Spatial proximity and object-based grouping effects on visual working memory

Performance in visual working memory is strongly influenced by how visual features are chunked across and within objects. For example, previous studies have demonstrated improved performance when items are presented spatially close to each other (Xu, 2006), as well as when multiple features are combined into the same object (Luck and Vogel, 1997; Oberauer & Eichenberger, 2013). However, limited evidence exists about the interaction between spatial and object-based grouping. To address this gap, we conducted a delayed match-to-sample experiment (N=71) to investigate the interaction of these two grouping cues on visual working memory. Participants had to remember the colors of 6 items grouped into pairs by their spatial arrangement or connecting lines. There were 4 grouping conditions: no grouping, 3 pairs grouped by spatial arrangement only, 3 pairs using object-based grouping only, and both cues. Sub-conditions for the latter included a congruent condition (when the two cues grouped the same pairs) and an incongruent condition (when the two cues conflicted). After a 1000 ms delay, one of the items was probed and participants had to choose between the correct color and a foil, which was always one of the adjacent items. We found that performance was generally better with either spatial or object-based grouping cues than with no cues. Interestingly, the combination of these cues did not yield any additive benefit, with spatial cues exerting a stronger influence in conflict situations. This finding aligns with the idea that attending nearby locations is more efficient than spreading attention through an object. Moreover, errors occurred more frequently when foils were outside the grouped pairs across all conditions, suggesting that the underlying mechanism involves reducing confusion within the group rather than between groups.