More Features in Bound Representations Does Not Require Extra Object-based Attention in Working Memory

PURPOSE: Feature binding is a core concept in many research fields, including the study of working memory (WM).We recently proposed that binding in WM is not passive, but requires more object-based attention to actively bind distinct single features into a coherent unit (Gao et al., Attention, Perception, & Psychophysics, 2017; Shen, Huang, & Gao, Journal of Experimental Psychology: Human Perception and Performance, 2015). However, a hallmark of object-based attention—the amount of attention is not modulated by the number features contained in an object—has not been examined. In the current study, we closed this gap by examining whether this hallmark of object-based attention still holds in WM. METHODS: In two experiments, we required the participants to memorize three bound representations, and manipulated the number of features (2 vs. 3 features) contained in each binding. To examine the role of object-based attention in retaining bindings in WM, we also manipulated whether a secondary task consuming object-based attention was interpolated into the maintenance phase of WM (with vs. without secondary task). If more object-based attention was required after an extra feature was added into the bound representation, then the secondary task would result in worse performance for 3-featured binding than 2-featured binding. RESULTS: In two experiments, we consistently found that the added secondary task significantly impaired the binding performance. However, the added secondary task impaired the 2-featured and 3-featured bindings to the same extent. CONCLUSION: The number of features contained in binding does not modulate the required object-based attention for binding in WM, suggesting that WM and perception share the same hallmark of object-based attention.