Semantic negotiation in dialogue: the mechanisms of alignment

A key problem for models of dialogue is to explain how semantic co-ordination in dialogue is achieved and sustained. This paper presents findings from a series of Maze Task experiments which are not readily explained by the primary co-ordination mechanisms of existing models. It demonstrates that alignment in dialogue is not simply an outcome of successful interaction, but a communicative resource exploited by interlocutors in converging on a semantic model. We argue this suggests mechanisms of co-ordination in dialogue which are of relevance for a general account of how semantic co-ordination is achieved.