SUBCRITICAL COLLECTIVE ESCAPE WAVES IN FISH

Winnie Poel (Personal webpage)

Living systems such as neural networks or animal groups processinformation about their environment via the dynamics of many interactingunits and show transitions between sets of distinct macroscopicbehaviors. While many studies focus on the idea that for biologicalsystems being close to such a transition (or critical point) optimallymanages a trade-off between desired functions of the macroscopicbehaviors it separates, or that it yields optimal computationalabilities, little attention has been given to the fact that the`optimal' configuration will depend on the environmental context. Here,we combine experimental evidence and computer simulations to show thatfor escape waves in schooling fish the distance to a critical point ischanged via a modulation of school density according to group members'perceived risk of the environment. We find that the observed escapewaves are spreading subcritically making the schools collective responseless sensitive to cue intensity but more robust towards false alarmsthan at the critical school density.