Habituation or sensitization? Short-term adjustment of flight initiation distance in incubating common eiders

Optimal escape theory predicts that prey balance the costs and benefits of fleeing. However, the use of prior experience in prey escape decisions has been overlooked. Indeed, when repeatedly exposed to a nonlethal yet threatening stimulus, prey may adjust their escape response by habituating, delaying escape when repeatedly exposed to that stimulus, or by becoming sensitized, escaping sooner. Here, we investigated intrinsic and extrinsic drivers of short-term variation in flight initiation distance (FID, distance at initiating escape) in response to repeated human approaches in incubating female common eiders, Somateria mollissima. We predicted habituation to repeated exposure to threatening but nonlethal stimuli, reflected in decreased FID. We also expected increased habituation in females with a high current reproductive value (older breeders, females in good condition or laying large clutches) or brood value (early breeders). Last, we investigated whether changes in FID varied with threat level (i.e. different adult or nest predation risk or breeding habitat). The magnitude of change in FID was small and varied among individuals, depending on individual traits (body condition, clutch size and phenology) and perceived threat (island-specific adult predation risk). Accordingly, females in poorer body condition, that laid larger clutches or laid earlier in the season were more prone to become sensitized over repeated approaches. Consistent with the risk allocation hypothesis predicting a dampening of antipredator responses under chronic risk exposure, females were less prone to become sensitized when nesting on islands with high adult predation risk. These results underline the ability of animals to make plastic state- and context-dependent adjustments of antipredator behaviour when repeatedly confronted with nonlethal stimuli. Nevertheless, given the limited magnitude of within-individual, short-term adjustment of FID compared to the among-individual variation in FID, questions remain concerning the efficacy of these antipredator responses in a context of rapidly changing predation regimes.