The ability to extract and store ensemble representation helps us to efficiently process and remember complex visual information, despite the limited capacity of our visual working memory. Previous studies on visual working memory have focused mostly on objects, rather than on ensembles. The current study investigated the storage limitations and storage format of ensemble representation using a change detection paradigm. In a series of experiments, participants viewed multiple sets of circles grouped by spatial proximity and memorized the mean size of each set. Each set contained six circles, but with different sizes. Participants reported whether the mean size of memorized set was identical to the mean size of the probe set. Results showed that visual working memory could stably maintain mean sizes of approximately two sets for at least four seconds. The working memory performance of ensembles was not affected by the number of individuals in the ensemble, indicating that individuals were not stored in working memory. In addition, the memory of ensemble representations did not depend on the precise memory of the individual objects. Taken together, these results suggest that ensembles are not stored as a simple composition of their individual elements, but as more abstract statistical information.

Authors:
Chaoer Xu, Zhejiang University, China
Jifan Zhou, Zhejiang University, China

About the Presenter(s)
Ms Chaoer Xu is a University Doctoral Student at Zhejiang University in China

See this presentation on the full schedule – On Demand Schedule

Bookmark