Along with Grant Hopcraft, Colin Torney, and Dirk Husmeier, I'm currently seeking promising applicants for our bid for an MVLS/EPSRC PhD Studentship. In total, six studentships will be awarded from the pool of 24 potential projects. Our project would be an exciting blend of experimental biology and modelling to examine the flexible nature of leadership in animal social groups spanning wildebeest herds to fish schools. Here is a rundown:

Highly coordinated movement in social animals raises questions about how collective decisions are made and leadership roles. For example, a bait-ball of fish responding to predator attacks is likely a product of each individual balancing the information about their immediate environment (predator versus no predator) that is nested within a matrix of information from neighbours whose perception is be beyond that of the focal animal. This balanced but hierarchical information structure could lead to coordinated reactionary swarms. In reality, however, these swarms also navigate and problem solve as they move through complex environments. Coordinated navigation suggests that a group of naïve organisms may actually make intelligent and informed decisions about where and when to move via some form of consensus that may be governed by rules of social influence. This poses interesting questions that relate to complexity science, information and communication, and the statistical inference and applied probability in biological informatics. For instance; are small groups equally efficient at navigating complex habitats as large groups? Are multi-species groups better than single species swarms? Does the mantle of leadership change between individuals based on the type of decision required such that different individuals are more decisive under certain scenarios (e.g. some individuals in the group make the left-right decisions, while others are decisive in fight-flight)? Do leaders always assume certain positions in the group, such as core vs peripheral, and are there specific locations where they are most effectual? We propose a PhD project that uses hours of aerial drone videography of wildebeest and zebra migrations in the Serengeti in which the movement of hundreds individuals can be tracked at once using image recognition algorithms (Dr. Hopcraft [IBAHCM] & Dr Torney [Maths]). The detailed analysis of this footage using equation-free modelling techniques will illustrate how movement decisions vary across scales and under a variety of natural conditions these animals face (navigating river crossings, moving through risky habitats, etc). The student will then use schools of aquaria fish (Dr. Killen [IBAHCM]) to experimentally manipulate groups to test ideas about how group size, conspecific characteristics, and relative position alter the roles of leadership and the ability to navigate complex environments. The application of equation-free modelling will train the student in multiscale computation and computer-aided analysis that is particularly informative when the evolution of an emergent property of the system is observed at a macroscopic scale (such as collective decisions by groups), but is dependent on explicit dynamical models that operate at a more detailed fine-scale (individual decisions operating at short time steps). Understanding the mathematics of how natural systems optimize information transfer and group size provides a fertile environment for a PhD student to explore analogous applications in swarm robotics and sensor technology as well as crowd control and emergency evacuations (Dr Torney and Dr Husmeier).

For more info on how to apply, click here

Comment