Motion correspondence determined by structure

I. Mukai, M. A. Moreno, T. Watanabe

Research output: Contribution to journalArticlepeer-review


Purposes: Mukai & Watanabe (1994, ARVO) created a stimulus array that consisted of small moving rectangles. The surface of the array appeared to be either cylindrical or flat and the small rectangles appeared to move either only in one direction (one-way motion), or to and fro (oscillation), depending on frame duration. They found that one-way motion (vs. oscillation), and the formation of a cylindrical structure (vs. a flat surface), co-occurred. Using the same and similar stimuli, the present study seeks to explore how the mechanism for the one-way motion and that for the cylindrical structure interact. Methods: An array of small rectangles was sequentially presented in three horizontally different loci at different times (t0, t1 and t2). The whole array jumped rightward twice at t1 and t2 and there jumped leftward back to the original position at to (t3) repeatedly. The width of the rectangles at the left edge at t0 and at the right edge at t2 was half that of the other rectangles of the array. This change in width was intended to be a local depth cue. Motion correspondence for the rectangles was set to be ambiguous so that either one-way motion or oscillation was perceived depending on the frame duration. Experiment 1 examined whether there was any effect of motion correspondence on the depth structure of the surface. Rows of rectangles in the stimulus were painted in alternating colors so that one-way motion could be perceived more frequently with the two-color array than with the array of black rectangles. It the occurrence of one-way motion facilitates the perception of the cylindrical depth structure, this structure should be perceived more frequently with the two-color stimulus. Experiment 2, in turn, examined whether there was any effect of depth structure on motion correspondence. We compared the frequency of the occurrence of one-way motion in the stimulus with the local depth cue to that in the stimulus with no depth cue. In both experiments, the duration of frame presentation was varied. In one block of each experiment, the subjects were instructed to judge whether they perceived a cylindrical structure or not. In the other block, they were instructed to judge whether they perceived one-way motion or oscillation. Results. The results of Experiment 1 show that one-way motion was perceived more often with the array of two-colors than with the black array, whereas there was no significant difference in frequency of the perception of the cylindrical depth structure between the two color and the black stimuli. The results of Experiment 2 show that one-way motion was more frequently perceived when the cylindrical depth structure was perceived than when it was not. Conclusions. These results challenges the view that structure is generated from motion only through feedforward processing, that is, that depth structure a formed subsequently after motion correspondence is determined at a lower stage.

Original languageEnglish
Pages (from-to)S745
JournalInvestigative Ophthalmology and Visual Science
Issue number3
StatePublished - 15 Feb 1996
Externally publishedYes


Dive into the research topics of 'Motion correspondence determined by structure'. Together they form a unique fingerprint.

Cite this