Citation

Kline K, Holcombe AO, Eagleman DM. Vision Res. 2004; 44(23): 2653-2658.

Affiliation

Department of Neurobiology and Anatomy, Houston Medical School, University of Texas, 6431 Fannin St., Suite 7046, Houston, TX 77030, USA.

Copyright

(Copyright © 2004, Elsevier Publishing)

DOI

10.1016/j.visres.2004.05.030

PMID

15358060

Abstract

In stroboscopic conditions--such as motion pictures--rotating objects may appear to rotate in the reverse direction due to under-sampling (aliasing). A seemingly similar phenomenon occurs in constant sunlight, which has been taken as evidence that the visual system processes discrete "snapshots" of the outside world. But if snapshots are indeed taken of the visual field, then when a rotating drum appears to transiently reverse direction, its mirror image should always appeared to reverse direction simultaneously. Contrary to this hypothesis, we found that when observers watched a rotating drum and its mirror image, almost all illusory motion reversals occurred for only one image at a time. This result indicates that the motion reversal illusion cannot be explained by snapshots of the visual field. The same result is found when the two images are presented within one visual hemifield, further ruling out the possibility that discrete sampling of the visual field occurs separately in each hemisphere. The frequency distribution of illusory reversal durations approximates a gamma distribution, suggesting perceptual rivalry as a better explanation for illusory motion reversal. After adaptation of motion detectors coding for the correct direction, the activity of motion-sensitive neurons coding for motion in the reverse direction may intermittently become dominant and drive the perception of motion.