaetichin results until his 60 paper, which is fascinating only for the historians and not needed to be mentioned in this paper.
4. Again there is a deficiency in the description of how most color vision scientists consider the coding of color to depend on the midget system playing a dual role in transmitting signals for both achromatic and chromatic contrast. This point is very important and delves into the question of whether the Hering color opponent channels are formed in the retina or in the cerebral cortex as most of us now believe.
5. The very interesting discovery of “double opponency” is not mentioned which is a weakness because this neural operation begins to reveal how chromatic and achromatic contrast is untangled and how color constancy could be established.
6. The question of color constancy fails to cite the very interesting work of Edwin Land and his Retinex Theory of color vision which remains one of the best predictors of color.
7. Indeed they do bring up the point that the circuitry of color vision is still mysterious but it would be nice to try to describe why and where it is so mysterious.
8. I would have liked them to have cited my Chapter on Color Vision in the website, Webvision, which does deal more with the physiology and circuitry of color vision than they do. Peter Gouras