Schedule-induced polydipsia was studied using a behavioral contrast paradigm. Food pellets were delivered to food-deprived rats on a response-independent FT l-min sehedule. Licking on a tube produeed water on a MULT FR 10 FR 10, MULT FR 10 EXT, or MIXED FR 10 EXT for three rats (Experiment 1) and on a MULT VI VI, MULT VI EXT, or MIXED VI EXT sehedule for three other rats (Experiment 2). On the FR sehedules, rats eould drink more water by inereasing lick rates, but on the VI schedules the amount of drinking was fixed by the sehedule parameters and was relatively unaffected by liek rates. Relative to MULT FR FR, positive polydipsia eontrast was clearly demonstrated on MULT and MIXED FR EXT; but relative to MULT VI VI, eontrast was not demonstrated on MULT and MIXED VI EXT. These data suggest that polydipsia eontrast oceurs only if inereased lieking permits inereased drinking.Several studies (Allen, Porter, & Arazie, 1975;Jacquet, 1972; Porter, Arazie, Holbrook, Cheek, & Allen, 1975) have reported contrast-like effects with schedule-induced polydipsia (Falk, 1969(Falk, , 1971. However, none employed the standard multiple schedule paradigm used to investigate behavioral contrast (see Reynolds, 1%1). Recently, demonstrated positive polydipsia contrast using a procedure analogous to the standard behavioral contrast design. Food pellets were delivered on a fixed-interval (FI) I-min schedule. Drinking was initially stabilized on a MUL T CRF CRF schedule (schedule components las ted 1 min), and when access to water was prevented in one component (MULT CRF EXT), drin king increased substantially in the unchanged component. When access to water was reinstated in the EXT component, drinking in the unchanged component decreased to previous levels.The study is subject to two criticisms. First, aleverpress response was required to deliver .1 ml of water in a water dipper, rather than the more direct response of licking a drinking tube. This method apparently did not constrain the development of schedule-induced polydipsia, since milliliter/pellet intake values were within previously reported ranges. Second, the water-reinforcement schedule (CRF) permitted the amount of water received by the rats to increase during the unchanged component. It might be argued that the increased water intake, rather than the contrast condition, accounted for the increased leverpress rates in the unchanged component. This was unlikely, since leverpress rates (and drinking magnitudes) retumed to baseline levels when the MUL T CRF CRF schedule was reinstated.The present study was designed to avoid these criticisms. In Experiment 1, licking was substituted for leverpressing as the response that produced water, and licking was reinforced by water delivery on a fixed-ratio (FR) 10 schedule. This low schedule value allowed the rats relatively unlimited access to water, and increases in water consumption could occur in the unchanged eomponent of the MULT schedule, as they had in the CRF component of the study. In Experiment 2, licking was reinforced...