Incentive contrast following repeated shifts in magnitude of food reward in the Skinner box

Abstract

Thirty-two rats were trained in the Skinner box and received either large or small food reward in Phase 1. In Phase 2, half of each group was shifted to the other reward condition while the other half continued to receive the same reward magnitude. In Phase 3, Phase 1 conditions were reinstated for the shifted groups. Neither a positive contrast effect (PCE) nor a negative contrast effect (NCE) was obtained in Phase 2. Following a repeated shift in Phase 3, a strong NCE was obtained, but only a graphical, albeit statistically nonsignificant, PCE was observed. The results were discussed in terms of the older relative theories of Crespi and Helson as well as the more recent relativistic accounts of Amsel and Capaldi. It has consistently been found in runway studies that subjects shifted from a large to a smaller food reward run Significantly slower than control subjects receiving small reward all the time, thus reflecting a negative contrast effect (NCE). The evidence for the opposite phenomenon, namely for a positive contrast effect (PCE) in the runway is virtually nonexistent (Dunham, 1968). However, with appropriate controls for the assumed ceiling effects inherent in the running response, numerous studies have recently reported Significant PCE based on shifts in magnitude of food reward (cf. Mellgren, 1972; Shanab, Birnbaum, & Cavallaro, 1974; Shanab, Sanders, & Premack, 1969). In the studies just cited, positive contrast was studied as a function of a single shift in reward magnitude. In other words, sub­ jects were shifted from a small to a larger reward only once. In a repeated-shift deSign, Capaldi and Lynch (1967) reported no PCE regardless of whether their subjects were shifted once or twice from a small to a larger reward magnitude. These results are inconsistent with either the earlier fmdings of PCE (Crespi, 1942) or the more recent findings of PCE based on repeated shifts (Benefield, Oscos, & Ehrenfreund, 1974; Shanab & Ferrell, 1975; Shanab, Young, & France, 1975). More recently, Shanab, France, and Young (1976), using sucrose reward, reported Significant PCE and NCE following a repeated-shift procedure similar to that used by Capaldi and Lynch (1967). However, no NCE in the runway has been obtained based on single downshifts in either magnitude or concentration of liquid sucrose, or both (Barnes & Tombaugh, 1973; Flaherty, Riley, & Spear, 1973; Goodrich, 1962; Homzie & Ross, 1962; Ison & Rosen, 1968; Rosen, 1966; Rosen & Ison, 1965; Shanab, Young, & France, 1975). But Significant NCE with sucrose was obtained in the licking response (Vogel, Mikulka, & Spear, 1968), and the