The role of the visual information of fish schooling via selective decision-making

Visual cues play crucial roles in the collective motion of animals, birds, fish, and insects. Recently, experiments have revealed that organisms such as fish selectively utilize a portion, rather than the entirety, of visual information. This method of the visual interaction avoids heavy load for small brain of the organisms. However, the previous models using visual interaction implicitly assume that an agent interacts with all visible neighbors. Therefore, we study the effect of the selective decision-making on the collective motion via the agent-based model and the coarse grained continuous model. In the former study, we have constructed a visual model which takes into account the motion of visual attention of agents induced by the visual stimuli, and our model can simultaneously show the spontaneous appearance of various collective patterns and the bifurcation process of the tracking of a neighbor. The later study, the agents corresponds to the density field by the coarse graining, and the visual occlusion is treated in a self-consistent manner via a coarse-grained density field, which renders the interaction effectively pairwise. The model exhibits a discontinuous transition as in the conventional models by the local collision, and but the discontinuity is weakened by the non-locality of visual interaction. Our studies clarify the comprehensive coincidence with experimental results via selective decision-making and the essential role of non-locality in the visual interactions.