Abstract
There is no generally agreed upon formula for expressing saturation and chromatic contrast. Here we investigate saturation coding in early visual cortex by way of the McCollough orientation-contingent color-aftereffect. We have previously shown the McCollough aftereffect strength is linear with induction time, suggesting that the effect is proportional to stimulus effectiveness (Vul, Krizay, & MacLeod, 2008). Is the same true for the saturation of the inducer? Subjects adapted to a fully saturated red and green pair of equal-luminance inducing gratings, produced by a wide gamut LCD. Aftereffect strength was assessed using a nulling technique, and comparing pre- and post-adapt nulls. Halving the inducing saturation of the gratings (by halving the difference in cone excitations) reduced the aftereffect strength only slightly. This suggests a compressive coding of saturation, where equal successive increments in cone contrast produce diminishing returns in the cortical code for saturation. Our result predicts that anomalous trichromats should see a larger McCollough effect. In anomalous trichromats, the apparent difference between red and green is greatly reduced owing to the proximity in peak spectral sensitivity (λmax) between the long-wave and middle-wave cones. During adaptation this only slightly reduces the induced aftereffect, because of the saturation non-linearity we found. In order to null the resulting aftereffect, however, the color-deficiency of the retina must be overcome by adding significantly more chromaticity to the null than a normal viewer would require. Initial evidence supports this prediction; the one anomalous subject tested so far had a McCollough effect several times larger than normal.
Supported by NIH Grant EY01711. Alan Robinson also was supported by the Temporal Dynamics of Learning Center at UCSD (NSF Science of Learning Center SBE-0542013).