The objecthood of visual working memory representations influences their interaction with attention

Visual working memory (VWM) guides attentional selection toward memory-matching information in the environment. However, in the classical VWM-driven attentional capture paradigm, this effect is substantially reduced when multiple representations are simultaneously maintained. Although this capacity limitation has been attributed to a limited number of active VWM representations, it remains unclear what constitutes the functional unit underlying this limitation. Here, we examined whether objecthood enables multiple features to function as a single unit for attentional guidance. Across five experiments, participants memorized multiple features that either belonged to the same object or to separate objects before performing a visual search task containing memory-matching distractors. Replicating previous findings, multiple features failed to produce VWM-driven attentional capture when they belonged to separate objects. Critically, when the same features were integrated into a single object, they reliably guided attention. This objecthood effect generalized from two to three colors, extended to features from different dimensions (color and orientation) and different modalities (visual and verbal color representations), and could not be explained by differences in memory fidelity. These findings suggest that the capacity limitation in VWM-driven attentional capture operates over integrated representational units rather than individual features, with objecthood serving as an effective organizational principle that integrates multiple features into a single unit for attentional guidance.