Binaural fusion, apparent motion, and the precedence effect
David R. Perrott, Thomas Z. Strybel · The Journal of the Acoustical Society of America · 1986
Wideband noise pulses, 50 ms in duration, were presented from pairs of speakers. Across sessions, the angular separation of the sources was varied between 6 and 160°. Within sessions, interstimulus onset intervals (ISOI) varied between 0.1 to 500 ms. Eight listeners were required to perform two concurrent discrimination tasks. The first was an objective procedure in which they were required to report the temporal order in which the sources were activated. The second task required them to assign the event to one of five descriptive categories: single stationary image; multiple stationary images; single moving image; an interrupted but moving image; or two successive images. Discrimination of the temporal order of speaker activation was only moderately affected by the angular separation of the sources (thresholds ranged between 12–20 ms). These results are well in line with other experiments in which temporal order judgments are required. However, when subjects report “apparent motion,” a relatively rare event with small ISOI's, judgments of temporal order are remarkably accurate (exceeding 80% correct) with all but the shortest ISOI's employed. The implications of these results will be discussed. [Work supported by NSF.]