Personality, Individuality, and the Social Lives of Bats

Abstract

Individual animals are unique and often differ substantially in their behaviour. Explaining how and why behavioural variation persists among individuals is central to behavioural ecology and requires methods that can capture both stability and uncertainty in complex data. In this thesis, I use a modelling approach to estimate the repeatability of behaviours observed from subjective assessments, standardized personality assays, and long-term monitoring of social interactions in the wild. Using bats as a study system, I investigate the extent of individual behavioural variation, how consistent differences contribute to sociality, and how individuality persists over evolutionary time. In Chapter 2, “Conserving cryptic complexity: Bats with broker roles structure the maternity societies of an endangered bat species”, I examine how individuals with distinct social roles contribute to the cohesion of Little Brown Bat (Myotis lucifugus) maternity colonies. I use a permutation-based network analysis to determine the positions of individuals in the social network across multiple years. My results reveal that a small subset of individuals can consistently act as brokers, meaning they bridge otherwise disconnected roost-area subgroups and maintain population-level social cohesion. Given that the same phenomenon is observed in two separate maternity societies of different sizes, this result suggests that these individualized social roles may be a feature of little brown bat sociality. Further, these results simulate how the loss of a small number of individuals from bat populations post-white-nose syndrome could have devastating impacts on social cohesion. This highlights the need to protect surviving individuals that persist on the landscape following population collapse. In Chapter 3, “Experimentally quantified personality traits are consistent across multiple years yet cannot explain observed differences in the social behaviour of bats on the landscape”, I test whether consistent individual differences in social behaviour among Little Brown Bats are linked to personality. Here I combine Bayesian social relations models of network structure with Bayesian hierarchical models of personality test scores to statistically propagate uncertainty from both social network estimates and behavioural assays into a joint analysis. I find strong evidence that behavioural tendencies are repeatable within individuals across multiple years. This finding is consistent with the concept of personality. However, assayed personality traits do not predict individual differences in social network positions, despite the repeatability of both. These findings suggest that personality and social network position, though both individually consistent, may be shaped by different processes. In Chapter 4, “Statistically upset: Fear response as a species-typical behaviour among neotropical bats”, I extend my behavioural comparison to apply a subjective personality assay to quantify fear-response behaviour in >3400 individuals from 55 Neotropical bat species across four countries. Using Bayesian hierarchical and phylogenetic mixed models, I investigate the sources of among-individual behavioural variation across species at multiple levels. I find that fear-response exhibits weak phylogenetic signal, with closely related species often differing markedly. I also find that species-typical fear-responses remain consistently distinct across geographic regions, suggesting that fear-response is a result of species-level differences and not shared environmental factors. None of the explanations for variation in fear-response investigated yielded positive results. That is, neither individual physiology nor species’ ecological and social traits explained the observed variation. Finally, I show that consistent individual differences are detectable across species, suggesting that species-level divergence in fear-response may arise through the accumulation of within-species individuality.