It has been widely reported that the
anxiolytic efficacy of
benzodiazepines in the elevated plus-maze test is abolished in subjects (rats or mice) that have been given a single prior undrugged experience of the test apparatus. The present series of experiments was designed to further characterise the key experiential determinants of this intriguing phenomenon in Swiss Webster mice. Using a standard 5 min test duration for both trials, Experiment 1 confirmed the
anxiolytic efficacy of
chlordiazepoxide (
CDP; 5-20 mg/kg) in mice naive to the plus-maze, but a virtual elimination of
drug effects in animals that had been pre-exposed to the maze 24 h earlier. Experiments 2 and 3 demonstrated that, while extending the duration of initial exposure to 10 min did not prevent the loss of
CDP (10 mg/kg) efficacy in a standard-duration second trial, increasing the duration of both trials reinstated an
anxiolytic profile for the compound. Finally, although trial 1 confinement to an open arm did not compromise
CDP efficacy when animals were subsequently allowed to freely explore the maze (Experiment 4), closed arm confinement during initial exposure abolished the
drug's
anxiolytic action upon retest (Experiment 5). In contrast to previous findings in rats, these data suggest that the experientially induced loss of
benzodiazepine efficacy in the mouse plus-maze depends rather critically upon prior discovery and exploration of relatively safe areas of the maze (i.e. closed arms). Results are discussed in relation to the hypothesis that the absence of an
anxiolytic response to
benzodiazepines in plus-maze-experienced subjects reflects the acquisition of an open arm
phobia during trial 1.