Abstract
The efficiency of averaging properties of sets of objects without encoding redundant details bears many similarities to previous Gestalt proposals that perception is parsimoniously organized as a function of recurrent order in the physical world, suggesting that grouping and averaging are part of a broader set of strategies the human visual system has developed to alleviate capacity limitations to represent visual information in the external environment. To examine how Gestalt grouping affects the manner in which information is averaged and remembered, I compared the error in observers' adjustments of remembered sizes of individual objects within and between sets defined by different Gestalt groupings. Observers viewed a study display of two sets of 8 heterogeneously-sized circles with different mean sizes, grouped by proximity, similarity, connectedness, or common region, followed by a test display of 6 homogeneously-sized test circles (3 per set) and adjusted the test sizes to match the remembered sizes of corresponding circles in the previous study display. As predicted, errors for items within the same Gestalt-defined group were more similar than errors between groups, such that individual circle sizes were recalled with bias towards respective set averages. Surprisingly, the duration of the study displays (500 ms or 5 seconds) had no significant effects on participants' errors. Furthermore, observers' error patterns could be classified by Gestalt grouping condition with significantly greater than chance (25%) accuracy. Taken together, results suggest that Gestalt grouping facilitates perceptual averaging to minimize the error with which individual items are encoded, optimizing the efficiency of VSTM. Overall, these findings support the proposal that the visual system relies on the canonical structure and statistical redundancy inherent in the surrounding environment to support our illusion of rich and stable perception, and raise intriguing possibilities for predicting how observers will encode and store sets of visual information.
Meeting abstract presented at VSS 2016