THE EFFECTS OF PROGRESSIVELY THINNING HIGH-PREFERENCE STIMULUS DELIVERY ON RESPONDING: AN EMPIRICAL ANALYSIS AND HYPOTHETICAL APPLICATION

Abstract

Delivering high preference reinforcers in an organization-wide behavior change program is optimal, but may prove difficult. Depending upon the number of employees participating in the program, there may be many high-preference stimuli; these may make it difficult for managers to track and deliver all of the high preference stimuli. The current investigation examined a systematically thinning high preference delivery model using a modified progressive ratio procedure. Mean responding during the first baseline phase was used to determine response requirements for earning stimuli during intervention phases. During each session in the intervention phases each occurrence of a participant completing the mean number of responses found in baseline resulted in a decreasing opportunity to earn $3 worth of a preferred stimulus (and a corresponding increasing chance of earning a low-preference stimulus). By averaging the percentages reached in all intervention sessions a breaking percentage was calculated for each participant. Results indicated that across five participants the mean breaking percentage was 78.24%. The range of percentages reached during individual intervention sessions was 8% to 100%. The number of stimuli required to account for high preferences in the participant group, as well as separate groups of 5, 10, 25, and 40 participants, suggest that the obtained mean breaking percentage would not maintain responding. From the current data set, the random delivery of high-preference stimuli to a group is not recommended.