Odor-mediated runway performance following extended training and grdual changes in reinforcement magnitude.

Abstract

Many researchers in the area of odor-mediated runway performance have suggested that frustration (as spoken of by Amsel, 1958, 1962) may be the mechanism whereby nonreward odors are produced. The present research was designed to test this idea. Three groups of rats received a daily 8-trial double-alternation (DA) schedule of reward and nonreward in a straight runway during a three-phase experiment. Only subjects receiving large (12 pellet) rewards developed appropriate DA patterning during Phase 1 (12 days). During Phase 2 (33 days) one group continued to receive large reward training, while a second group underwent gradual reward reduction, and a third group experienced gradual reward increase. Appropriate patterning was maintained throughout the entire phase by the group continued on large reward, and until the final block of trials (small-reward level) by the gradualdecrease animals. Patterning was established by the gradual-increase animals. During Phase 3 (6 days) all groups received 1 pellet on reward trials. Although R trials speeds for all groups on all days were faster than N trial speeds, they were not significantly faster on all days. The results indicate that frustration is not the underlying mechanism of odor production. Frustration has been shown to decrease after 240 trials. In the present study Group E continued to maintain patterning as long as 360 trials. When reward size was decreased abruptly (a condition which should result in frustration) Group E continued to respond appropriately in the run section. Thus, reward odors were present during frustrative conditions. Because Group I was trained on a 1 pellet reward schedule they should not have experienced frustration. However, they developed patterning during Phase 2. Two predictions could be made based on frustration for Group DIS behavior. However, neither prediction can account for the behavior of subjects in Group D.