Dyadic social behaviour in the hamster: Evidence for contextual mediation

Dyadic social interactions among 34 adult male golden hamsters reared in either standard wire cages (Experiment 1) or cages with sandy substrates (Experiment 2) were observed for 15 min on consecutive days in two neutral arenas with either a bare Plexiglas or a sandy substrate. The tests showed that in the arena with the sandy substrate the hamsters spent less time engaged in agonistic encounters, initiated fewer agonistic encounters of all kinds and engaged in fewer agonistic encounters of a potentially physically damaging nature (i.e. violent encounters) than they did in the arena with the Plexiglas substrate. These effects were the same in animals reared under either condition. This finding argues against the behavioural effects being due to the novelty of a sand substrate to laboratory-reared hamsters. Consequently, the social behaviour of hamsters in laboratory settings is highly labile across various contexts. Therefore, it is necessary to examine the behavioural effects of contextual variables if general behavioural principles are to be derived from laboratory indices of social behaviour in hamsters, and possibly a number of other species.