ABSTRACT
Joint language production is the study of the mechanisms involved in producing language jointly with another real or assumed speaker. We review work in this area from the last decade. We specifically highlight how comparisons between the production of language in joint versus individual tasks have begun to uncover when and how speakers engage in co-representation of others’ utterances. We conclude that this growing body of experimental work shows that the process of producing language is susceptible to influence from representations of others’ utterances (whether real or assumed). There is also good evidence that co-representation makes use of language production mechanisms, but typically stops short of engaging in detailed simulation of what another person is saying. We also identify key open questions and suggest that future work should systematically investigate which factors influence the flexible use of simulation-based mechanisms (e.g., nature of the communicative context, partner's identity, cognitive load) to further our understanding of how speaking jointly differs from speaking in isolation.
