Structural consistency of behavioural syndromes: does predator training lead to multi-contextual behavioural change?

Behavioural syndromes are suites of behaviours that are correlated across multiple contexts. Syndromes may occur in populations because behaviours are tightly linked by underlying mechanisms, such as genetics or physiology, which constrain flexibility and preclude multi-contextual plasticity. Alternatively, correlated behaviours may not share a common mechanism and may be able to change independently, allowing for potentially maladaptive combinations of traits to be broken apart. We tested these two hypotheses by training Trinidadian guppies (Poecilia reticulata) which possessed a behavioural syndrome encompassing three contexts, to avoid a potential predator. While we found no difference in magnitude of behavioural change between the trained and control groups, we did find that all subjects generally became shyer toward a potential predator following training and hypothesise that this resulted from sensitization and a predisposition to quickly recognize and adjust behaviour to predator-like stimuli. Importantly, behavioural changes in response to a potential predator did not generate changes in 'general activity' or 'exploration', and a tri-contextual syndrome broke apart. Our results suggest that in this population of guppies, individuals differ in behavioural plasticity in terms of their response to experience with predation risk and behaviours across contexts are domain specific and are able to change independently of each other. Future research should focus on populations that evolved in high predation environments and, therefore, may possess more rigid syndromes, to determine whether behavioural flexibility is limited.