Abstract
Adaptation optimizes vision to a dynamic world, which changes at many timescales, from very transient to semi-permanent. To adapt optimally, the visual system also adjusts at different timescales, with longer-lasting environmental changes producing longer-lasting effects. But how the visual system adapts in this way remains unknown. We examined whether contrast adaptation is controlled by multiple mechanisms, each operating over a different time scale, by testing for spontaneous recovery, a hallmark of multiple controllers first shown in the animal learning literature. We used an objective task, two spatial alternative forced choice detection of a briefly presented (200 msec) Gabor patch (1 cpd). Test trials were alternated with 1.4 sec adapting Gabor presentations, in a "top-up" design. Three adapter contrasts were used: 25% (baseline), 90% (adaptation) and 5% (deadaptation). In each session, subjects (N = 3) first completed six minutes of baseline trials to determine a threshold contrast, where performance was 70% correct, which was used for calculating test contrasts in the remainder of the experiment. Subjects then performed two minutes of trials in the baseline condition, followed by 10 minutes in the adaptation condition, 80 seconds in the deadaptation condition, and a second eight minutes in the baseline condition. Adaptation produced large, reliable decreases in performance from baseline levels. Deadaptation counteracted these effects, as performance quickly increased to return to baseline levels. However, continued testing in the second baseline period revealed a striking second decrease in performance. This second decrease represents spontaneous recovery of the effects of adaptation. Deadaptation likely produced effects in a distinct, shorter-term controller that cancelled effects of initial adaptation in a longer-term mechanism. As the shorter-term effects decayed, the ongoing longer-term effects reemerged. These multiple temporally tuned controllers may allow vision to adapt optimally to environmental effects that arise at different timescales.
Meeting abstract presented at VSS 2013