Conference on Cognitive Computational Neuroscience

The superior temporal sulcus (STS) is involved in the processing of changeable aspects of faces as well as in auditive and multimodal processing. We measured fMRI while participants attended to naturalistic, emotional videos of people telling their life-stories in auditive, audiovisual and visual conditions. We used inter-subject correlation and functional connectivity to look how the audiovisual presentation of faces and voices modulates responses in the brain. We found a cluster in the dorsal STS responding more to audiovisual than visual or auditive stimuli. Connectivity between this and lateral occipital areas was heightened when audiovisual stimuli was present, and it seemed to process facial action units of the mouth region. Our results suggest that, specifically when speech is present, dorsal STS connects to the more posterior face processing areas and processes perceived mouth movements. We thus replicate and add up to earlier similar findings in a more naturalistic context with different analysis methods.