Improving communication skills of children with autism through support of applied behavioral analysis treatments using multimedia computing: a survey

Naturalistic applied behavior analysis (ABA) techniques have been shown to help children with autism improve their communication skills. Recognizing that individuals who interact with children regularly are in the position to utilize treatments with profound effects, researchers have examined methodologies for training parents, teachers, and peers to implement treatments. These programs are time intensive and often unable to support trainees after training. Technologies need to be examined to determine how they can aid in the educational and support process. Academic publications and publicly available training programs were reviewed to determine the types of participants, methodologies, and training durations that have been reported for instructing interventionists. These resources illustrate a need to make programs more accessible. To address this, selected computer science research is applied to methods of evaluating ABA implementations in order to recommend how the technologies could be utilized to make training and support programs more accessible. Review results of instructional programs, both in research and available in the community, illustrate the challenges in providing training in ABA methodologies. Modern research in multimedia data processing and machine learning could be applied to reduce the human cost of training and support individuals implementing ABA techniques. Utilizing machine learning techniques to analyze video probes of naturalistic ABA treatment implementation could alleviate the human cost of evaluating fidelity, allowing for greater support for individuals interested in the treatments. These technologies could be used in the future to expand data collection to provide more perspective on the treatments.