A neurocognitive framework for comparing linguistic and musical interaction

Most cognitive research concerned with the relationship between language and music asks whether isolated individuals represent and process them in similar ways. In this paper, we focus instead on the relationship between interactive language and interactive music, and suggest that speakers engaged in dialogue and musicians engaged in joint performance face similar difficulties – how to relate their contributions to their partners in terms of both timing and content. We propose a model that spans interactive language and music in which each interactor constructs a single joint prediction of their own and their partner’s behaviour, and then compares that prediction against the actual behaviour when it occurs. We discuss how predictions differ depending on turn organisation, as well as message spontaneity. We relate this proposal to behavioural and neuroscientific data from interaction research in the domains of both music and language.