Dresden 2026 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
SOE: Fachverband Physik sozio-ökonomischer Systeme
SOE 4: Poster Session
SOE 4.4: Poster
Montag, 9. März 2026, 18:00–21:00, P4
Dynamics of Behavioral Contagion in Human and Artificial Agent Collectives: A Drift-diffusion Simulation Study — •Maryam Karimian1,2 and Pawel Romanczuk1,2 — 1Institute for Theoretical Biology, Department of Biology, Humboldt-Universität zu Berlin, 10115 Berlin, Germany — 2Science of Intelligence, Research Cluster of Excellence, Berlin, Germany
Behavioral contagion, the spread of behavior within a group, is a central topic in cognitive and collective behavior research. Prior work has either relied on first-person perspective that overlook dynamic social structure or on idealized collective models that neglect individual decision processes. To bridge this gap, we use an extended drift-diffusion model to simulate virtual-reality (VR) experiments, integrating cognitive mechanisms with group-level dynamics. We model agents’ decision-making processes, responding to uncertain environmental signals, and influencing one another through an increment drift-diffusion process. Decision states evolve based on accumulated environmental evidence and socially conveyed information. Social information is derived from visually grounded interaction networks capturing first-person sensory constraints shaped by distance and occlusion. These networks are implemented as weighted, directed graphs modulated by group size and density. Using controlled experimental data, we parameterize the model and generate testable predictions on decision dynamics across different environmental and social conditions, highlighting how visual network properties shape vision-mediated contagion and related cognitive processes such as selective attention.
Keywords: behavioral contagion; Drift-diffusion model; visual network