Abstract
The ‘colour-shading effect’ (Nat. Neuro., 2003, 6, 641) describes how perceived 3D-shape-from-shading in flat luminance patterns is enhanced by non-aligned, and suppressed by aligned chromatic patterns. We investigated the colour specificity of the effect using mixed colour-and-luminance-defined sine-wave plaids, in which the chromatic contrast components were defined along the cardinal axes of colour space: L-M (red to bluish-green) and S-(L+M) (violet to yellowish-green). L-M and S-(L+M) contrast was found to be equally effective at triggering apparent depth in the superimposed, orthogonal-in-orientation (i.e. non-aligned) luminance contrast component, providing the L-M and S-(LUM) contrasts were equated for visibility. As with previous findings, adding colour contrast in alignment, and in phase, to the luminance component suppressed apparent depth. We found that the amount of suppression was the same irrespective of whether the enhancing and suppressing colour contrasts were defined along the same, or along different cardinal axes. These results suggest that the colour-shading effect is agnostic with regard to the particular type and combination of enhancing and suppressing colour contrasts. In turn this implies that the assumptions made by the visual system that arguably underlie the colour-shading effect, namely that chromatic variations arise primarily from variations in spectral reflectance, whereas near-pure luminance variations primarily arise from inhomogenous illumination, are of a general nature and are not limited to certain colour contrasts, or colour-contrast combinations.
Supported by Canadian Institute of Health Research grant #MOP-11554 awarded to FK.