Contact patterns reveal a stable dynamic community structure with fission-fusion dynamics in wild house mice

Abstract

Living in groups is a widely adopted strategy in gregarious species. For group-living individuals it is crucial to be capable to integrate into a social structure. While there is an intuitive understanding that the concept of a group arises through some form of cohesion between its members, the exact definition of what constitutes a group and thus tasks like the detection of the dynamics of a group over time is a challenge. One way of measuring cohesion is through direct interactions between individuals. However, there is increasing evidence that associations between individuals can be mediated by others, and thus, that the drivers for group cohesion extend beyond direct individual interactions. We use dynamic community detection, allowing to relate individuals beyond direct contacts, both structurally and temporally, to study the social structure in a long-term study of a population of free-ranging house mice in a barn in Switzerland. During the 2-year study period, mice had unlimited access to food, and population density increased by 50%. Despite strong fluctuations in individual contact behaviour, population demography and structure embed into long-lived dynamic communities that are characterised by spatial fidelity, persist over several seasons and reproduction cycles, and considerably extend the life-span of single individuals. Within these multi-male and multi-female communities, seasonal changes strongly affect their structure, leading to fission-fusion like dynamics. We identify femalefemale interactions as the main driver for the longevity of these communities, a finding that contrasts with prior reports of the importance of a dominant male for the stability of a group. Moreover, males have a drastically shorter presence time in the study population and more often move between communities than females. Nevertheless, interacting with other breeding males in stable communities increases the duration of male presence and thus, potentially, reproductive success. Our analysis of contact patterns in a rodent that uses shelters to rest, hide and rear offspring emphasises the importance of female-bonded communities in the structuring of the population.

Long-term overlap of social and genetic structure in free-ranging house mice reveals dynamic seasonal and group size effects

“…Though the increase in population size had less effect than predicted, in the within-group networks a larger group size clearly resulted in a lower overlap between social and genetic networks. As there were more social groups present in the barn toward the end of the study period than at the beginning, some groups clearly underwent fission as the population grew (see also Liechti et al. 2020 for detailed quantification of changes in group structure during some of the years analyzed here).…”

“…2012 ). The number of individuals in a box can also vary depending on the level of breeding activity, with a single or small group of males more likely to monopolize a nest box during periods of high breeding activity ( Liechti et al. 2020 ).…”

Quantifying the influence of space on social group structure

“…We constructed a series of networks based on the sharing of nest boxes, each consisting of 14 days of antennae data over the duration of 2 years (population size during this time period ranged between 52 to 188 tagged adult house mice). Inactivity periods of the data collection system extended this time window, so that each time window consists of a similar period of active data collection (Liechti, Qian, König, & Bonhoeffer, 2020). We used total time spent sharing a nest box in seconds as our measure of association strength.…”

“…In order to be able to partition a network spatially, we first inferred a spatial location for each individual in a network. This was done using a bilayer network approach (Liechti et al, 2020), with the first layer consisting of nest boxes and the second layer consisting of individuals. The between-layer links wn,k were given by the total time individual k spent in the nest box n. We then defined the spatial location of individual k, as the interpolations of spatial locations of all nest boxes linked to this individual:…”

Family dynamics reveal that female house mice preferentially breed in their maternal community

“…Female house mice typically live in social groups with overlapping generations, demic structure and female philopatry (Baker, 1981; Gerlach, 1996; König and Lindholm, 2012; Lewontin and Dunn, 1960). In our study population, groups are relatively stable over time and have proven to be resilient even after a major disturbance (Evans et al, 2020a; Liechti et al, 2020). While there is strong intrasexual female reproductive competition, several females usually breed simultaneously in the same group and even engage in communal offspring care (Ferrari et al, 2019).…”