Evaluation of a self-instructional package for teaching tutors to conduct discrete-trials teaching with children with autism

Abstract

A widely used instructional method for teaching children with autism is Applied Behaviour Analysis (ABA), and a main component of ABA programming is discretetrials teaching (DTT). DTT is made up of a series of brief teaching trials, with each trial including an antecedent (e.g., an instruction from the teacher), a response from the child, and an immediate consequence (e.g., a reinforcer provided for a correct response). Using a modified multiple-baseline design across participants, Experiment 1 assessed the effectiveness of a DTT self-instructional package for teaching 4 pairs of newly-hired ABA tutors how to apply 21 components of DTT to teach 3 tasks to a confederate role-playing a child with autism. In Experiment 2, a group of 8 additional tutors were each independently presented with the same procedure. In both experiments, in Phase 1(baseline), tutors attempted to teach the 3 tasks to the confederate. In Phase 2 (manual), tutors mastered a 37-page self-instructional manual and once again attempted to teach the same 3 tasks to the confederate. Phase 3 was a within-subject AB design component. That is, if tutors did not meet a DTT mastery criterion of 80% after studying the manual in Phase 2, then they watched a brief video demonstration of a DTT expert teaching a task to a child role-playing a child with autism (Fazzio, 2007), and then once again attempted to teach the 3 tasks to the confederate. Across both experiments: a) The 16 tutors averaged 4.6 hours to master the self-instructional manual, and showed an average improvement in DTT accuracy of 32.2% from baseline; b) For the 13 tutors who did not meet the mastery criterion after reading the manual, their average DTT performance improved an additional 12% after watching the video; c) 13 of the 16 participants met mastery (3 after the manual and 10 after the manual plus video), and the other 3 tutors were very close to mastery. The results suggest that the training package is

Review and Discussion of Strategies to Address Low Trainer‐to‐Staff Ratios

“…Other extensions of the Salem et al study demonstrated the efficacy of SI plus video modeling for eight tutors to master delivery of three discrete-trial programs with confederates (M = 4.4 h of training; Thomson et al, 2012) and the efficacy of a package including SI plus video modeling plus role-play with feedback for parents to master delivery of three discrete-trial programs with their children (M = 4.68 h of training; Young, Boris, . Both studies demonstrated mixed effects of SI, alone (e.g., five of eight participants required addition of video model; Thomson et al).…”

Training Novice Instructors to Implement Errorless Discrete-Trial Teaching: A Sequential Analysis

“…Participants are often administered a quiz on the manual content, and then asked to demonstrate their skills with a confederate. Although this method of instruction often results in substantial improvements in DTT skills (e.g., Arnal et al, 2007;Salem et al, 2009;Thompson, Martin, Fazzio, Salem, Young, & Yu, 2012), most participants require additional instruction (e.g., video demonstration, rehearsal, feedback) in order to demonstrate mastery and generalization of relevant skills Thiessen et al, 2009). Some studies have demonstrated the effectiveness of rehearsal and performance feedback without prior selfinstruction and have resulted in mastery of DTT skills (Gilligan, Luiselli, & Pace, 2007;LeBlanc, Ricciardi, & Luiselli, 2005).…”

Preference assessment training via self-instruction: A replication and extension

“…Specifically, technicians had experience providing behavior‐analytic services to children. As a result, they may have a history of using instructional packages to learn how to implement various behavior‐analytic procedures (e.g., Thomson et al, ; Vladescu, Carroll, Paden, & Kodak, 2012; Weldy et al, ), whereas the undergraduate students did not share that same history. In Experiment 2, our main objective was to identify the effective components of the feedback intervention used in Experiment 1.…”