Three-systems theory of human visual motion perception: review and update: comment.

This comment addresses two issues raised by Lu and Sperling [J. Opt. Soc. Am. A 18, 2331 (2001)]. These authors stated that stereomotion (movement of binocular disparity) is processed exclusively by a third-order motion system that involves feature tracking. This comment discusses evidence that clearly shows that stereomotion is processed by a low-level mechanism that does not track features. Lu and Sperling also claimed that motion signals from binocular rivalry have confounded many stereomotion experiments in the past. This comment discusses how stimuli employed in most previous studies of stereomotion processing would not produce rivalry.