Automatic Facial Mimicry of emotion during perception of Song

We examined facial responses to audio-visual presentations of emotional singing. Although many studies have now found evidence for facial responses to emotional stimuli, most have involved static facial expressions and none have involved singing. Singing represents a dynamic ecologically valid emotional stimulus with unique demands on orofacial motion that are independent of emotion, related to pitch and linguistic production. Observers’ facial muscles were recorded with electromyography while they saw and heard recordings of a vocalist’s performance sung with different emotional intentions (happy, neutral, and sad). Audio-visual presentations successfully elicited facial mimicry in observers that were congruent with the performer’s intended emotions. Happy singing performances elicited increased activity in the zygomaticus major muscle region of observers, while sad performances evoked increased activity in the corrugator supercilii muscle region. These spontaneous facial muscle responses occurred within the first three seconds following onset of video presentation indicating that emotional nuances of singing performances can elicit dynamic facial responses from observers.