Purchase this article with an account.
Bong-sun Lee, Zygmunt Pizlo, Jan P. Allebach; Characterization of red-green and yellow-blue opponent channels. Journal of Vision 2004;4(8):314. doi: 10.1167/4.8.314.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: Responses of opponent-channels have been modeled in the past as a linear transformation of cone absorption values L,M,S (and of tristimulus values CIEXYZ). Different authors proposed different linear transformations. We asked two related questions: (i) which of these transformations is psychologically most plausible; (ii) is a linear transformation the right model, in the first place. Linear transformations have commonly been considered acceptable in view of substantial individual differences with respect to the position of pure hues. To shed more light on the two questions, we tested positions of pure colors for individual subjects. Methods: The subject viewed a patch in the center of a calibrated CRT monitor with neutral gray background. The subject's task was to adjust the intensity of one gun so that the patch looked as having pure color (red, green, blue or yellow). The starting hue of the patch was produced by a weighted combination of the other two guns. By changing the weights of the two guns, the experimenter manipulated the saturation of the final adjustment. Each of the five subjects ran four sessions (one session per hue), 10 trials per session. Results: Neither of the two opponent channels can be adequately approximated by a single straight line. The red-green channel can be approximated by two straight lines, forming an angle that did not vary much across subjects. The yellow-blue channel can be approximated by a quadratic function, whose middle section coincides closely with the daylight locus. Conclusions: Linear transformation of cone responses (or of tristimulus values) does not seem to be an adequate model for describing the characteristics of opponent channels. Furthermore, it does not seem that any simple non-linear transformations could be an adequate model, either.
This PDF is available to Subscribers Only