Abstract
V1 encodes both the sum of, and the difference between, the visual
inputs to the two eyes, (Li and Atick, 1994, Network,5(2),157-174);
however, perception favours the sum. Consider a dichoptic stimulus
for which the binocular sum and difference signals are both sinu-
soidal gratings with the same amplitude, but which drift in opposite
directions. This arises from a flashing, but not drifting, grating to
each eye, with the spatial and temporal phases of the left eye grating
different from those of the right eye grating by 90 degrees. In
central vision, subjects more likely see the drift direction associ-
ated with the sum rather than the difference (Shadlen and Carney,
1986, Science,232(4746),95-97). However, I recently found that
this bias for binocular summation is absent when the gratings
were viewed peripherally (about 10 degree eccentricity; stimulus
enlarged to counter acuity change), and that reducing the speed of
the drifting increases this bias in central, but not peripheral,
vision (Zhaoping, submission to ECVP 2013). Here, we extend
this study to the case that the binocular sum and difference
signals are static gratings, but tilted in opposite directions from
horizontal. When the two gratings have equal amplitudes and are
viewed centrally and briefly, subjects more likely report the ori-
entation of the sum rather than the difference grating; however,
when viewed peripherally, the two orientations are roughly
equally likely to be perceived. Furthermore, reducing the duration
of stimulus presentation, e.g., from 100 to 50 milliseconds,
reduced the bias for binocular summation in central viewing only.
Just as for drifting gratings, the probability of perceiving the ori-
entation associated with the sum increases with the amplitude of
the sum relative to the difference grating, but for any amplitude,
this probability tends to be higher for central than peripheral vision, unless it is too close to zero or one.
