How Is Flocking Like Computing? | Quanta Magazine

collective-behaviorswarm-intelligencebiological-computationemergencecomplex-systems

Abstraction: Animal flocking as distributed biological computation and emergence

Key points:

Iain Couzin (Max Planck Institute) argues the common thread across all collectives — from placozoan cell aggregates to bird flocks — is computation: individuals process information together in ways no individual can alone
Information waves in fish schools and bird flocks propagate ~10x faster than a predator's maximum speed, allowing individuals to respond to threats they cannot directly detect
Natural selection tunes collective systems near phase transition / bifurcation points to simultaneously achieve robustness and hypersensitivity to inputs
Under predator threat, schooling fish do not become more individually jittery; instead they reorganize the social network topology to optimize information propagation
Locust marching swarms emerge from cannibalism: individuals follow fleeing neighbors and flee from approaching ones, producing directed collective motion out of nutrient-poor areas
Fish brains represent space in a non-Euclidean coordinate system; near bifurcation points this amplifies differences between remaining options, reducing decision complexity

Connections: Iain Couzin · Steven Strogatz · Collective Behavior · Emergence · Biological Computation · Phase Transitions