Abstract
It is well known that Gestalt grouping principles facilitate visual perception. However, it is unclear how or to what extent these grouping principles benefit visual working memory (VWM). Because VWM is subject to strict capacity limitations, anything that facilitates VWM could serve to significantly expand VWM capacity. Previous findings show that at least two grouping mechanisms, proximity and physical connectedness, can improve VWM accuracy (e.g. Woodman, Vecera, & Luck, 2003). Here, we tested VWM performance when visual stimuli could be grouped by proximity and/or similarity of color. In Experiment 1, we briefly presented super-capacity arrays (16 items: 4 items of 4 colors) to undergraduate participants. Stimuli were either grouped according to color or the colors were intermingled. After a brief delay (1000 ms) a single probe item appeared. Participants judged whether the color of the probe item was the same or different. When the stimuli were grouped according to color VWM accuracy was significantly higher than when items were not grouped. These data replicated previous findings showing VWM can also benefit from Gestalt grouping. Yet, other grouping manipulations were not so robust. In Experiment 2, we investigated whether the presence of multiple grouping principles in the same visual array could result in additive benefits for VWM performance. Using 6 horizontally arrayed color patches we tested grouping by similarity and proximity using four conditions: no grouping, color grouping, proximity, and color + proximity. VWM accuracy was significantly modulated by grouping, but in an unexpected direction. VWM accuracy in the color + proximity condition was significantly lower than either of the other two grouping conditions. These results indicate that Gestalt principles of perceptual grouping are not automatically instantiated in VWM and consequently, do not always enhance VWM capacity.
Meeting abstract presented at VSS 2012