Fission-fusion behavior, which is widely reported in social animals, has been considered as a mechanism for adapting to changing environmental conditions. Although several hypotheses have been proposed to explain the potential benefits of fission-fusion behavior, there are only a few theoretical studies that have systematically explored its mechanism or quantitatively examined the potential forces shaping its evolution. We developed a social learning model to investigate the mechanism and evolutionary forces that underlie a fission-fusion society. In particular, we focused on the day-roost choices of bat individuals because bat societies represent one of the most sophisticated fission-fusion systems. The assumptions of the study were as follows. Each individual selects a single day-roost to use, and forms a roosting group with roost mates. Bats randomly choose a roost to visit in order to inspect its quality. Inspection is not always accurate, i.e., it includes some error. After inspection, bats return to the current day-roost and share the new information with roost mates. Each bat estimates the quality of each potential roost by social learning and chooses which one to use based on the relative value of expected roost quality. The size distribution of sub-colonies is determined by this choice behavior. Three roost-switching behaviors (settlement, synchronized movement, and fission-fusion grouping) were predicted depending on two factors (the level of difficulty of evaluating roost quality and the capacity to remember roost quality information). Settlement behavior, in which most bats remain in the best roost, led to the highest fitness because the accuracy of estimating roost quality was improved when bats exchanged information with members in a large group. However, when disease transmission was combined with learning dynamics, the cost of infection significantly increased under both settlement and synchronized movement behaviors, and eventually fission-fusion behavior led to the highest fitness. These results highlight two conflicting factors: learning in a large group improves information accuracy, but living in a small group effectively reduces the risk of spreading disease. Dynamic change of group size by fission-fusion can resolve the dilemma between these two conflicting factors.
All Science Journal Classification (ASJC) codes
- Statistics and Probability
- Modelling and Simulation
- Biochemistry, Genetics and Molecular Biology(all)
- Immunology and Microbiology(all)
- Agricultural and Biological Sciences(all)
- Applied Mathematics