|dc.description.abstract||The alignment of our linguistic representations with those that we interpret from other’s utterances allows us to form accurate predictions of their upcoming utterances using our own language production mechanism. The integration of these predictions with our own production of utterances enable us to produce more fluent and coherent responses, allowing for more effective communication of ideas in dialogue (Pickering & Garrod, 2007). Past research has suggested that when cooperating with a partner, we integrate the others’ action representations with ours more strongly, whereas in competition we keep self and others’ representations separate (e.g. Decety et al., 2004). A shared two-colour Stroop task investigated whether this occurs in language. It was predicted that when the colour of the word did not match the description of the word (e.g. ‘red’ presented in the colour green, incongruent condition) there would be interference when the two participants were cooperating due to strong integration of prediction of their partner’s word with the production of their own. Conversely when competing, participants would integrate to a lesser extent the representations of the other’s predicted response with the representations of their own responses, leading to reduced interference in the incongruent condition. This effect was found in the group of participants who performed the cooperative condition first, but not in those performing the competitive condition first.
The implications of these results are discussed in relation to past research on language and action processes that argue that in competition, we use additional processes to keep our own and other’s representations separate. In cooperation however, we merge other’s representations with our own to effectively coordinate our responses with others, leading to interference in the Stroop task, but arguably leading to more coherent communication in natural dialogue.||en