Purchase this article with an account.
Hörmet Yiltiz, David Heeger, Michael Landy; Contingent adaptation in masking and surround suppression. Journal of Vision 2018;18(10):259. doi: https://doi.org/10.1167/18.10.259.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Background: Adaptation describes changes in neuronal gain in response to recent input history. The classic gain-control model of adaptation predicts that a neuron's output depends only on its own recent input. A new model, the Hebbian normalization model (Westrick, Heeger and Landy, 2016), suggests that the normalization weight, whereby one neuron inhibits another, is correlated with the recent history of the product of their responses. This suggests a behavioral correlate: If two stimuli co-occur during adaptation ("contingent adaptation"), each stimulus should be more effective at suppressing the other. We test this prediction in two experiments. Method: Experiment 1 (overlap masking): observers adapted to two co-located high-contrast gratings (e.g., ±15 deg orientation) presented either (1) simultaneously (a plaid) alternated in time with a blank (±15 plaid, blank, ±15 plaid, blank, …; contingent adaptation) or (2) alternating in time (+15, -15, +15, -15, …; asynchronous). After adaptation, observers performed a contrast-detection task for one of the gratings at low contrast masked by a co-located, high-contrast grating at the other orientation (spatial 2IFC). Experiment 2 (perceived contrast): Adapters were a pair of gratings (a center disk and surrounding ring) alternating between the two orientations. In one location, center and surround had the same orientations (+15/+15 alternated with -15/-15); in the other location, they had the opposite contingency (+15/-15 alternated with -15/+15). After adaptation, observers were shown high-contrast surround gratings with intermediate contrast center gratings, and chose the side with higher center contrast (spatial 2IFC). Results: Experiment 1: Contingent adaptation raised detection threshold compared to asynchronous adaptation. Experiment 2: Perceived contrast of the center grating was reduced when the pair of center/surround orientations matched the contingency in that location present during adaptation. Conclusions: Our results support the Hebbian normalization model of adaptation.
Meeting abstract presented at VSS 2018
This PDF is available to Subscribers Only