Ensemble coding of crowd emotion: Differential hemispheric and visual stream contributions

The visual system takes advantage of redundancies in the scene by extracting summary statistics from a set of items. Similarly, in many social situations where scrutinizing each individuals expression is inefficient, human observers make snap judgments of crowds of people by reading \"crowd emotion\" to avoid danger (e.g., mass panic or violent mobs) or to seek help. However, how the brain accomplishes this feat remains unaddressed. Here we report a set of behavioral and fMRI studies in which participants made avoidance or approach decisions by choosing between two facial crowds presented in the left and right visual fields (LVF/RVF). Participants were most accurate for crowds containing task-relevant cues: avoiding angry crowds and approaching happy crowds. This effect was amplified by sex-linked facial cues (angry male/happy female crowds) and highly lateralized, with better recognition of the task-congruent facial crowd when presented in LVF. fMRI results showed that the dorsal visual stream was preferentially activated in crowd emotion processing, with intraparietal sulcus and superior frontal gyrus predicting behavioral crowd emotion efficiency, whereas the ventral visual stream showed greater involvement in individual face emotion processing, with fusiform cortex activity predicting the accuracy of decisions about individual face emotion. Our results shed new light on the distinction between global vs. local processing of face stimuli, revealing differential involvement of the left and right hemispheres and the dorsal and ventral pathways in reading crowd vs. individual emotion.