Salience of spatiochromatic patterns.

Abstract

We investigated the dependence of perceived contrast on cone-opponent stimulus content and its spatial distribution. Participants matched a comparison patch to a light gray standard of fixed contrast. The first experiment determined the point of iso-salience for gratings, Gabors and Gaussians along cardinal directions in cone-opponent color space for two-alternative forced choice (2AFC) and adjustment tasks. No difference was found between adjustment and 2AFC tasks, meaning that adjustment tasks provide a quick and robust way to measure perceived contrast, at least for relatively large suprathreshold stimuli. In line with the differences in contrast energy between Gaussians, Gabors, and gratings, Gaussians required less contrast to achieve equal perceived salience with a standard irrespective of color. More surprisingly, bluish Gaussians were found to have higher salience than yellowish Gaussians at equal levels of contrast. Although perceived contrast of grating and Gabor patterns likely depends on spatial frequency channels that at 1 cycle-per-degree are not too dissimilarly tuned for color and luminance, for Gaussians the contribution of single-opponent neurons would be greater for color than for luminance. In a follow-up experiment, we found that the bluish/yellowish asymmetry decreased as we reduced the proportion of the lowpass non-flat contrast distribution in the stimulus, with minimal asymmetry for the stimulus with a flat contrast distribution (i.e., uniform patch). Combined, this means that differential engagement of spatial frequency channels, single-opponent and double-opponent neurons impacts on perceived contrast of chromatic suprathreshold stimuli. Perceived contrast thus provides a window into neural computations enacted by low-level cone-opponent mechanisms.