Association for Behavior Analysis International

Educational policy in the United States advocates for empirically validated teaching procedures to be utilized in the classroom. As educational systems continue to be levied with added responsibilities and diminishing resources, it becomes imperative to identify Evidence-Based Practices (EBP) that are efficacious and easily implemented in educational systems. Additionally, empirically based teaching is not frequently observed when considering University level instruction. This symposium presents recent research on evidence-based practice in both primary and University level teaching with a focus on functional changes in both teacher and child behavior. Data-based presentations will outline the critical features of EBP in educational settings, summarize methods for effectively training teachers on EBP, identify critical features for implementation of EBP, and evaluate the effects on student behavior.

We examined the effectiveness of a Behavioral Skills Training (BST) package to instruct 3 parents to teach manding skills to young children with autism. Sessions included 5 bi-weekly modules with instructions, video modeling and role plays. Following completion of each training module, we observed parents implementing procedures in family homes or at a university-affiliated autism treatment center. Data were collected on the accuracy of implementing training steps for parents and the frequency of manding responses in children. We provided feedback and modeling of correct procedures following training. A changing-criterion design was used to evaluate the effects of the training. Results showed the training was effective for all parents, and findings were maintained during 1- and 2-month probes. We observed variable increases in manding responses for the children. Findings are discussed in terms of future applications of a BST model for enhancing parent skills to teach verbal behavior.

Behavior Skills Training (BST) typically consists of an initial informational component presented to the learners either vocally, through a handout, presentation, or both. Results from the active student responding literature indicates these methods as the least effective means of conveying important information to learners. This study sought to utilize an alternative instructional component, fluency training, and to evaluate if any effects are observed on implementation of the behavior chain of Discrete Trial Training (DTT). Teachers had previous training and experience on implementing DTT prior to this studyhowever, all teachers implemented strategies with low integrity. Teachers were trained to fluent levels on verbally stating the component steps of DTT and were then observed during probe sessions to evaluate percentage of steps implemented correctly. The probes indicate an initial improvement, but decreases over time that are consistent with results on other passive in-service trainings. Teachers then took part in a single session of Modeling, Role-Play, and Feedback. Results suggest that while fluency training had an impact on participants verbal performance on discrete trial information, and affected overt performance during subsequent probes, the effects were small and transient. Performance improved only after training on the components of BST and additional training had been completed in-situ.

This project assessed the use of behavioral momentum on the latency and accuracy of math problems. Behavioral momentum involved the sequencing of high probability tasks (e.g., two single digit math problems) before a student was presented with a low probability behavior (e.g., triple digit math problem). Assessment modules (3 to 5 sessions each) were used to assess baseline measures: (1) ten 3+3 digit math problems, and (2) twenty 1+1 digit math problems. Following assessment, students were given a 30 problem math test with 3 + 3 digit math problems embedded with a sequence of 1 + 1 digit math problems. Latency to each keystroke, duration of problem completion, and accuracy of each problem was monitored and analyzed. The use of behavioral momentum resulted in a decrease in latency in the completion of triple-digit addition problems with a concurrent increase in accuracy of triple-digit addition problems.