As behavioral scientists working in the field of autism, we often like to brag about the “gold-standard treatment” that is Applied Behavior Analysis (ABA). Yes, we have empirical evidence for the effectiveness of our interventions. Therefore, we employ ABA in our clinical work with children with autism. But is the existing evidence enough? More specifically, do we know enough about the conditions under which our interventions are effective? Are our clinical interventions supported by empirical evidence and guided by best practices? Given most of the research conducted by behavior analysts uses single-subject design, it is not usually possible to draw broad conclusions regarding the relevant client characteristics, features of the intervention (e.g., intensity, duration), or features of the programming that lead to success or failure of our intervention. When we say our interventions are supported by evidence, we often mean a handful of behavior analytic articles published studies with three to five participants demonstrating generally positive effects of a particular behavioral intervention. We know little about the participant characteristics and whether these are relevant to the outcomes obtained.
A recent paper published in Autism Research summarized findings of a Special Interest Group (SIG) sponsored by the International Society for Autism Research (INSAR) tasked with identifying knowledge gaps and research priorities in the field of early intervention for children with autism. Dr. Giacomo Vivanti and colleagues highlighted several gaps in knowledge such as intervention theory translating to applied practice (see the chapter by Dr. Critchfield and colleagues in Autism Service Delivery, for related discussion). Perhaps most importantly, Dr. Vivanti and colleagues highlighted the need for studies beyond proof-of-concept, small-scale studies (e.g., single-subject design) that allow proper evaluation of whether or not an intervention is effective. These studies would require large samples of participants to provide enough power and draw conclusions about features of the intervention and/or client that determine positive treatment outcomes.
As behavioral scientists and practitioners, we can (and should) be contributing to this literature. To do so, we should frame our research questions in a manner amenable to comparison with other interventions and that allows one to draw general conclusions about the usefulness of particular interventions. In the process, we should contact relevant literature outside our field to inform our research and practice. Finally, we should disseminate our findings more broadly, outside our flagship journals.
My goal is for this blog to be a space to discuss areas of behavioral science that can inform the work we do as behavior analysts in the field of autism and how we can add to the broader field of autism science. Thus, the goal is to connect the dots between behavior analysis, autism science, and clinical practice.
From time to time, we’ll feature guest bloggers and contributors from our Translational and Applied Behavioral Science Lab. Join us as we connect the dots!