Abstract
The stream bounce effect (Sekuler, Sekuler & Lau, 1997) typically depicts two identical targets that move toward one another from opposite sides of a 2-D display, coincide, and continue past one another along linear trajectories. The targets can be perceived to either stream past or bounce off of one another with equal likelihood. However, streaming is the dominant perception in motion displays free of additional transients while a transient (e.g. auditory or visual) presented at the point of coincidence shifts this bias towards bouncing. From a psychophysicist's point of view, this transient induced shift in perceptual bias could be caused by at least three factors: 1) integration at the sensory level where the transient signal and the visual motion sequence are integrated at an early stage of processing and that combination becomes available to conscious awareness determining the response; 2) a decisional level of processing where the two inputs are available independently and the observer's response is based upon decisional biases or strategies; 3) a combination of 1 and 2. We quantified the relative contributions of sensory and decisional factors in resolving stream/bounce displays using novel motion sequences in which the motion targets were visually distinguishable. We measured sensitivity (d') and criterion (c) using a signal detection paradigm in which targets objectively streamed on 50% of the trials and bounced in the remaining trials. We presented an auditory click at coincidence on 50% of the streaming and 50% of the bouncing sequences. Observers were required to indicate whether the targets streamed or bounced. Mean d' values for detecting bounces were little different between the sound and no sound conditions while c was significantly more liberal in reporting bounces in the sound condition compared to the no sound condition. These results suggest decisional factors dominate the resolution of stream bounce displays.
Grant-in-Aid of MEXT for Specially Promoted Research (no. 19001004).