The landscape of fear has individual layers: an experimental test of among-individual differences in perceived predation risk during foraging

Perceived predation risk varies in space and time creating a landscape of fear. This key feature of an animal’s environment is classically studied as a species-specific property. However, individuals differ in how they solve the trade-off between safety and reward and may, hence, differ consistently and predictively in perceived predation risk across landscapes. To test this hypothesis, we quantified among-individual differences in boldness and activity and exposed behaviourally phenotyped male bank voles (Myodes glareolus) individually to two different experimental landscapes of risks in large outdoor enclosures and provided resources as discrete food patches. We manipulated perceived predation risk via vegetation height between 2 and >30 cm and quantified patch use indirectly via RFID-logging and giving-up densities. We statistically disentangled among-individual differences in microhabitat use from spatially varying perceived risk, i.e. landscape of fear.  We found that individuals varied in mean vegetation height of their foraging microhabitats and that this microhabitat selection matched the intrinsic individual differences in perceived risk. As predicted by the patch use model, all individual’s perceived higher risks when foraging in lower vegetation. However, individuals differed in their reaction norm slopes of perceived risk to vegetation height, and these differences in slopes were consistent across two different landscapes of risks and resources. We interpret these results as evidence for individual landscapes of fear, which could be predicted by among-individual differences in activity and boldness. Since perceived predation risk affects when and where to forage, among-individual differences in fear responses could act as a mode of intraspecific niche complementarity (i.e. individual niche specialization), help explain behavioural type by environment correlations, and will likely have cascading indirect effects on lower trophic levels.

Methods
Please see accompanying paper.