Binocular summation of suprathreshold color

Two eyes are better than one for detecting both achromatic and chromatic contrast, an advantage termed binocular summation. It is widely believed that binocular summation is mediated by neurons that sum the signals from the two eyes. While binocular summation for luminance-defined stimuli has been measured for both threshold and increment-threshold contrasts, those for chromatic stimuli have only been measured at threshold contrasts. Here we measured binocular summation for both threshold and increment-threshold contrasts defined along the black-white, red-cyan, and violet-lime axes of cone-contrast color space. Stimuli were 1.0 cycles per degree Gabor patches at a range of pedestal contrasts, each presented to both eyes, added to which was an increment to the left, right or both eyes. Left eye and right eye data were combined, and binocular summation was calculated as the ratio of monocular to binocular increment thresholds. For all three color directions, binocular summation was observed throughout the range of pedestal contrasts, with greater summation for the red-cyan and violet-lime compared with the black-white color directions. Four models of binocular summation were tested: a probability and an additive summation model each with two free parameters, a three-free-parameter model of additive summation, and a four-free-parameter model based on the modelled responses of binocular adding and binocular differencing channels, termed Max(S + S -). The three-free-parameter additive and Max(S+S-) models fared best and equally well.