Context. Predator-controlled environments can lead to prey species losing costly antipredator behaviours as they exploit their low-risk environment, creating a ‘predator-naïve’ population. If individuals lacking suitable antipredator behaviours are used as source populations for reintroductions to environments where predators are present, their behaviour could result in high post-release predation. In contrast, animals sourced from environments with predators (‘predator-exposed’) may show effective antipredator behaviours and thus higher survival post-release.

Aims. The aim was to compare the antipredator behaviour of brushtail possums (Trichosurus vulpecula) at predator-exposed and predator-naïve source populations, and then compare post-release survival after their reintroduction to a low predator environment.

Methods. Data were collected from possums at two sites, one with and one without mammalian predators. The behavioural responses of possums to a spotlighter, their willingness to use supplementary feeders at ‘safe’ and ‘risky’ heights, whether they avoided predator odour at traps and their general willingness to enter traps were recorded.

Key results. Predator-naïve possums showed weaker antipredator responses, were often found at ground level, engaged with novel objects, did not avoid predator scents and utilised different habitats regardless of associated predation risk. In contrast, predator-exposed possums had higher antipredator responses, chose connected trees, were rarely found at ground level and were generally difficult to capture. Post-translocation survival was high for both source populations. Predator-naïve-sourced female possums began to avoid predator urine (feral cat; Felis catus) 12 months after translocation.

Conclusions. Our research demonstrates that environmental predation risk can predict prey naïvety in brushtail possums. Some aspects of prey naïvety behaviour appear to be able to change in response to altered predation risk.

Implications. With many threatened species now existing only in feral predator-free areas, these results have implications for future reintroductions into unbounded areas where feral predators are present, and for the management of fenced reserves. The addition of a small number of predators to fenced reserves may aid in retaining antipredator behaviours in fenced prey populations.