Abstract

Models of motion-sensing units that are local or nearly local in their measurements suffer from a problem that the actual meaning of a unit’s firing is ambiguous. In order for this ambiguity to be resolved, the outputs of many units must be combined and compared. In the experiments described here the predictions of one such model of this process were tested. The effect of vertical motion was measured for patterns moving horizontally. It was found that there was a strong suppression from the vertical motion that could not be accounted for by the simple addition of noise to the motion signal. Further, this suppressive effect was found to be asymmetric in time, in that motions occurring for as much as 300 msec after the vertical motion were affected. whereas those almost immediately before the motion were unaffected. I suggest that such results are predicted by a cooperative–competitive network of interactions between local motion measurements.

© 1989 Optical Society of America

Motion interference in speed discrimination

Samuel F. Bowne, Suzanne P. McKee, and Donald A. Glaser
J. Opt. Soc. Am. A 6(7) 1112-1121 (1989)

Discrimination of opposite directions measured with stroboscopically illuminated random-dot patterns

B. De Bruyn and G. A. Orban
J. Opt. Soc. Am. A 6(2) 323-328 (1989)

Interaction of first- and second-order direction in motion-defined motion

Johannes M. Zanker and Nicholas R. Burns
J. Opt. Soc. Am. A 18(9) 2321-2330 (2001)

Motion processing by chromatic and achromatic visual pathways

Andrei Gorea and Thomas V. Papathomas
J. Opt. Soc. Am. A 6(4) 590-602 (1989)

Computational structure of a biological motion-detection system as revealed by local detector analysis in the fly’s nervous system

Martin Egelhaaf, Alexander Borst, and Werner Reichardt
J. Opt. Soc. Am. A 6(7) 1070-1087 (1989)