A comparison of schematic and taxonomic iPad® AAC systems for teaching multistep navigational AAC requests to children with ASD

The variety of augmentative and alternative communication (AAC) applications available on devices such as the Apple iPad®, necessitates research comparing different application components. AAC applications can include a variety of display formats such as: visual scene displays (VSDs; with vocabulary embedded into images of a scene or context), grid displays (with rows and columns of symbol buttons representing vocabulary), and hybrid formats (combining elements of VSDs and grids). To navigate through multiple pages of vocabulary, VSDs and hybrids are often organized schematically (i.e., by context or location) and grids are commonly organized taxonomically (i.e., by category). This study compared how four young children (ages 4 to 8) with autism spectrum disorder (ASD) acquired two-step navigational requesting with an iPad® AAC application using a schematic VSD, or hybrid pop-up grid, and a taxonomic grid. Using a multielement design, acquisition was compared across two settings (e.g., living room, kitchen), and three categories of preferred items (e.g., drinks, food, toys). Intervention involved behaviorally-based strategies (e.g., time delay, least-to-most prompting). During intervention, three participants mastered the schematic systems (VSD or hybrid), but did not master the taxonomic grid. Two of these participants also generalized requesting with schematic systems to an untrained location with a new preferred item, and maintained responding across all three settings. A fourth participant mastered both a schematic VSD and a taxonomic grid during training. During generalization, she rapidly acquired requesting in the new environment with the schematic VSD, but did not meet mastery criterion with the taxonomic grid. Across participants, the most common error with schematic systems was selecting the wrong scene (i.e., selecting an image of a location that did not match the location of the given session). In contrast, all participants showed a greater variety of error types with the taxonomic grid (including selecting the wrong category symbol, pressing the screen multiple times, trying to activate the screen with the wrong motion, and selecting the wrong item symbol). Differences in the types of errors observed suggest possible advantages and disadvantages with each system. Results have important implications for the development of AAC assessment and implementation protocols.