Meeting Abstract
Animal collective behavior is a result of the intricate conflict between the natural variability among the individuals comprising a group, and the homogenizing effect of the group, working to generate synchronization and maintain coherence. Deciphering the bi-directional interactions between individual and group properties is essential for understanding the swarm phenomenon. Here, these conflicting, complex interactions were studied using marching locust nymphs under controlled laboratory settings. Novel experimental and analysis methods were applied to compare single animals, small groups of locusts, and virtual groups composed of real, randomly shuffled and thus non-interacting, members. This unique approach revealed two types of behavioral characteristics: 1) Traits that were under the homogenizing effect of the group, i.e. differing between single animals and groups, but not between group types. These traits were classified as essential for the formation of the swarm and the collective behavior; and 2) Traits retaining individual heterogeneity. These were responsible for the distinctive, group-specific behavioral characteristics, reflected in large inter-group and small intra-group variance (when compared to the virtual groups). Markov-chain models were used to identify social interaction networks within the group, as well as to confirm the hypothesized interplay between variance at the individual level and the emerging group-specific behavioral properties.
