Purchase this article with an account.
David C. Burr, John Ross; Direct evidence that ‘speedlines’ aid perception of motion direction. Journal of Vision 2002;2(7):259. doi: 10.1167/2.7.259.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Geisler (Nature, 1999) proposed that streaks left in the wake of a moving object may be detected by orientationally selective mechanisms (with appropriate temporal integration) and used to help disambiguate motion direction. We have measured thresholds for discriminating the direction of motion of a field of moving dots superimposed on orientationally-filtered noise, oriented either parallel or orthogonal to the dot motion. Orthogonal noise had little effect on direction discrimination thresholds, while parallel noise raised them by a factor of about 8. The detrimental effect decreased with increasing orientation-bandwidth for the parallel noise, and increased with bandwidth for the orthogonal noise. Parallel noise specifically impeded motion direction discrimination, having virtually no effect on either contrast thresholds or speed discrimination. We also measured direction discrimination thresholds for random fields of dot pairs, that were oriented either coherently to form Glass patterns, or at random. Motion direction thresholds were considerably higher when the dot-pairs were oriented coherently and near the direction of motion than when they were oriented randomly. Similar results were obtained for motion along a spiral trajectory. The results provide direct evidence that motion traces, efectively ‘speedlines’, give information about motion direction that is used by the visual system. The presence of oriented masks, either in the form of orientation-filtered noise or coherent Glass patterns confounds this information source, either by reducing sensitivity to the motion traces or by producing false traces that mislead motion mechanisms.
This PDF is available to Subscribers Only