Abstract
Several past studies have demonstrated that internal motion signals that do not reflect in subjective perception do affect other aspects such as motion aftereffect. In this study, we examined whether such hidden motion signals affect vection by using pedestal and a type of missing fundamental stimuli (2f+3f motion). The pedestal stimulus was generated by adding static gratings to drifting gratings of the same spatial frequency. These stimuli contained motion energy to one direction but no directionality for feature-based system, and motion perception actually became ambiguous when the contrast of static gratings was high. In these occasions, vection should occur according to the first-order direction, if vection is mediated solely by low-level motions. In the experiment we asked subjects to judge the direction of vection and perceived motion in separate sessions. The pedestal stimuli and regular drifting gratings were presented for 30 sec each in a randomized order. Whereas as normal vection was perceived with regular gratings, no systematic vection was found for pedestal stimuli when perceived motion was ambiguous. A similar experiment was conducted with 2f+3f stimuli. These stimuli had 2nd and 3rd harmonics of a fundamental frequency, and the first- and second-order components moved in opposite directions when motion was created by shifting the pattern a quarter wave-length of the fundamental at a time. The dominance of the two types of motion, i.e. the direction of perceived motion depended on SOA. The direction of vection should follow the first-order direction regardless of the perceived motion, if vection is controlled by low-level motion. As a result, we found a high correlation between the directions of vection and perceived motion. These results together indicate relatively weak contribution of low-level motion to vection. Rather, they demonstrate a strong influence of higher-order motion systems.