地点：心理学院301会议室

Biological motion (BM), which broadly refers to the movements of animate entities, is a social signal that carries information critical for survival and interpersonal interactions. Processing such movements has immense value for successful daily activities, including normal social interaction (e.g., pro-social behavior and nonverbal communication). As noted by Aristotle, “Man is by nature a social animal”. To conform to our social nature, as well as promote our abilities related to BM perception, it is important to store BM information in working memory. This ensures that BM information will not decay from the visual system, and allows us to have coherent visual perception of dynamic BM. A series of studies were performed to investigate the processing mechanisms of BM in working memory. In the first study, we explored the working memory capacity and architecture of BM information by examining the maintenance of BM stimuli in working memory. In the second study, 2 critical issues of BM-related binding in working memory were explored: (a) how many BM binding units can be retained in working memory, (b) whether involuntarily object-based binding occurs during BM binding. In the third study, we investigated how interactive BM (e.g., two people shaking hands) is stored in working memory by testing two opposite hypotheses: (a) that working memory stores the interactions individually (the individual-storage hypothesis) and (b) that working memory stores the interactions as chunks (the chunk-storage hypothesis). The results are as follows. First, an independent storage buffer of working memory exists for BM information, which can hold up to 3 to 4 motion units. Second, working memory capacity for BM bindings (e.g., BM-color binding) is rather low; only 1 or 2 BM bindings could be retained in working memory regardless of the presentation manners. Furthermore, no object-based encoding takes place for BM stimuli regardless of the processed dimensions. Third, the working memory automatically and efficiently uses semantic knowledge about interactive biological movements to store them, and the chunk-storage hypothesis is supported.