Emotional speech synchronizes brains across listeners and engages large-scale dynamic brain networks

摘要: Speech provides a powerful means for sharing emotions. Here we implement novel intersubject phase synchronization and whole-brain dynamic connectivity measures to show that networks of brain areas become synchronized across participants who are listening emotional episodes in spoken narratives. Twenty participants' hemodynamic activity was measured with functional magnetic resonance imaging (fMRI) while they listened 45-s narratives describing unpleasant, neutral, pleasant events neutral voice. After scanning, the again rated continuously their feelings pleasantness–unpleasantness (valence) arousal–calmness. Instantaneous (ISPS) were computed derive both multi-subject voxel-wise similarity inter-area (seed-based synchronization, SBPS). Valence arousal time series subsequently used predict ISPS SBPS series. High associated increased auditory cortices Broca's area, negative valence enhanced thalamus, anterior cingulate, lateral prefrontal, orbitofrontal cortices. Negative affected fronto-parietal, limbic (insula, cingulum) fronto-opercular circuitries, positive striatum, amygdala, cerebellum, dorsal frontal cortex. Positive had markedly smaller effects. We propose high synchronizes listeners' sound-processing speech-comprehension networks, whereas circuitries supporting self-referential processing.