Association for Behavior Analysis International

Function-based treatments for problem behavior are well studied using single-subject designs, but there is less research using larger samples. While single-subject designs are an important cornerstone of behavior analysis, larger sample sizes allow us to answer different questions and disseminate results to novel audiences. Two of the studies in this symposium analyze resurgence and relapse in clinical populations. These concepts are well studied in basic research but the prevalence in applied settings is unclear. These studies used consecutive-case review designs to answer this question and identify the prevalence in patients undergoing treatment for problem behavior. The third study also used a consecutive case series design to analyze treatment results for inappropriate sexual behavior, which is an incredibly problematic, but understudied, topography. By using a large-N design, the researchers were able to assess treatment efficacy in a format that easily disseminates to professionals outside of our field. The last talk focuses on a literature review for automatically-maintained problem behavior. In using this large sample found in past research, they identified trends in assessment that may predict the types of treatment components necessary. The implications of use of large-N designs in these projects allowed for solving novel and important research questions.

Professional clinicians and researchers in the field of psychology and behavior analysis who work with children who engage in problem behavior. I also believe this presentation qualified for psychology CEUS (PSY), but this was not an option listed above.

A consecutive controlled case-series design (Hagopian, Rooker, Jessel, & DeLeon, 2013) was used to examine resurgence of problem behavior during implementation of a multiple schedule of reinforcement and during subsequent thinning steps. Multiple schedules signal the availability of reinforcement by correlating a unique stimulus with the presence or absence of reinforcement following a targeted response (Saini, Miller, Fisher, 2014). Resurgence was defined as the occurrence of problem behavior at a rate that exceeded levels observed during baseline or the prior multiple-schedule thinning step in at least one of the next 10 sessions (Volkert, Lerman, Call, & Trosclair-Lasserre, 2009). Data were collected from clients of an intensive day treatment program over an eight year period whose treatment for problem behavior included a multiple schedule component. Resurgence was coded after the multiple schedule was introduced and following each subsequent thinning step. Data were collected on a variety of different participant characteristics including: age, topography of problem behavior(s) (e.g., aggression, self-injury, pica), and confirmed function(s) of problem behavior. Implications for thinning the schedule of reinforcement in a multiple schedule will be discussed.

This study used a consecutive controlled case-series design (Hagopian, Rooker, Jessel, & DeLeon, 2013) to examine the prevalence of renewal when treatments for problem behavior were generalized. Renewal is defined as the reemergence of a previously extinguished response when the stimulus context changes (Kelley, Liddon, Ribeiro, & Greif, 2015). For example, when renewal is applied to the treatment of problem behavior, problem behavior is initially observed in the home. Next, treatment is implemented in a different setting and problem behavior decreases or extinguishes. Finally, when treatment is generalized back to the home, problem behavior may reemerge. This treatment relapse effect has significant implications for clinicians generalizing treatments for problem behavior. Data were analyzed from all clients over a six year period who were admitted to a day treatment program for the treatment of problem behavior. Renewal was coded if problem behavior was observed at a rate in the first 10 sessions in the new stimulus context higher than that of the 10 sessions preceding generalization. Data were also collected on participant and treatment characteristics including: topography(s) of problem behavior, identified function(s) of problem behavior and treatment components. Implications for the treatment of problem behavior and generalization are discussed.