RESUMO
Different test series have been developed and used to measure behaviour in shelter dogs in order to reveal individuals not suitable for re-homing due to their aggressive tendencies. However, behavioural tests previously validated on pet dogs seem to have relatively low predictability in the case of shelter dogs. Here, we investigate the potential effects of (1) timing of the behaviour testing and (2) presence of a human companion on dogs' aggressive behaviour. In Study I, shelter dogs (n=25) showed more aggression when tested in a short test series two weeks after they had been placed in the shelter compared to their responses in the same test performed 1-2 days after arrival. In Study II, the occurrence of aggressive behaviour was more probable in pet dogs (n=50) in the presence than in the absence of their passive owner. We conclude that the sensitivity of aggression tests for shelter dogs can be increased by running the test in the presence of a caretaker, and after some period of acclimatisation to the new environment. This methodology could also provide better chances for successful adoption.