Tye and Byrne and Hilbert have proposed a solution to this last problem, one that depends on exploiting the opponent-processing model of color vision (see Supplement Color Science: Some Complexities.). It is to specify the relevant groups of spectral reflectances, associated with each color, in terms of their capacity to produce suitable responses of the visual system. The argument, by Byrne and Hilbert, proceeds in two steps: (1) it is argued that color experience are characterized as having a