Conversational production and comprehension: fMRI-evidence reminiscent of the classic Broca-Wernicke model

A key question in neurolinguistics is whether language production and comprehension share neural infrastructure, but this question has not been addressed in the context of actual conversation. We utilized a public fMRI dataset where participants (N=24) engaged in unscripted conversations with a confederate outside the scanner via an audio-video link. We provide evidence indicating that production and comprehension, in a conversational setting, diverge with respect to how they modulate the recruitment of regions in the left-lateralized perisylvian language network. Activity in the left inferior frontal gyrus was stronger in production than in comprehension. Compared to production, comprehension showed stronger recruitment of the left anterior middle temporal gyrus and superior temporal sulcus, but this was not the case for the posterior aspect of these loci. Although our results are reminiscent of the classic Broca-Wernicke model, the anterior temporal activation is a notable difference from that model. This is one of the findings which may be a consequence of the conversational setting, another being that conversational production activated what we interpret as higher-level socio-pragmatic processes. In conclusion, we present evidence supporting that the above-mentioned frontal vs temporal regions in the language network are functionally segregated during conversation.