Single and multiple schedules of reinforcement in developmentally retarded children.

Free-operant conditioning methods are being used with increasing frequency in a wide range of research applications with human subjects. Although some of the schedules of reinforcement which are of central importance in this kind of research have been systematically evaluated with infrahuman organisms (Ferster & Skinner, 1957), as yet few studies bear directly on questions of generality, variability, and special effects of schedule control in human Ss (Bijou, 1958; Ellis, Barnett, & Pryer, 1960; Green, Sanders, & Equier, 1959; Holland, 1957; Holland, 1958; Long, Hammack, May, & Campbell, 1958). Aside from academic issues involved in assumptions that infrahuman data provide sufficient bases for application in human behavior (Beach, 1960) and the need for genuine comparative and developmental data, practical research problems arise. Typically, extensive "trying out" of schedule types and values is necessary to find appropriate rate, pause distribution, resistance to extinction, and other features of free-operant behavior for studies using human Ss. The results of these time-consuming preliminary investigations are potentially valuable for other researchers because they provide empirical foundation for the selection of schedules with maximum sensitivity and control. The purpose of the present paper is to show characteristic performances under four basic schedules (VI, Fl, VR, and FR) and two multiple schedules (mult VR ext and mult CRF ext) in human Ss, and to illustrate some of the range and variability of these features in a population of institutionalized developmentally retarded children.3 However, we did not attempt to demonstrate the full range of possible schedule values or limiting conditions and thresholds. Examples are from schedules that have proven useful in the study of retarded children.