Purchase this article with an account.
A. C. Dobbins, J. K. Grossmann; An orthogonal axis aftereffect and its implications for perceptual grouping. Journal of Vision 2001;1(3):396. doi: https://doi.org/10.1167/1.3.396.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
We describe a new aftereffect and use it to explore perceptual grouping. Dynamic dot quartets appear to jump along the horizontal or vertical sides of an implicit square. The axis of motion is ambiguous and bistable. Ramachandran and Anstis (1983) reported that an array of quartets all appear to share a single axis of motion at any instant. Quartet motion can be biased by changing the ratio of path lengths in the two directions. Observers viewed an array of identically biased quartets, followed by an array of unbiased quartets. The transition from biased to unbiased quartets usually elicited a short latency switch to perception of the orthogonal motion axis. Increasing adaptation time increased both the probability of a switch, and the time until next reversal. Introducing a delay between adapt and test phases did not diminish the effect, nor did using complementary arrays in which the test phase quartets were positioned in the interstices of the adapting array's quartets. To examine the basis of grouping using the orthogonal axis aftereffect, observers adapted to biased quartets in one hemifield and were tested in the opposite hemifield. In a second experiment observers simultaneously viewed two orthogonally biased quartet arrays and were tested with two unbiased arrays. The orthogonal axis aftereffect was undiminished by spatial separation of the adapt and test arrays. However, simultaneous adaptation did not produce simultaneous aftereffects —- the two test arrays rapidly assumed a shared axis of motion. Perceptual grouping of dynamic ambiguous objects does not depend critically on local motion detector adaptation, but rather reflects a global bias on inferred motion. We hypothesize that the visual system favors interpretations consistent with those that could accompany perspective shifts during movement in a natural environment. Thus, we fail to see global motion in radial quartet arrays, refusing to be at once coming and going.
This PDF is available to Subscribers Only