Association for Behavior Analysis International

Aggregations of single-case research studies are useful in assessing the external validity of evidence in a given area. Aggregations of the literature can provide behavior analysts with a cohesive understanding of a certain practice; they can also inform us of the most effective treatments demonstrated for a specific domain of behavior. This symposium includes quantitative syntheses and meta-analyses of single-case literature on specific interventions (i.e., errorless teaching and embedding preferences) as well as on certain domains of behavior (i.e., academic skills and functional living skills). All of these reviews are focused on learning in individuals with autism spectrum disorder (ASD). Individuals with ASD face unique challenges in learning and development. A common purpose to each of the reviews here is to identify the current state of the evidence for individual practices. Standards for aggregating and analyzing single-case research have undergone much refinement in recent years. Meta-analytic reviews of single-case research are only recently becoming prevalent in the special education and applied behavior analytic literature. Thus, this symposium will be useful for attendees with interest in educational interventions for individuals with ASD as well as for those interested in advancements in methods for study aggregation.

This review of single-case research is on interventions aiming to enhance academic/pre-academic skills and/or task completion and engagement during academic contexts for children with ASD in school settings. Forty-eight studies published between 1995 and 2014 were included. These articles were coded according to (a) design; (b) participant demographics; (c) setting type; (d) intervention type and implementer; (e) duration and dosage; (f) outcomes; (g) inter-rater reliability, social validity, treatment integrity, and (h) generalization and maintenance. Tau-U was used to determine overlap indices for each study and across intervention categories of literacy, math, science, social studies, engagement and task completion, and prerequisite skills. Findings showed that most studies included students in grades K-12. The mean Tau-U score across all studies was moderate to high (Tau-U = 0.74, range 0.15 - 1.00). All interventions were effective or very effective in improving targeted pre-academic or academic skills. Interventions targeting task engagement and completion had the lowest mean Tau-U (Tau-U = 0.59); approximately half of the interventions were not effective for improving task engagement and completion (Tau-U = 0.65). These results reveal the need for further research targeting academic skills for preschool students with ASD and interventions targeting on task engagement and completion.

Errorless learning, or errorless teaching, is a technology that has been developed as a way to teach individuals with autism a variety of skills. The purpose of this systematic review was to review the extant literature on errorless teaching procedures that have been used with individuals with autism. Our goal was to determine which forms of errorless teaching have been researched and what the effects of each form were in terms of learning outcomes. We conducted a comprehensive search using 7 databases and studies were included in the review if they included participants with autism, were primary research, were written in English, and directly tested the effects of errorless teaching procedures. Twenty studies met the inclusion criteria and we conducted an independent appraisal of the results and methodological quality of each study (based of the What Works Clearinghouse Procedures and Standards Handbook). We found that five different forms of errorless teaching have been researched and that, in general, errorless teaching procedures are effective at teaching discrete skills to individuals with autism. We also found that errorless procedures seem to produce learning slower than non-errorless procedures in terms of sessions to mastery. Implications for research and practice will be discussed.

A body of research has explored how embedding interests into the learning environment by modifying aspects of the original activity may facilitate positive outcomes in an area of need for individuals with ASD. We reviewed this literature base to (a) determine the quality of the evidence and (b) evaluate potential moderators of results according to variations in participant characteristics and dependent variables. Studies were included that targeted participants with ASD, included quantitative data, and compared conditions with high- versus lesser-preferred objects/activities embedded into the environment. Studies were rated on quality and evidence using two rubrics. A total of 21 studies with 86 participants were included. Thirteen studies with 42 participants targeted children in early childhood, 12 of which were single-case studies with 28 participants and 74 effects/targeted dependent variables. In evaluating the quality of this initial sample of studies, findings were mixed. There was emerging support among studies targeting social-communication type skills (e.g., joint attention). It is possible that stereotyped or repetitive behaviors associated with restricted interests of children with ASD could deter the effects of this intervention at times, such as during instructor-led activities. Professionals should structure the environment to monitor added benefit when considering use of this intervention.

This systematic review and meta-analysis analyzed peer-reviewed research concerning the generalization and maintenance of functional-living skills for individuals with autism spectrum disorder. Following the application of predetermined inclusion criteria, 31 articles (containing 32 studies) were reviewed and summarized in terms of: (a) generalization dimension, (b) generalization assessment design, (c) maintenance assessment design, (d) maintenance and generalization teaching strategy, and (e) latency to maintenance probes. Using Tau effect size and post-hoc analyses, the type of generalization and maintenance teaching strategies were analyzed as potential moderators. Outcomes identified strong effects for generalization strategies employed. Resulting maintenance effects demonstrated strong effects for baseline to maintenance contrasts. Negative effects for the train and hope method were identified using intervention to maintenance contrasts. Results indicated statistically significant findings supporting the use of training in a natural setting and training to a criterion as compared to the train and hope strategy. Implications for future research and practice are discussed.