Simultaneously and sequentially presented arrays evoke similar visual working memory crowding

ABSTRACT In visual crowding, an item representation is degraded by adjacent flanker items. Recently, the related phenomenon of visual working memory (VWM) crowding has been used to evaluate shared mechanisms between memory and perception. However, some previous studies that investigated VWM crowding suggested that it stemmed from encoding, rather than memory maintenance. In the current study, we evaluated two measures in simultaneously-presented arrays: anisotropy for radially vs. tangentially configured arrays, and effect of target to array proximity (array middle vs. edge targets). Simultaneously presented arrays evoked effects in both measures. We then compared the data from the current study to that from our previous study that used sequential presentation, and thus avoided encoding-based explanations for crowding. We predicted that we would observe greater crowding for simultaneous than sequential presentation because simultaneous arrays allow for two opportunities for crowding—encoding and maintenance—while sequential arrays only allow for maintenance-based crowding. Surprisingly, we observed that both measures were similar across simultaneous and sequential arrays. These results indicate that VWM crowding does not have an additive error mechanism across encoding and maintenance. Moreover, the anisotropy result suggests that both simultaneous and sequential array VWM crowding is influenced by retinotopy in the early visual cortex.