Author ORCID Identifier

https://orcid.org/0000-0002-5144-2349

Semester

Spring

Date of Graduation

2023

Document Type

Dissertation

Degree Type

PhD

College

Eberly College of Arts and Sciences

Department

Psychology

Committee Chair

Kathryn Kestner

Committee Member

Barry Edelstein

Committee Member

Kimberly Floyd

Committee Member

Kennon Lattal

Committee Member

Claire St. Peter

Abstract

Differential reinforcement of other behavior (DRO) is a reinforcement schedule that commonly includes the delivery of a reinforcer following an interval during which a target behavior did not occur and extinction (i.e., the reinforcer is withheld following any instances of the target behavior). Although interventions using DRO schedules can decrease target behavior when implemented as designed, little is known about the efficacy of DRO interventions when they include fidelity errors. A growing field of literature has demonstrated different ways fidelity errors can affect the outcomes of other interventions (e.g., DRA, response cost, and skill acquisition). One study by Foreman et al. (2023) evaluated how commission errors may change the efficacy of DRO interventions in a human-operant arrangement with college students. For three participants, DRO lost its efficacy at 60% fidelity or lower. However, DRO remained effective in maintaining low rates of target responding at as low as 20% fidelity for the remaining three participants. These results suggest that DRO may have robust effects with commission errors for some but may vary across individuals or other factors. The study of the effects of omission errors during DRO is largely missing from current research. Additionally, the effects of fidelity errors during DRO have not yet been studied in a design arranged to identify potential order effects. Experiment 1 examined effects of DRO implemented with various percentages (100%, 80%, 60%) of fidelity with commission errors in a human-operant arrangement. Experiment 2 systematically replicated Experiment 1 with omission errors. During both experiments, points were delivered contingent on the target response (mouse clicks on a moving circle) during baseline phases, and points were delivered contingent on the absence of the target response during phases with DRO implemented with 100% fidelity. In Experiment 1, DRO was implemented with 60% and 80% fidelity with commission errors, during which DRO was implemented with points delivered on a probabilistic basis contingent on the target response. In Experiment 2, DRO was implemented with 60% and 80% fidelity with omission errors, during which point deliveries according to the DRO schedule were omitted on a probabilistic basis. The results from Experiment 1 suggest that commission errors are detrimental to DRO schedules for some, resulting in increased response rates that often exceeded the levels of responding observed during baseline. Additionally, participants exposed to ascending percentages of fidelity first engaged in more frequent increases in responding during degraded-fidelity phases than participants exposed to the descending percentages of fidelity first. The results from Experiment 2 suggest that omission errors degrade the effects of DRO at 60% fidelity for some, such that response rates increase compared to DRO with 100% fidelity; however, even when responding was elevated during these phases, response rates remained lower than those during baseline phases. The results of the current experiments provide preliminary support for recommendations regarding DRO implementation, training, and fidelity reporting.

Download

Share

COinS