|
Translational Research on Persistence and Relapse |
Saturday, May 23, 2015 |
3:00 PM–4:50 PM |
006C (CC) |
Area: EAB/PRA; Domain: Applied Research |
Chair: Christopher A. Podlesnik (Florida Institute of Technology) |
Discussant: Iser Guillermo DeLeon (Kennedy Krieger Institute) |
CE Instructor: Iser Guillermo DeLeon, Ph.D. |
Abstract: The success of behavioral interventions often requires both the persistence of desirable behavior and preventing relapse of problem behavior. Goals of translational research in behavior analysis are to identify fundamental behavioral processes at work when implementing behavioral treatments and to use the understanding of those processes to improve treatment implementation. Therefore, translational research forms a continuum, with basic and applied research informing one another better serves the goal of improving the efficacy of behavioral treatments than either alone. This symposium includes presentations spanning the range of translational research assessing factors impacting behavioral persistence and relapse. Two presentations assess basic questions involving differential reinforcement using animal models (Bai, Craig), one assesses potentially converging measures of persistence in individuals with developmental disabilities (Kelley), and another assesses persistence of academic performance (Schieltz). All four presentations have a common focus toward highlighting the relevance of translational research to understanding how reinforcement processes influence outcomes of behavioral treatments. |
Keyword(s): persistence, relapse, translational research |
|
Local patterns of resurgence following repeated contingency reversals between target and alternative responding |
JOHN BAI (University of Auckland), Sarah J. Cowie (University of Auckland), Christopher A. Podlesnik (Florida Institute of Technology) |
Abstract: Animal models of resurgence suggest discontinuation of alternative sources of reinforcement can result in relapse of extinguished behaviors, potentially providing insight into clinical relapse of problem behavior. Procedurally, resurgence is typically studied across three successive phases: (1) reinforcement of a target response, (2) extinction of the target and concurrently reinforced training of an alternative response, and (3) re-emergence of the target response when extinguishing the alternative response. The present study used pigeons in 50-s discrete trials to explore the effects of gradual versus discrete stimulus changes on local patterns of resurgence. Within each trial, target (left-key) responding was reinforced and alternative (right-key) responding was not reinforced in the first half of the trial the contingencies then reversed in the second half of the trial. Consistent with baseline reinforcement contingencies, alternative responding replaced target responding across time since trial onset. Moreover, extending trials beyond training durations during extinction probes revealed a resurgence of target responding. The magnitude of resurgence was greater, and the latency was earlier, when signaling the onset of the extended extinction probes. These findings reveal a role for stimulus control in resurgence, which could be useful in understanding factors involved in relapse of problem behaviors. |
|
Effects of Separating Target- and Alternative-Reinforcement Contexts on Relapse Following Differential Reinforcement of Alternative Behavior |
ANDREW R. CRAIG (Utah State University), Mary Margaret Sweeney (Utah State University), Timothy A. Shahan (Utah State University), John A. Nevin (University of New Hampshire) |
Abstract: The behavioral-momentum-based approach to understanding relapse suggests alternative reinforcement may contribute to relapse if it is delivered in the same discriminative-stimulus context as reinforcement for target responding. In this experiment, effects of separating target- and alternative-reinforcement contexts on relapse were examined. Pigeons pecked keys in a two-component multiple schedule. During baseline, food was delivered in both components according to variable-interval 120-s schedules. During treatment, no contingency changes were introduced in one, No-Treatment, component. Food deliveries for target responding were suspended in the other, Treatment, component, and access to an alternative context was delivered according to a differential-reinforcement-of-other-behavior 15-s schedule for target responding. In this context, an alternative-response key was available, pecks to which produced food according to a variable-interval 30-s schedule. During resurgence testing, food was suspended in the No-Treatment component and in the alternative context of the Treatment component. Finally, access to the alternative context was suspended in the Treatment component, and free food was delivered 2 and 8 s into each component presentation to test for target-response reinstatement. During resurgence testing, little-to-no resurgence was observed, and reinstatement was greater in the No-Treatment component than in the Treatment component. Practical and theoretical implications of these data will be discussed. |
|
Convergence of progressive-ratio analyses and momentum as a measure of response strength |
CLARE LIDDON (Florida Institute of Technology), Michael E. Kelley (The Scott Center for Autism Treatment, Florida Institute of Technology), Aurelia Ribeiro (Florida Institute of Technology) |
Abstract: Response strength may be measured in a variety of ways, including in the context of a behavioral economic analysis (e.g., measures of demand; Hursh, 1980) or as response persistence (tendency of a response to continue to occur in the face of a disrupting event such as extinction; Nevin, Mandall, & Atak, 1983). In the current study, we sought to evaluate the convergence of these measures of response strength. First, we established two stimuli that were similar in preference in the context of formal stimulus preference assessments (Fisher et al., 1992). Next, we conducted progressive-ratio assessments to establish breaking points for each stimulus (Roane, Lerman, & Vordran, 2001). Finally, we exposed responding to disruptive events (e.g., extinction, satiation, pre-feeding, or distraction) to test the extent to which responding would persist. Similar outcomes across the analyses would suggest convergence of these two measures of response strength. Results are discussed in the context of translational research. |
|
An evaluation of behavioral persistence with academic performance |
KELLY M. SCHIELTZ (The University Of Iowa), David P. Wacker (The University of Iowa), Brooke Natchev (University of Iowa) |
Abstract: The purpose of this study was to evaluate the effects of positive reinforcement on academic behavior that had historically been related to problem behavior maintained by negative reinforcement. Sam was 12 years old with ASD. During Phases 1 and 2, a choice assessment was conducted within a concurrent schedules design to demonstrate escape responding and low academic performance in math. During Phase 3, contingent positive reinforcement (SR+) was evaluated within an AB design, and showed that SR+ increased Sam’s academic performance. During Phase 4, SR+ was evaluated within a concurrent schedules design embedded within a reversal design. Results showed that SR+ functioned as an abolishing operation for negative reinforcement (i.e., increase in choice allocation towards math, number of problems attempted, and percentage correct). During Phase 5, behavioral persistence was evaluated within a reversal design. Results showed behavioral persistence with academic performance, specifically, the number of problems attempted. As Sam received greater amounts of positive reinforcement (for answering bonus questions) he completed more math problems across sessions during subsequent extinction sessions. This application of behavioral persistence provides a novel application for improving academic behavior. |
|
|