Abstract
The effects of selectively attending to moving stimuli appear to depend on the type of motion involved; while some high level motion relies on attending to the stimuli other, lower level motion requires minimal attention. However, it is unclear whether perception of biological motion, a complex yet apparently automatic process, requires attention and if so, where in the processing stream do attention's effects occur? Here we used functional magnetic resonance imaging (fMRI) and high-density event related potentials (ERPs) in the same subjects (N=13) during separate testing sessions to examine the role that selective attention plays in modulating the response to biological motion. Participants viewed point-light animations of spatially overlapping human and tool motion or either category of motion overlapped with scrambled versions of the other category of motion. Attention was directed to either human or tool motion in separate blocks using a 1-back task. Group analysis of fMRI data indicated that right superior temporal sulcus and left intraparietal lobule showed greater response to biological motion than tool motion. Alternately, there was greater response to tool motion in bilateral ventral temporal areas and left middle temporal gyrus compared to biological motion. However, the response in these motion category-preferring regions did not appear to be reduced when attention was directed away from the preferred motion category. ERPs showed a similar pattern of results and revealed larger N1 response to biological than tool motion (“N1bm”) over left centro-parietal and right temporo-parietal sites. A polarity-reversed component at a similar latency to the N1bm was observed over midline posterior sites that was greater to tool motion than biological motion. Consistent with the fMRI data, the early responses were not dependent on attention; instead, attention modulated responses after 400ms. These results indicate that initial responses to biological motion and tool motion are not reliant on attention.
This study was supported by grant DAAD-19-01-C0065 from the US Army Research Laboratory.