Abstract
Object identities appear stable and continuous over time despite visual noise, disruptions in visibility, and constantly changing visual input. Recent results have demonstrated that the perception of stimuli such as orientation, facial identity, and facial expression is systematically biased (i.e., pulled) toward visual input from the recent past (Fischer & Whitney, 2014; Cicchini et al., 2014; Liberman et al. 2014; Liberman & Whitney, VSS 2015). The spatio-temporal region over which current orientations or face identities are pulled by previous orientations or identities, respectively, is known as the continuity field. This serial dependence of perception could contribute to the visual stability of objects over short time periods, but it is unknown whether serial dependence can occur for multiple object dimensions in parallel and whether it requires a decision. Here, we tested whether the continuity field plays a role in maintaining the stability of facial expression and identity simultaneously. Subjects saw a series of faces that varied randomly in both identity and expression. Before subjects saw each face, they were cued to which of the two facial dimensions they should attend to (identity or expression). Subjects then had to match either the expression or identity of a test face to the target dimension they had previously attended. Subjects made consistent errors when reporting the perceived expression or identity of the target face, seeing it as more similar to the facial expression or identity presented on the previous trial, respectively. Furthermore, this perceptual pull within a facial dimension occurred regardless of which dimension subjects attended to on the previous trial. Perceived expression or identity was pulled toward previous expressions or identities, independent of any decisions or judgments made on previous trials. These results indicate that the continuity field can operate on multiple facial dimensions in parallel in order to maintain perceived object stability.
Meeting abstract presented at VSS 2016