Recognizing dynamic facial expressions of emotion: Specificity and intensity effects in event-related brain potentials.

Emotional facial expressions usually arise dynamically from a neutral expression. Yet, most previous research focused on static images. The present study investigated basic aspects of processing dynamic facial expressions. In two experiments, we presented short videos of facial expressions of six basic emotions and non-emotional facial movements emerging at variable and fixed rise times, attaining different intensity levels. In event-related brain potentials (ERP), effects of emotion but also for non-emotional movements appeared as early posterior negativity (EPN) between 200 and 350ms, suggesting an overall facilitation of early visual encoding for all facial movements. These EPN effects were emotion-unspecific. In contrast, relative to happiness and neutral expressions, negative emotional expressions elicited larger late positive ERP components (LPCs), indicating a more elaborate processing. Both EPN and LPC amplitudes increased with expression intensity. Effects of emotion and intensity were additive, indicating that intensity (understood as the degree of motion) increases the impact of emotional expressions but not its quality. These processes can be driven by all basic emotions, and there is little emotion-specificity even when statistical power is considerable (N (Experiment 2)=102).