Abstract
Visual objects presented under interocular noise suppression take different amounts of time to emerge from suppression depending on their familiarity and recognizability (e.g., upright faces emerge from suppression sooner compared to inverted faces. Jiang, Costello, & He, in press). This paradigm is capable of revealing automatic processing of certain stimulus features. Here we examined whether the local motion component of the point-light biological motion (BM) sequence could be extracted automatically. A standard high contrast dynamic noise pattern was presented to one eye, and a test motion sequence was introduced to the other eye. The test motion sequences could be upright BM, inverted BM, or their spatially scrambled counterparts. Interestingly, an upright scrambled BM took less time to emerge from suppression compared to an inverted scrambled BM against the identical suppression noise. This result suggests that the visual system can extract local motion information from scrambled BM in the absence of any global configural information. This is even more surprising given that observers could not explicitly distinguish between upright and inverted orientations of scrambled BM in a 2AFC task. We further measured the cortical responses using fMRI while observers explicitly viewed the BM sequences. Contrasting between intact point-light BM and scrambled control motion revealed strong activation in pSTS, LOC, hMT+, and EBA. More importantly, V3/V3a and pIPS were found to be sensitive to the orientation of the scrambled BM, responding more strongly to upright than inverted scrambled BM, even though these areas showed stronger activation to scrambled than intact BM. These results suggest that the local motion information in BM, independent of the global configuration, can be extracted automatically by the visual system. The current study also highlights the usefulness of the interocular suppression paradigm in isolating the feed-forward processes of visual information processing.
This research was supported by the James S. McDonnell foundation, the US National Institutes of Health Grant R01 EY015261-01, and the Eva O. Miller Fellowship from University of Minnesota.