The Effects Of Systematically Fading Reinforcement While Maintaining Skills

A review of the literature describing treatment programs for young children with autism revealed two distinctly different strategies. One, the Lovaas approach (Lovaas, 1987), focused on an intense, one-on-one discrete trial training format implemented for approximately 40 hours per week. A prescribed curriculum targeted at social and communication skills is incorporated. The second program, TEACCH, emphasized an individualized curriculum selected after a careful assessment process to determine a person's strengths and limitations. Goals and objectives were subsequently implemented in highly structured special education classrooms where the focus was on antecedent control and the use of visually mediated strategies. Interviews with classroom teachers and educational consultants indicated problems with generalization, and specifically, the use of reinforcement when implementing the TEACCH program. This study describes a single subject multiple baseline across skills with a five year old male child with autism who was placed in a self-contained classroom where components of structured teaching and the TEACC~ philosophy were implemented. The purpose of the study was to determine the effectiveness of a fading procedure regarding reinforcement for skill acquisition, maintenance, and generalization. Initially, a reinforcer assessment was conducted to identify the participant's preferred reinforcers. A dense schedule of reinforcement was implemented during initial stages of skill acquisition followed by a structured fading procedure after the skill was acquired at a predetermined level of mastery (i.e., 80% for three consecutive days). Skills targeted for instruction included (a) the concept of 'beside' and (b) matching pictures of food items to actual food items. The first skill was acquired after only six days of direct instruction. The participant was then able to demonstrate the skill in both the direct instruction setting and the generalization setting of independent work even after reinforcement was faded and only offered at the end of each session. The second skill was monitored in three settings: (a) the direct instruction setting, (b) generalization setting of independent work, and ( c) the generalization setting of lunch. This skill was monitored during lunch to observe whether the participant was utilizing the skill to communicate items desired within a natural context. The participant made steady progress acquiring this skill, however he did not reach mastery before the close of the academic school year. Thus, the reinforcement fading procedure was not implemented for this skill. Data collected during lunch, however, were promising and indicated the participant utilized this skill between 65%-85% of the time. In conclusion, a discussion focuses on possible implications for further research;