The roundtable: an agent-based model of conversation dynamics

What mechanisms drive the group size evolution of social conversations? How does schisming take place in dynamically-changing conversations? Are there any universal driving mechanisms taking place? In this work we address such questions by means of an agent-based model of conversation dynamics. The proposed model develops from simple yet realistic assumptions derived from empirical evidence, it abstracts from conversation content and semantics while including topological and simple psychological information, and it is driven by stochastic dynamics. It describes the turn-taking dynamics of an abstract N-body conversation. We find that a single mechanism, i.e. the dynamics of individual fitness thresholds that characterize the happiness level of each participant, enables the onset of complex collective phenomena such as the self-organized schisming phenomenon. Potential generalizations of the model - including individual traits and preferences, memory effects and more complex conversational topologies - may find useful applications also in other fields of research, where dynamically-interacting and networked agents play a fundamental role.