Percentile Reinforcement: Paradigms for Experimental Analysis of Response Shaping11Preparation of this paper was supported by Grant A8269 from the National Research Council of Canada. The author wishes to thank H. M. Jenkins for critical readings of earlier versions of this paper, although remaining inadequacies are solely the fault of the author.

Publisher Conditioning paradigms allow experimental control of selected classes of variables, which potentially affect behavioral change. Instrumental or operant conditioning paradigms allow examination of how the consequences of an organism's behavior change subsequent behavior. This chapter presents a class of conditioning paradigms, which offer considerable promise for experimentally analyzing a class of variables that has long been accepted as fundamental to behavior change, but which remained inextricably confounded within traditional paradigms of instrumental and operant conditioning. Additionally, some early results utilizing these new paradigms are reviewed, and some examples of applications of these paradigms to specific problems in behavior theory are proposed. The paradigms presented in this paper have been developed to deal with problems of controlling contact with, and selectiveness of reinforcement criteria during response shaping, but similar behavior dependent schedules seem particularly well suited to a number of problem areas in which effects of behaviorally confounded relationships between events have to be separated.