Abstract
The luminance-defined (LD) motion influences the perceived position of a moving object itself (Kinetic edge effect, Ramachandran & Anstis, 1990). However, little is known about the perceived position of more complex motion. In this study, we investigated the perceived position of motion-defined (MD) motion stimuli. In the experiment, observers viewed LD or MD motion displayed in a rectangular patch at the center. Both stimuli consisted of random dots and were surrounded by a dynamic random-noise field. The dot luminance in the LD stimuli was modulated to generate a square-wave pattern. The dots in the MD stimuli moved either upward or downward to generate a motion-defined square wavepattern. The modulation wave was shifted leftward or rightward to generate horizontal LD/MD motion. Two white wire-frame rectangles were displayed above and below the motion patch, and served as references. The horizontal offsets between the motion patch and the references were varied randomly across trials, and observers were asked to judge the direction of the offset. We pooled data from each subject and estimated the point for each motion direction condition where the perceived position of the central patch is aligned with the references. The perceived offset was estimated as a half of the distance between the estimated points in two motion direction conditions. Consistent with previous findings, we obtained a significant kinetic edge displacement for the LD stimuli with the effect size being several arc minutes. However, no significant displacement was found for the MD stimuli. The same pattern of results was found even when the observers accurately (around 95%) judged the direction of MD motion concurrently with the main position judgment task. Our results show that the activation of higher-motion system, at least in this particular case of MD motion, does not influence the position coding of a visual object.