Australia has the worst record of mammal extinction of any continent. In order to prevent further extinctions, we rely on predator-free havens; islands and fenced areas within which vulnerable fauna can persist. While this strategy has successfully secured many species, the predator-free conditions inside havens can cause rapid evolutionary loss of antipredator behaviours. Once these behaviours are lost, havened mammals become easy prey for even their natural predators, and so my be completely unfit for release into unfenced habitats.

Rapid loss of predator awareness may be jeopardising national conservation strategies for Australian mammals. Unfortunately, we are currently blind to this trait loss because, by and large, we have never looked for it. This lack of information stems from the facts that a) we have only recently become aware of this risk; and b) accurately measuring behavioural change in wild populations is a real challenge. We must urgently quantify the extent to which antipredator traits have been lost, but to do so requires the development of ground-truthed behavioural assays that can be applied to a range of threatened mammal species.

This project will develop robust behavioural assays to measure anti-predator responses in critical-weight-range mammals. We will use a bettong - the woylie (Bettongia penicillata) and a bandicoot - the quenda (Isoodon fusciventer) as model systems. We will develop a library of well-characterised assays, and use these to quantify behavioural shifts between populations of these species inside and outside of predator-free havens. Our work will not only provide direct measures of trait change in these two species, but will provide a battery of assays that can be applied to many other critical-weight-range mammal species under intensive conservation management.


This project is well underway, led by the highly capable duo of Nicki Mitchell and Natasha Harrison. The project has already resulted in several publications, not only proofing up behavioural assays, but applying them to investigate trait shift in havened populations of Woylies.

Some fairly simple behavioural assays exhibit reasonable repeatability and exhibit clear shifts between havened and non-havened populations. The work has attracted considerable media attention. The finding that havened animals may be losing critical traits is of clear public interest and will likely lead to systems not only for monitoring and managing populations, but also the traits these populations carry. Hopefully the result will be a future-proof conservation strategy!

Selected publications

Harrison, Natasha. D., Phillips, Ben. L., Mitchell, Nicola. J., Wayne, Julia. C., Maxwell, Marika. A., Ward, Colin. G., Wayne, Adrian. F. (2023). Perverse outcomes from fencing fauna: Loss of antipredator traits in a havened mammal population Biological Conservation. 281: 110000.

Harrison, Natasha D., Rochelle Steven, Ben L. Phillips, Jan M. Hemmi, Adrian F. Wayne, and Nicola J. Mitchell. (2023) Identifying the Most Effective Behavioural Assays and Predator Cues for Quantifying Anti-Predator Responses in Mammals: A Systematic Review. Environmental Evidence 12: 5.

Harrison, Natasha. D., Phillips, Ben. L., Hemmi, Jan. M., Wayne, Adrian. F., Steven, Rochelle, Mitchell, Nicola. J. (2021). Identifying the most effective behavioural assays and predator cues for quantifying anti-predator responses in mammals: a systematic review protocol Environmental Evidence. 10: 38.