Centering human cognition in epidemiological models

Brian Beckage , Louis J Gross, Ari Freedman, Laurent Hébert-Dufresne, Suzanne Lenhart, Chadi Saadroy, Charles Sims

Epidemiological models (EMs) have traditionally treated human behavior as static or as a simple function of disease prevalence, limiting their ability to capture disease trajectories such as those observed during the COVID-19 pandemic. We argue that infectious disease is fundamentally a coupled human-pathogen system in which cognition and behavior operate on different timescales and must be represented as distinct dynamic processes. We propose cognitive epidemiological models (CEMs) that represent cognitive states as continuously evolving state variables alongside traditional disease compartments. We identify four cognitive processes that mediate behavioral response to disease: perceived risk, social influence, culture, and learning. Separating cognition from behavior allows CEMs to distinguish information-based interventions that shape cognitive states from resource-based interventions that enable behaviors, providing insight into why intervention strategies succeed or fail and how they might be optimally timed and targeted.