Attack priming and aggressive arousal in female Syrian golden hamsters, Mesocricetus auratus

Earlier studies have found that priming a resident female hamster by allowing it to attack a conspecific intruder transiently reduced the latency of its attack on a second (probe) intruder. The present series of experiments showed that the time the subject spent in contact with each intruder prior to attack revealed this priming effect more clearly than did the conventional total elapsed time measure. The influence that stimuli encountered during the first few minutes of intruder exploration would heighten the subject’s aggressive arousal was confirmed in experiments showing that increasing exposure to an anaesthetized intruder from 0 to 90 s systematically reduced subsequent attack latency. Ninety seconds of contact with an anaesthetized intruder just prior to testing on a pair of priming and probe trials significantly reduced the priming effect. However, such exposure may not reproduce the full reduction in latency that follows an overt attack. Consecutive priming and probe attack latencies were uncorrelated even though the latter is routinely shorter than the former. Attacks were therefore modelled as stochastic events. Standard log survivor analysis suggests that attack probabilities increase to an asymptote during both priming and probe trials. A novel θ(t) transformation of the data showed more clearly that the priming effect results from a probability of attack which starts at a higher level on probe trials and rise to asymptote faster.