Learning Time-Varying Turn-Taking Behavior in Group Conversations
This work addresses the challenge of generalizing turn-taking predictions beyond single groups for researchers in conversation analysis, though it appears incremental as a generalization of prior models.
The authors tackled the problem of predicting turn-taking patterns in group conversations by developing a probabilistic model that uses individual characteristics and past speaking behavior. Their approach demonstrated that previous behavioral models may not always be realistic, motivating a data-driven yet theoretically grounded method.
We propose a flexible probabilistic model for predicting turn-taking patterns in group conversations based solely on individual characteristics and past speaking behavior. Many models of conversation dynamics cannot yield insights that generalize beyond a single group. Moreover, past works often aim to characterize speaking behavior through a universal formulation that may not be suitable for all groups. We thus develop a generalization of prior conversation models that predicts speaking turns among individuals in any group based on their individual characteristics, that is, personality traits, and prior speaking behavior. Importantly, our approach provides the novel ability to learn how speaking inclination varies based on when individuals last spoke. We apply our model to synthetic and real-world conversation data to verify the proposed approach and characterize real group interactions. Our results demonstrate that previous behavioral models may not always be realistic, motivating our data-driven yet theoretically grounded approach.