Disorders of fear and anxiety are significant behavioral and physiological concerns in veterinary medicine. Thus, the present studies sought to develop and validate novel preclinical laboratory models for the development of anxiolytics for the veterinary market. Specifically, noise-induced sensitivity and aversion using thunderstorm recordings on an open-field task in Beagle dogs was used to establish this model. This thunderstorm task was based on the open-field tests previously described for the dog, except that a thunderstorm recording, compiled from desensitization compact discs, played during the middle of the test was used to elicit anxious or fearful responses in laboratory dogs. Initially, we compared the behavioral response on this test with that on an open-field test, in which no sound stimulus was provided. When compared with the open field, the thunderstorm recording increased inactivity duration and frequency, which was related to freezing behaviors and near-door duration. This suggests that we can objectively characterize a fear and anxious response to noise. We then attempted to pharmacologically validate this model by testing 0.5, 1, or 2 mg/kg doses of diazepam on this behavioral outcome. In test-naïve subjects, diazepam reduced the increase in inactivity seen at washout compared with that seen at baseline, suggesting that this paradigm may be useful for establishing the effect of drugs that reduce or prevent sensitization to fear-invoking events. In test-experienced subjects, diazepam reduced inactivity compared with both baseline and washout, indicating that this model may be useful for screening drugs counteracting fear and phobia, as well as anticipatory anxiety responses. Interestingly, diazepam stimulated behavioral activity as measured with the Actiwatch method. The test described here provides a laboratory method for testing therapeutics targeted at reducing either anxiety or fear related to noise sensitivity and aversion in pet dogs.