Abstract
Our performance of distinguishing between leftward and rightward motions is limited by internal noise. This study aimed at clarifying what is the optimal stimulus to counteract this noise and to yield the best performance. First, we found that slightly tilted (e.g., by 22 deg) directions deviating from the horizontal were better discernable than purely leftward and rightward directions. Within a blurred window, a random-dot pattern moved in one of two possible directions that were mirror-symmetric about the vertical, and the correct response rate was measured in the two-alternative forced response paradigm. Compared with purely horizontal directions (0 deg versus 180 deg), the judgment for two oblique directions (e.g. 22 deg versus 158 deg) was slightly better, even when the physical velocity was identical. This advantage became weak as the angular deviation from the horizontal increased, and the performance became worse when the task was to distinguish between 67 and 113 deg. Second, we found a similar advantage of slightly oblique directions when they were not displayed physically but produced internally in the brain as a result of motion integration between horizontal physical motion and vertical illusory motion. A central Gabor patch with a horizontally drifting carrier was surrounded by another grating drifting vertically. Subjects perceived the Gabor patch as moving in an oblique direction due to integration between horizontal carrier motion and vertical induced motion. The detection performance as determined by the correct response rate of leftward/rightward judgment was enhanced at a relatively slow surrounding speed, when the Gabor was perceived as moving obliquely. A similar modulation of detection performance was also observed when we replaced induced motion with motion aftereffect. These results suggest that interactions with the critical internal noise for the present tasks reside in a higher-order motion processing stage, where vectorial integrations and center-surround interactions take place.
Nissan Science Foundation.