Predator naivety negatively affects reintroduction success and this threat is exacerbated when prey encounters predators with which they have had no evolutionary experience. While methods have been developed to inculcate fear into such predator-naïve individuals, none have been uniformly successful. Exposing ontogenetically- and evolutionary-naïve individuals firstly to native predators may be an effective stepping stone to improved responses to evolutionarily novel predators. We focused on greater bilbies (Macrotis lagotis) and capitalised on a multi-year mammalian recovery experiment whereby western quolls (Dasyurus geoffroii) were reintroduced into parts of a large fenced reserve that contained a population of naïve bilbies. We quantified a suite of antipredator behaviours and measures of general wariness across quoll-exposed and quoll-naive bilby populations. We then translocated both quoll-exposed and quoll-naïve individuals into a large enclosure that contained feral cats (Felis catus) and monitored several behaviours. We found that bilbies can respond appropriately to quolls but found only limited support that experience with quolls better prepared bilbies to respond to cats. Both populations of bilbies rapidly modified their behaviour in a similar manner following their reintroduction to a novel environment. These results may have emerged due to insufficient prior exposure to quolls, inappropriate behavioural tests or insufficient predation risk during cat exposure. Alternatively, quolls and cats are only distantly related and may not share sufficient similarities in their predatory cues or behaviour to support such a learning transfer. Testing this stepping stone hypothesis with more closely-related predator species and under higher predation risk would be informative.

Data from each of four experiments are provided. The data was collected from a variety of methods conducted under field settings. These primarily were from radio-tracking data, camera trap data and filming of behaviour. Raw data of burrow sites are provided. Data collected from videos from either camera traps or hand held cameras were analysed using BORIS to determine the time spent engaged in various behaviours. The proportion of time engaged in specific behaviours were calculated from the total time the individual was visible.

The data is provided as a csv file and can be opened using MS excel or R.