Iain Couzin - Cognitive collectives - IPAM at UCLA

Collective behavior research is shifting from simple statistical mechanics models toward neurobiological frameworks that account for individual cognition. Traditional models, such as the V-check and Cousin models, rely on explicit alignment rules that lack empirical support in nature. Instead, animal navigation and decision-making emerge from ring attractor networks in the brain, which process spatial information through dynamical bifurcations. These bifurcations allow individuals to break symmetry and make rapid, accurate decisions when faced with complex environmental choices. Experimental data from fruit flies, locusts, and zebrafish—often using virtual reality to isolate variables—demonstrate that this embodied process is a fundamental, non-Euclidean computational principle. This neuro-centric approach challenges long-held assumptions about collective motion, offering a more robust, data-driven understanding of phenomena like locust swarming and providing a path toward more effective, science-based interventions for global agricultural crises.