Inferring the statistical physics of collective biological behaviour
The coordinated flight of bird flocks is a striking example of
collective behavior in biology. Using 3D positions and velocities of
large natural flocks of starlings, I will show how to build an
explicit mapping of flock behaviour onto statistical physics models of
magnetism. Learning the parameters of these models allows us to infer
the local rules of alignment, and to reveal that flocks are poised
close to a critical point, where susceptibility to external
perturbations is maximal. Extending the approach to alignment dynamics
shows that flocks are in a state of local quasi-equilibrium.