Answers are grounded in this paper and cite their sources.
In dialogue, interlocutors are mutually responsible for communicating timely, efficiently and successfully. To account for this observation, Clark and Wilkes-Gibbs (1986) proposed their collaborative model, based on the finding that, over time, speakers in a referential matching work together in producing definite references, leading to fewer words and speaking turns to describe abstract figures repeatedly. We aimed to replicate and extend their study with more naturalistic stimuli: affective pictures varying in valence and arousal. We assessed the influence of these pictures on the interlocutor’s affective state and their collaborative reference production. Results support the generalisability of the collaborative model: word use and turn taking declined over the course of the interactions, as did filled pauses, hedges and indefinite references. Curiously, the affective state of participants generally improved during the experiment, regardless of type of affective content discussed. Overall, only limited effects of affective content on collaborative reference were observed.
No takes yet. Share an insight, caveat, or question.
Out et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: