Motion-transparent inducers have different effects on induced motion and motion capture

To assess the relationship among the underlying mechanisms of induced motio n, motion capture, and motion transparency, directions of the former two il lusions in the presence of motion-transparent inducers were examined. Two r andom-dot patterns (inducers) were superimposed upon a stationary disk (tar get), and moved in orthogonal directions. Either a high-contrast target (fo r induced motion) or a low-contrast target (for motion capture) was used. T he task was to report the perceived direction of the target. The depth orde r of inducers was controlled either by adding binocular disparity or by ask ing the subject to report subjective depth order. For induced motion, the t arget appeared to move in the direction opposite to the inducer that had a disparity closer to the target; when there was no difference in disparity, induced motion occurred oppositely to the 'vector sum' of the inducers' dir ections. For motion capture, the target was captured by the inducer that su bjectively appeared behind. These results suggest that the underlying mecha nism of motion capture utilizes the output from the process for motion tran sparency, whereas induced motion has no clear relationship to the output of the process for motion transparency. (C) 1999 Elsevier Science Ltd. All ri ghts reserved.