Association for Behavior Analysis International

The papers in this symposium review the current status of the literature with respect to treatment integrity and describe the outcomes of investigations of procedures to train caregivers to implement behavior change procedures with integrity. The extant literature was reviewed to identify 1) the number of studies in which levels of treatment integrity were manipulated systematically, 2) the types of errors investigated, 3) which parts of the intervention procedure were manipulated, and 4) the degree to which these errors affected participant behavior. The effects of a video modeling program to increase procedural integrity with graduated guidance to 3 direct care teachers were investigated, with results showing benefits in both treatment integrity and student outcomes. Two studies investigated the development, implementation, and evaluation of a large-scale behavioral skills training program in a large residential/training facility. Following initial development and establishment of a pyramidal training program, maintenance of trainer and direct caregiver skills were assessed and, when necessary, remedial training was provided. Procedures for assessing generalization of caregiver skills to natural environments were developed, implemented, and evaluated.

Treatment integrity measures the extent to which direct care staff implement procedures consistent with the prescribed protocols. Errors made by direct care staff when implementing teaching or treatment procedures may impede progress or harm consumers. In recent years, treatment integrity research has begun to assess how specific types of treatment integrity errors affect consumer behavior. Studies of this type involve manipulating systematically the degree to which treatment integrity errors are administered and measuring their effect on consumer behavior. The current review evaluated articles published across seven behavior analytic journals between 2000 and 2014. The main objectives of this review was to identify 1) the number of studies in which levels of treatment integrity were manipulated systematically, 2) the types of errors investigated, 3) which parts of the intervention procedure were manipulated, and 4) the degree to which these errors affected participant behavior. Twelve studies from eight articles met inclusionary criteria. Results showed that a majority of studies involved children with disabilities, took place in a school setting, and manipulated errors during the consequence component of treatment.

The purpose of the present investigation was to evaluate the effectiveness of video modeling as a teacher training tool to improve procedural integrity of graduated guidance. Additionally, the effectiveness of a prescribed graduated guidance procedure was assessed. The implementation of the video modeling training procedure was assessed with a non-concurrent multiple baseline design. Participants were 3 teenaged students diagnosed with autism and 3 direct care teachers. Narrated video models of the lead experimenter and a confederate were used to train the teachers to implement a graduated guidance procedure with their students for three house-hold chores. Results showed that the video models were an effective training tool and that increases in procedural integrity generalized to untrained tasks. Furthermore, the students learned the tasks with the prescribed graduated guidance procedure. The social validity of the video modeling training procedure was assessed with a participant completed survey. The importance of using effective training methods for both teachers and students are discussed. Data reliability was collected in 46% of sessions and range from 91% to 100% agreement.

Cooper, Heron and Heward (2007) define maintenance as the extent to which the learner continues to perform the target behavior after the intervention has been terminated. Maintenance is important because it ensures that long-lasting behavior change is occurring, and that gains were sustained following the termination of a treatment program. In addition, once it is proven that a learner’s skills have remained in the repertoire the assessment of generalization is possible. Previous literature in behavior skills training has assessed maintenance in a variety of settings and for a variety of skills. Following maintenance assessments, booster sessions are commonly used to re-train skills that did not maintain at criterion levels. The current project assessed the maintenance of caregiver’s skills following a training package used to teach behavior management skills at a large, residential facility as well as the maintenance of the professional staff’s skills in implementing the training package. The project developed, implemented, and evaluated procedures to assess the caregivers maintenance of the skills and, if needed, to re-establish those skills using 5-15 minute booster sessions.

In order for the benefits of a behavior management package to reach clients, the caregivers must use the behavior management package in the natural environment. Caregivers at a state residential facility were previously taught three behavior management tools to competency. Initial training and subsequent maintenance probes with booster sessions (2-22 months after training) were assessed through contrived role-plays. Generalization of behavior management tools in the natural environment is difficult to assess because opportunities to utilize the tools are not programmed by the researcher, rather they are contingent on client behavior. Therefore, the current project systematically defined opportunities to use the tools prior to assessing generalization. Generalization of the behavior management package was assessed by observing caregivers use of the tools when the opportunities arose in the natural environment. Training procedures included prompting the staff when to use behavior management tools in the natural environment followed by immediate feedback and using a constant prompt delay to teach staff to identify opportunities to use behavior management tools.