Abstract
When added, two identical sinusoidal gratings moving in opposite directions appear as a single grating that is flickering in place with no net motion. This is taken as evidence for the existence of motion opponent mechanisms. Single neurons in MT are often inhibited by motion in the anti-preferred direction, indicative of motion opponency.
Most neurons in MT can be driven strongly through either eye. We wondered whether the motion opponency expressed so strongly under monocular conditions still occurs under dichoptic conditions. We recorded from MT neurons in anesthetized, paralyzed macaques. For each neuron, we measured the contrast response function to a preferred grating presented alone, and in the presence of a high contrast grating moving in the opposite direction. The two gratings were either presented in the same eye (monocular) or one to each eye (dichoptic). The anti-preferred stimulus caused the contrast response function to shift rightward on a log-contrast axis, with the effect being largest in the monocular condition. To characterize this effect we fit each contrast response function with a Naka-Rushton equation in order to identify the contrast at which the response of the neuron was half-maximum (c50). In the monocular condition, the anti-preferred grating increased the c50 by a factor of 4, in the dichoptic condition it increased the c50 by a factor of 1.5. Control experiments confirmed that this dichoptic effect represents genuine opponency and not just a dichoptically exerted contrast gain change.
This distinction between dichoptic and monocular motion opponency mirrors recent results obtained psychophysically (Gorea et al. 2001, Vision Res.). Taken together, the psychophysical and physiological results suggest that motion opponency is for the most part generated prior to binocular combination.