Mechanisms underlying navigational leadership hierarchies: studies with avian and fish models

For animals moving together in groups, navigating collectively requires group members to reach consensus decisions about, among other things, the direction in which to travel. This thesis focuses on how consensus collective decisions are made when certain individuals have a disproportionate influence on the group’s movement decisions. The specific focus is on hierarchical decision-making during collective movement, whereby an individual's position in the hierarchy corresponds to its relative contribution to the directional decision-making process. I used two model species, homing pigeons (Columba livia) and the Characin fish (Astyanax mexicanus) to explore the mechanisms and consequences of leadership hierarchies, using a combination of GPS tracking and high-resolution video recording. Across several experiments, the following key results emerged: Previous navigational experience contributes to structuring navigational leadership hierarchies in homing pigeons, but only at the most influential positions. Birds with local experience occupy the top hierarchical positions significantly more often than inexperienced birds. This suggests there is an advantage to individuals in having an experienced leader when homing as a group. When leader birds hold highly inaccurate information, leadership hierarchies are spontaneously rearranged causing the original leaders to lose influence over flocks. This demonstrates the importance flexibility in decision-making structures, which may thus prevent the propagation of navigational errors – detrimental to all individuals within the flock – through the hierarchy. Homing pigeon navigational leadership hierarchies can be stable over long periods of time, but only in flocks with high levels of inter-individual variation in age and experience. Flocks homogeneous in age and experience showed no hierarchy stability. This highlights the importance of flock composition in determining the stability of navigational leadership hierarchies, which potentially impacts on the advantages of these decision-making structures. Three-dimensional data are not a necessary requirement for identifying leadership hierarchies in animals that move and interact in three dimensions. A two-dimensional directional correlation delay model was able to correctly identify leader-follower interactions in three-dimensional shoals of Astyanax mexicanus. Overall, this thesis has shed light on the mechanisms of consensus decision-making, involving hierarchies of influence, during group movement. Through this knowledge we can further our understanding of the adaptive significance of collective behaviour.