Plasticity and consistency of behavioural responses to predation risk in laboratory environments

Abstract

The individual animal is currently a major focus of behavioural research and an increasing number of studies raise the question of how between-individual behavioural consistency and behavioural plasticity interact. Applying the reaction norm concept on groups, our study addresses both of these aspects in one framework and within an animal's natural social environment. Risk-taking behaviour in 1-year-old perchPerca fluviatiliswas assayed in aquarium experiments before and after the fish were subjected to the presence or absence of a piscivorous predator for 3weeks. To analyse the inter-individual behavioural variation across the repeated measurements, we dissected the behavioural change across the predator treatment into individual constant and plastic components using hierarchical mixed-effects models. During the predator treatment, juvenile perch increased in boldness and decreased in vigilance, the magnitude of these behavioural changes was influenced by group composition. However, the behavioural changes were not influenced by the presence of a predator, indicating the difficulties in generating realistic long-term predation pressure in the laboratory. Individuals differed in the relative increase in boldness across the predator treatment and, thus, varied in the shape of their reaction norms. In accordance, the best linear unbiased predictors, extracted from the random effects of separate linear mixed-effects models for the data before and after the predator treatment were only weakly correlated. Hence, between-individual variation seems to change under laboratory conditions and therewith not necessarily represents the initially present ‘natural’ variation, giving important implications for the conduction and interpretation of behavioural experiments.