Abstract
The static envelope of a Gabor patch with a moving carrier appears shifted in the direction of the carrier motion (motion-induced position shift; MIPS). Such a conventional configuration contains at least three co-varying factors, i.e., the retinal carrier velocity, the environmental carrier velocity, and the carrier-envelope velocity difference. To elucidate which factor is critical, we independently manipulated them and measured the perceived position of the moving Gabor patch. We presented two vertically aligned Gabor patches, a reference patch and a test patch. The reference contained a horizontally oriented static carrier, whereas the test contained a vertical carrier drifting in various velocities. The envelopes of the reference and test patches moved coherently to the left or right at 2.5 deg/s. By asking each subject to judge the relative horizontal position between the reference and test, the position of subjective alignment was established as the index of illusion strength. In the first experiment, the MIPS of the moving envelope was observed during fixation; the perceived position of the moving envelope shifted in the direction of the carrier. Furthermore, the MIPS was greater when the carrier moved oppositely to the envelope motion. In the second experiment, we measured the MIPS during smooth pursuit eye movements to the left or right at 2.5 deg/s, with the envelopes of the patches being either static or moving in the pursuit velocity, thereby dissociating retinal and environmental velocities. Under all conditions the MIPS was induced in the retinal direction of the carrier. Also, the MIPS was greater when the movements of the carrier and envelope were opposite to each other in retina-centered coordinates. We conclude that the retinal velocity of the carrier is the primary determinant for the MIPS, and we will discuss possible additional contribution of mechanisms detecting the motion contrast between the carrier and envelope.