Matthew J.S. Creasey’s research while affiliated with University of Exeter and other places

The factors selecting for territoriality and their relative importance are poorly resolved. Theoretical models predict that territoriality will be selected when resources of intermediate abundance are distributed variably and predictably in time and space, but can be selected against if the resource-holding potential of individuals is low or the risk of predation is high. Here we used a model averaging approach in a mixed modelling framework to analyse 5 years of observational and experimental data collected on group responses to actual and perceived intruders in the cooperatively breeding chestnut-crowned babbler, Pomatostomus ruficeps, in order to provide a rare test of the relative importance of resource value, resource-holding potential and predation risk in territorial behaviour. We found that babblers were highly plastic in their responses to actual and simulated intruders: on average, approaches occurred on 55% of occasions, and aggression ensued in 55% of approaches (observational and experimental results combined). Whether or not babbler groups approached, and if so were aggressive towards, actual or simulated intrusions was explained by time of day, location, group sizes, predator encounter rate and habitat characteristics, but not by reproductive status. Consideration of each of these effects regarding the three hypotheses above suggested comparable roles of group competitive advantage and predation risk on approach probability, whereas ensuing aggression was mostly explained by correlates of resource value. Our study provides compelling evidence to suggest that the risk of predation can affect the incidence of territorial and agonistic behaviour between social groups of animals by moderating the effects of resource value and group competitiveness, and might partly explain the high plasticity in group responses to intrusions.