Examining task-dependent changes of visual working memory representations

Undergraduate Just-In-Time Abstract

Poster Presentation 43.358: Monday, May 20, 2024, 8:30 am – 12:30 pm, Banyan Breezeway
Session: Undergraduate Just-In-Time 2

Safiya grant1, Chunyue Teng1; 1Lawrence University

Task-relevant information maintained in visual working memory (WM) may be used flexibly to address different task demands. This study investigates how the functional relevance of information within visual WM – distinguished as task-relevant content (e.g., actively used and reported) and context (e.g., cues guiding memory retrieval) – may differentially affect visual processing and consequently be influenced by distractor processing. In a dual-task paradigm, participants performed a visual search task during the delay period of a visual WM task. They memorized both the location and orientation of a grating stimulus, followed by a delay, after which they were prompted to recall either the orientation or location of the memory sample. The other stimulus dimension served as context, signaling whether the recalled content required mental manipulation. The visual search task included distractors that matched the content of WM sample, its context, both, or neither, allowing for an examination of WM-perception interaction. Results showed that both content and context within visual WM guide visual processing, evidenced by increased response time in the visual search task when the distractor matched with either dimension of the WM representation. Further, interference to memory recall was modulated by the information’s task-specific function. Recall was most precise when the search distractor was consistent with the WM item in both dimensions, and suffered the most when the distractor matched the context but mismatched the content dimension. Conversely, distractors with a mismatched context caused more swap errors, indicating a disruption to the content-to-context binding. This pattern was consistent across different feature dimensions. Altogether, these results suggest that both content and context of WM representations interact with visual perception bidirectionally. They further underscore the importance of understand the mechanisms of visual WM from a functional perspective.