Comparing embedded and non-embedded visual scene displays for one adult diagnosed with autism spectrum disorder: A clinical application of single case design

This study provides an example of using single case experimental design to guide data-driven, low-inference interventions for an individual using customized displays on a speech-generating device and can serve as an example for interventionists who need to compare intervention strategies to identify approaches that are most appropriate for a specific learner. This study compared the performance of locating vocabulary in embedded and non-embedded visual scene displays (VSDs) on a speech-generating device by one 19-year-old male diagnosed with autism spectrum disorder. A single case experimental alternating treatment design was applied and the participant was taught to locate 48 vocabulary items that were equally divided among eight VSDs (i.e. four embedded and four non-embedded VSDs, each containing six vocabulary items). Accuracy and latency data for locating target VSDs and vocabulary items were measured over 12 intervention sessions and one maintenance session. Results show that the participant was initially more accurate and faster in locating vocabulary among embedded VSDs, when compared to non-embedded VSDs. However, over the course of 12 intervention sessions and the maintenance session, his performance became similar among both embedded and non-embedded VSDs. Implications regarding the clinical utility of the single case alternating treatment design and limitations are discussed.