Chlordiazepoxide (5 and 10 mg/kg) and diazepam (2.5 mg/kg) reduced the latency to eat and enhanced feeding response to familiar food in a food-preference test. The increased feeding response resulted from an increased frequency of individual eating episodes (bouts) without significant change in the episode duration. Experiment II indicated that this facilitatory action of the two drugs was not dependent on the presence of principal novelty cues in the food-preference test. Diazepam (5 mg/kg), in contrast, enhanced the feeding responses to novel food in the food preference test. Experiment III demonstrated that although diazepam can counteract induced food neophobia, its facilitory action on the response to novel foods may be mediated by a mechanism which can operate independently of food neophobia.