Two experiments examined the counterconditioning of an aversively motivated response in rats. Presentation of a highly palatable sugar solution (maltose in Experiment 1; sucrose in Experiment 2) to thirsty rats was used as the counterconditioning treatment. In Experiment 1, the counterconditioning procedure was found to be effective in modifying both a newly acquired and a cue-reactivated fear memory. In Experiment 2, the counterconditioning effect was shown to occur when the fear memory was reactivated with a noncontingent exposure to the unconditioned stimulus rather than the conditioned stimulus. This outcome supports the interpretation of counterconditioning as a modification of some central representation of the original training memory, rather than the acquisition of a competing peripheral response. The methodological implications of the present approach to counterconditioning are considered.