The Neural Mechanisms Supporting Structure and Inter-Brain Connectivity In Natural Conversation

Abstract

Conversation is the height of human communication and social interaction, yet little is known about the neural mechanisms supporting it. To date, there have been no ecologically valid neuroimaging studies of conversation, and for good reason. Until recently, imaging techniques were hindered by artifact related to speech production. Now that we can circumvent this problem, I attempt to uncover the neural correlates of multiple aspects of conversation, including coordinating speaker change, the effect of conversation type (e.g. cooperative or argumentative) on inter-brain coupling, and the relationship between this coupling and social coherence. Pairs of individuals underwent simultaneous fMRI brain scans while they engaged in a series of unscripted conversations, for a total of 40 pairs (80 individuals).

The first two studies in this dissertation lay a foundation by outlining brain regions supporting comprehension and production in both narrative and conversation - two aspects of discourse level communication. The subsequent studies focus on two unique features of conversation: alternating turns-at-talk and establishing inter-brain coherence through speech. The results show that at the moment of speaker change, both people are engaging attentional and mentalizing systems - which likely support orienting toward implicit cues signaling speaker change as well as anticipating the other person's intention to either begin or end his turn. Four networks were identified that are significantly predicted by a novel measure of social coherence; they include the posterior parietal cortex, medial prefrontal cortex, and right angular gyrus.

Taken together, the findings reveal that natural conversation relies on multiple cognitive networks besides language to coordinate or enhance social interaction.