Decreasing the intensity of reinforcement-based interventions for reducing behavior: Conceptual issues and a proposed model for clinical practice.

Abstract

Behavioral interventions that include reinforcement as a treatment component have proven quite effective in decreasing problem behavior in children and individuals with developmental disabilities. These interventions are typically initiated with frequent, immediate reinforcement to increase the likelihood of success and schedules may then be thinned to more clinically manageable schedules to promote generalization and maintenance of treatment effects. Immediate reinforcement can also be delayed to the same effect. However, there are currently no specific procedural guidelines for decreasing the intensity of effective behavioral interventions. The current paper examines several conceptual issues regarding procedures for decreasing the intensity of behavioral interventions and presents clinical and research suggestions. ********** Behavioral interventions have proven quite effective in decreasing problem behaviors of children, especially those with developmental disabilities (Carr, Coriaty, & Dozier, 2000; Carr, Yarbrough, & Langdon 1997; Watson & Gresham, 1998). Many behavioral interventions used to reduce behavior incorporate reinforcement as a treatment component in the form of differential reinforcement of appropriate behavior (DRA), differential reinforcement of other behavior (DRO) or noncontingent reinforcement (NCR) (Carr, Coriaty, Wilder et al., 2000; LeBlanc, Le, & Carpenter, 2000). In an attempt to increase the likelihood of success, such interventions typically involve frequent and immediate delivery of high-quality reinforcers when treatment is initiated (Hagopian, Fisher, & Legacy, 1994; Miltenberger, 1997). Once the intervention has proven effective, the intensity of the intervention is gradually decreased over time by either reducing the frequency of reinforcement or implementing a delay to reinforcement. Interventions of decreased intensity are beneficial for several reasons. First, frequent, and immediate delivery of reinforcers often is unmanageable in natural settings (Fisher et al, 2000). For example, a parent may not be able to provide continuous attention for their child during each visit to the doctor or during an important telephone conversation. Second, less labor intensive interventions may result in greater overall treatment integrity. Interventions that are labor intensive or attract undue attention in natural settings are less likely to be implemented consistently than interventions that appear more natural and are easier to implement. Third, less intensive interventions characteristically are more similar to naturally occurring contingencies than highly intensive interventions, promoting generalization of treatment effects to everyday settings (Stokes & Baer, 1977). Unfortunately, several problems may arise when attempting to decrease the intensity of behavioral interventions. First, treatment gains may be lost as interventions become less intensive. For example, Hagopian, Fisher, Sullivan, Acquisto, and LeBlanc (1998) found that effective functional communication training with extinction interventions failed to remain effective in 60% of cases when schedule thinning or delay to reinforcement was initiated. Second, there are no accepted algorithms or even general strategies that identify specifically how the intensity of interventions should be reduced. The intensity of interventions can be manipulated along several dimensions, including the frequency of reinforcement, the delay of reinforcement, and the characteristics (e.g., magnitude, quality) of the reinforcing events. Researchers have failed to compare the effects of altering these dimensions. Most studies that have systematically examined strategies for reducing the intensity of reinforcement-based interventions for reducing behavior have manipulated the frequency of reinforcement (Lalli, Casey & Kates, 1997; Piazza, Moes, & Fisher, 1996). Even in this area, little evidence exists to guide how quickly we should decrease intensity, or even what the initial intensity of the intervention (i. …