Abstract
Many direction-selective neurons have a receptive field structure that promotes suppressive interactions between center and surround regions. These interactions sculpt the overall pattern of activity among those neurons and, therefore, presumably impact perceived direction of motion. To test this conjecture, we have assessed the effect of motion signals produced by a moving stimulus on perceived motion within a neighboring region. On each trial a vertical bar (inducer) appeared at 8 eccentricity in the upper visual field, moving either leftward or rightward, and a circular shaped random dot kinematogram (test) appeared at 4 eccentricity. The test dots moved randomly except when the inducer passed nearby the test, at which time a pulse of coherent motion occurred in one of the two directions within the test. Coherence strength was adjusted by QUEST to maintain equal likelihood (point of subjective equality: PSE) of leftward and rightward reports of perceived direction during this motion pulse. The inducer caused a substantial shift in PSE: it was necessary for the test to contain 50% coherent motion in the same direction as that of the inducer to nullify the illusory motion within the test caused by the inducer. The effect of the inducer could also be offset by simultaneously presenting a second inducer moving in the opposite direction. This pattern of results implies substantial suppressive interactions between neighboring moving stimuli, interactions whose strength and direction can be assessed psychophysically using nulling procedures.