Abstract
Here, we report a new cross-modal effect. When observers had to time the occurrence of a brief sound (S) relative to a moving visual stimulus, the S appeared to be dragged later and later in time over successive motion sweeps, despite the fact that its timing relative to the motion was the same on each sweep. Over the course of the trial, the perceived drag would abruptly and randomly reset to a value close to the initial one, and then start all over again on the succeeding sweeps. Method: A bright circular disk (100.4 cd/m2, 1 deg. dia., 10 deg. eccentricity below fixation) drifted 21 deg. from left to right (or right to left, depending on the trial) over 465 ms on a dark background. This motion repeated 15 times (sweeps) on each trial. A 2.2 ms long S (65 dBa SPL) occurred at the same instant relative to the visual motion on all sweeps. After all sweeps were concluded on the trial, observers (n=5) had to indicate where the disk was in space when the Ss occurred. Os made two reports with a mouse: the locations of the disks coincident with the first and the last S. Results: Congruent with informal observations, the perceived disk location coincident with the final S was further along in the motion than that of the first S. (difference = 3.84 deg. +/− 1.56 deg., p<.001 for n=3, p<.05 for n=1). We asked whether multiple Ss or multiple motion sweeps contributed to the sound drag. In one experiment (2 sounds), we played the S only in the second and last motion sweeps. In another (2 sweeps), we showed the moving stimulus only for the second and last Ss. The sound drag in the motion direction was not as large in either experiment compared to the original (2 sounds: n = 3, difference = −2.27 deg.; 2 sweeps: n = 2, difference = 1.81 deg.). Thus, both the sound and the motion contributed to the sound drag. Other factors and the relationship of the effect with the flash-lag effect (Nijhawan, 1994) and the flash-drag effect (Whitney & Cavanagh, 2000) will be discussed.