Validation of a simple, ethologically relevant paradigm for assessing anxiety in mice.

Although numerous behavioral tests are available to assess anxiety, we introduce a simplified version of a previously established test that is exquisitely sensitive and reliable. The latency to consume a palatable snack (graham wafer crumbs) was assessed among mice in their home cage and in an unfamiliar environment, as well as in the presence or absence of predator scent. The effects of various anxiolytics and nonanxiolytics were evaluated in these paradigms. When offered the palatable snack in a familiar environment, mice readily approached and began consumption; however, in a novel environment (cage with fresh bedding), or in the presence of predator scent (rat feces), response latencies increased 10-fold. Anxiolytics, including diazepam, chlordiazepoxide, propranolol, or chronic treatment with either buspirone or the antidepressant desmethylimipramine attenuated the effects in the novel environment without affecting home-cage responding. In contrast, nonanxiolytic agents (haloperidol, amphetamine, acute buspirone or desmethylimipramine) failed to exclusively affect novel environment-induced prolongation of response latencies. The simplicity of design, the absence of food deprivation or neurogenic stressors, the possibility of using it in a repeated measures design, the reliability and magnitude of response, and the specificity and sensitivity to anxiolytic drugs makes this an ideal preparation with which to assess anxiety and anxiety-altering manipulations.