Abstract
When the internal texture of a Gabor patch drifts in the orthogonal direction of the patch itself, the perceived path can deviate from its physical path by 45° or more (Tse & Hsieh, 2006; Lisi & Cavanagh, 2015). Recently, Adamian and Cavanagh (ECVP 2015) reported that the size of this double-drift illusion, when tested with an elliptical trajectory, is reduced when presented at either the vertical or horizontal meridian. In the present study, we used a Gabor patch that oscillated back and forth on a linear path, reversing its internal motion at the same time as the external motion reversed. We measured the strength of the illusion by asking subjects to add a physical tilt to the perceived path angle until the trajectory appeared to be aligned with two reference dots that were set at 45° from the initial motion axis. For a strong illusion, the path would already appear aligned with the dots and no additional physical tilt would have to be added. The weaker the illusion, the more physical tilt would have to be added to reach alignment. The physical path of the stimulus was centered at one of the eight isoeccentric peripheral locations (8° from fixation) with one of four external motion orientations (vertical, horizontal, right diagonal and left diagonal) and one of two adjusting directions (45° clockwise or counterclockwise from the initial physical path). In contrast to the previous finding, our results show that the magnitude of the motion-induced position shift of the double-drift stimulus was similar across all eight locations in the visual field for all the physical or the adjusted orientations of the motion path. Measured with a linear rather than curved path, the illusion appears to be based on its physical configuration and independent of retinal location or orientation at a given eccentricity.
Meeting abstract presented at VSS 2016