The purpose of the present study was to test two predictions derived from the application of Pavlovian conditioning principles to the schedule-induced polydipsia (SIP) phenomena. The first prediction was that a pretrained excitatory signal for pellet delivery would facilitate the development of SIP. A second prediction was that phenomena which are typically observed in Pavlovian conditioning experiments, such as learned irrelevance, latent inhibition, and the US preexposure effect, would also be observed in the SIP paradigm. The results indicated that a group given excitatory pretraining to a tone failed to develop SIP more rapidly than controls when this tone was used to signal pellet deliveries during SIP training. However, learned irrelevance, latent inhibition, and the US preexposure effect were demonstrated. Thus, these data generally support the predictions derived from the Pavlovian conditioning view of SIP.