Abstract
I will review some of the presumed successes, and some acknowledged and unacknowledged obscurities, in our physiological understanding of the dimensionality of color space. As we trace the flow of information from object to image and thence through the retina and visual pathway, it becomes harder, not easier, to discern an isomorphism between color as we perceive it and the neural representation of color in the visual system. Even the physiological explanation of trichromacy presents as yet unanswered challenges in the common situation where more than 3 visual pigments are involved.
Color charts commonly adopt a coordinate system with two bipolar axes, one for redness (positive) vs. greenness (negative) and the other for yellowness vs blueness. Perceptually unitary (“unique”) red and green, and blue and yellow, are found in opposite directions from white at the origin. But the pervasive rectifying nonlinearity of neural responses, and the marked asymmetry between excitatory and inhibitor response dictated by the relatively low spontaneous firing rate, are equally supportive of an idealization with multiple monopolar signals for redness, greenness, yellowness and blueness. With mutually orthogonal monopolar coordinates for the four primary signals, the isoluminant colors lie on the surface of a hypercube in the 4D space, with the achromatic point at one corner. To test this scheme, MacLeod, Pallett and Krizay asked whether red, green, yellow, and blue colors equidistant from white are perceptually equidistant from each other, as the hypercube model predicts. Results are close to the predictions of the hypercube surface model.
Supported by Grant No. EY01711.