Purchase this article with an account.
Alan L. F. Lee, Hongjing Lu; Phantom motion aftereffect using multiple-aperture stimuli: A dynamic Bayesian model. Journal of Vision 2010;10(7):822. doi: https://doi.org/10.1167/10.7.822.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Using random-dot kinematograms, previous studies found that motion aftereffect (MAE) exists not only in the area of adaptation, but also in non-adapted visual field, a phenomenon termed “phantom MAE”. The present study examined whether phantom MAE also exists for a stimulus comprised of multiple drifting gratings, and to what extent MAE can affect local motion processing. In addition, we developed a computational account of phantom MAE within the framework of Bayesian sequential learning. In Experiment 1, an adapting stimulus exhibited global motion via randomly-oriented drifting gratings in two non-adjacent quadrants. In a subsequent testing stimulus, drifting gratings were shown either in the adapted (to measure concrete MAE) or in non-adapted areas (to measure phantom MAE). For translational, circular and radial adapting motion, phantom MAE was found to be significant, although weaker than concrete MAE. The existence of phantom MAE demonstrates that motion aftereffect is processed in a global manner. In Experiment 2, random motion flow was assigned to the testing stimuli. Grating orientations in the testing stimuli were sampled from uniform distributions (range = ±30 degrees), centered at directions either orthogonal or parallel to illusory motion direction. Observers were asked to discriminate motion direction after adapting to coherent or random motion stimuli. Responses to testing stimuli with orthogonal orientations were different from responses to parallel orientations only after adapting to coherent motion, but not after adapting to random motion, indicating a top-down influence of MAE on local motion processing. A dynamic Bayesian model was developed to quantify adaptation-induced changes on multiple motion channels, each selective to a specific velocity. The simulation showed that prolonged exposure to a moving stimulus changed the width of the tuning function differently for different motion channels. The model predicted the existence of phantom MAE, as well as the qualitative difference between phantom and concrete MAEs.
This PDF is available to Subscribers Only