Collective biological motion: connecting empirical and theoretical approaches

点击翻译按钮获取中文摘要

From cells to humans, aggregates form striking collective patterns, displaying self-organized coordination, cohesion, and collective intelligence. Due to the inherent complexity in these aggregates, deducing the underlying interactions and effects that determine individual movement paths remains a challenge for most groups. However, recent technological advances in image acquisition have led to new datasets that can completely and accurately track each individual member of large collectives (even at the scale of cells), and so we are poised to answer a host of questions concerning how form relates to function of the group, how local and global information is reconciled, what precise interactions govern individual motion, and how the collective is shaped by the ecology in which it exists. My research program uses individual-based mathematical models, coupled with empirical data of individual-based collective biological motion, to address these questions. I use data to generate statistical measures that can, in turn, be used as a blueprint for constructing and validating theoretical models; conversely, models allow formulation, exploration and testing of hypotheses to be verified by observations. Where empirical data are scarce or missing, I design experiments, and collect and analyze collective motion datasets. In one application, I have completely reconstructed trajectories of aquatic birds (surf scoters) within foraging flocks of a few hundred individuals (among the first of such detailed data sets), and used these observations to precisely infer governing interactions. Using the empirical-theoretical approach, I intend to study real groups, ranging from cell colonies to bird flocks, to look at phenomena such as phase transitions and directional information transfer, and to investigate collective behaviour in the context of biomechanics, environment, navigation, and how ecological context shapes these behaviors on evolutionary timescales.