Abstract
To measure each eye's individual contribution to a binocularly perceived (cyclopean) image, we present a horizontal sinewave grating in a different phase to each eye. The perceived phase of the cyclopean grating defines the relative contributions. These perceived phase data are encapsulated by a gain-contol model in which each eye exerts gain control on the other eye, and additionally gain control on the other eye's gain control (Ding and Sperling, Proc. Natl. Acad. Sciences, in press). Here we define the gain control by masking experiments. Temporal masking. Adding a moving sinewave to one eye's image, increased the masked eye's contribution to the cyclopean image equally for frequencies less than about 15 Hz, then less with increasing frequency. When bandpass-filtered noise was added to the one eye's grating, the more noise, the greater the relative contribution of the noisy eye. The most efficient masking noise was 4x higher in spatial frequency than the sinewave being judged. Similar results were obtained with perpendicular masking gratings as a function of spatial frequency. Barely visible masked gratings completely dominate clearly visible gratings of equal contrast. The masking stimulus perturbs the masked eye's contents but exerts gain control on the other eye. Masking as a function of the orientation angle of masking gratings was tested with 8x higher-spatial-frequency gratings. Effectiveness was somewhat greater for both vertical and horizontal gratings than for diagonal orientations. This orientation selectivity of masking indicates that at least some of the interocular gain control is of cortical (versus LGN) origin.
Air Force Office of Scientific Research, Life Sciences, Grant FA9550-04-1-0225