Date of Graduation


Document Type


Degree Type



Eberly College of Arts and Sciences



Committee Chair

Kennon A. Lattal.


The effects of reinforcement rates of alternative responding on resurgence were studied in six experiments with pigeons. In Experiment 1A, key pecking was maintained on a multiple variable-interval (VI) VI schedule during the first, Training, phase. In the second, Response-Elimination, phase, variable differential-reinforcement-of-other-behavior (DRO) schedules were in effect in each component and reinforcement rates were equal and then, higher in one (rich) component, and lower in the other (lean), than in the Training phase. In the third, Resurgence, phase, reinforcers were discontinued and more resurgence occurred in the lean than in the rich component. Differences in Training-phase response rates between components also could have produced these results. In subsequent experiments, rich and lean components were those in which higher and lower response rates occurred in the Training phase, respectively. These experiments differed in how reinforcement rates were programmed during the Response-Elimination phase. Experiment 1B was a replication of Experiment 1A with experimentally-naive pigeons. In Experiment 2, reinforcement rates were equal in one component and lower or higher in the other, relative to those programmed in the Training phase. In Experiment 3, reinforcers were discontinued in each component before differential reinforcement rates were in effect. In Experiment 4, differential reinforcement rates were in effect initially and, then, reinforcement rates in each component were equal to those in the Training phase. In Experiment 5, differential reinforcement rates were arranged by a fixed-DRO in each component and, in Experiment 6, alternative responding consisted of pecking a different key under VI schedules arranging higher and lower reinforcement rates than in the Training phase. Little to no resurgence occurred in each experiment, and differential resurgence was not systematically related to reinforcement rates of alternative responding. Schedule differences, topography of alternative responding, order of exposure to conditions or the length of Training and Response-Elimination phases could not explain these results, nor did current theoretical models of resurgence predict them. Reinforcement rates of alternative responding did not affect resurgence according to the schedule parameters in each experiment. Studies in which reinforcement rates of alternative responding were manipulated parametrically would clarify the present results and the effects of this variable on resurgence.