Associative responses to colour and pattern in the human visual system

Abstract

Psychophysical methods have been applied to the following four aspects of the McCollough effect with a view to the elucidation of its mechanism and its locus:
(1) time course of build-up and decay
(2) association of colour with scale of pattern
(3) interocular relations
(4) angular distribution function.
1. At normal light levels decay of the McCollough effect typically follows a power law with exponent = 1/3. Its growth follows a similar power law with exponent + 2/3. In darkness decay is arrested. After a period in the dark the course of decay is similar to that Immediately following exposure to the inducing stimuli. Unpatterned light is as effective as randomly patterned light in producing decay, but decay proceeds particularly rapidly during exposure to achromatic gratings at the same orientations as the inducing gratings,
2. A large pattern-contingent chromatic aftereffect can be induced and detected using stimuli which contain no straight oriented edges but which differ only in magnification or texture.
3. Although the McCollough effect transfers to a barely detectable degree to a covered or steadily Illuminated eye, pattern-contingent chromatic aftereffects can be induced dichoptically, and can also have a specifically binocular component. The former are visible when each eye is tested separately, and are opposite in the two eyes, ("normal" in the eye exposed to coloured unpatte>-ned fields, and "anomalous" in the eye exposed to achromatic patterns). The latter shows up as a difference between the strengths of effects seen monocularly and blnocularly. This difference Is positive or negative according as the McCollough stimuli concurred or conflicted in the two eyes during induction of the effects. k. The orientation contingent aftereffects induced by a single coloured grating have a Gaussian angular distribution function with points of inflection 42° apart and are of the complementary hue only. No signs of "opponent processing" of orthogonal orientations have been found. The nett aftereffect induced by successive exposures to two or more coloured gratings Is the linear sum of the effects of the Individual grattngs.
It is suggested that these effects are best explained in terms of synaptic modification rather than fatigue of "edge-detectors".