Evaluating an iPad Game to Address Overselectivity in Preliterate AAC Users with Minimal Verbal Behavior
LouAnne E. Boyd, Kathryn E. Ringland, Heather Faucett, Alexis Hiniker, Kimberley Klein, Kanika Patel, Gillian R. Hayes · 2017 · Proceedings of the 19th International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS '17) · doi:10.1145/3132525.3132551
Summary
This paper evaluates Go Go Games, an iPad-based therapeutic game suite designed to teach Multiple Cue Responding (MCR) to preliterate, minimally verbal children who exhibit overselectivity — a learning challenge where individuals attend to only one or a few features of a stimulus while ignoring other relevant features. Overselectivity is common in children with autism, intellectual disabilities, and other developmental conditions, and it directly impacts their ability to use AAC devices effectively because it interferes with learning to discriminate between symbols that share visual features. For example, a child who is overselective might learn to identify a symbol for "tennis" based only on the yellow ball in the image, then confuse it with the symbol for "balloons" which also contains a yellow circle. Currently, screening and intervention for overselectivity require trained specialists and are time-intensive, leaving most of the affected population untreated. Go Go Games translates the principles of Pivotal Response Training (PRT) into digital form: children observe a graphic prompt (e.g., a train car), scan multiple options, select the correct match, and drag it to the correct location. The game systematically scaffolds difficulty, randomises placement, provides differentiated feedback, and uses child-preferred stimuli (trains, cars, robots). The study deployed the game in a special education school with 11 children ages 6-14 with multiple learning challenges, using a single-case experimental design with staggered groups.
Key findings
Of 11 enrolled participants, 3 were screened out (already demonstrating MCR mastery), 1 was dismissed for aggression, and 1 dropped out. Six participants completed the study with enough data for analysis, all AAC users. Using the Nonoverlap of All Pairs (NAP) metric standard in behavioral research, three of the six participants who received the full treatment dose (9 days, ~10 minutes/day) showed medium to strong treatment effects: P5 achieved 100% NAP (strong effect, from zero correct at baseline to mid-range post-intervention), P11 achieved 88% NAP (medium effect), and P2 achieved 83% NAP (medium effect). The remaining three showed weak effects (22-54% NAP), attributable to motor challenges (P1), behavioural issues (P8), and attendance problems (P7). Staff interviews and observations revealed important implementation insights: the iPad's form factor was inherently engaging for some children but "not reinforcing" for others; corrective feedback was critical but tolerance for being wrong varied; difficulty transitioning away from the game sometimes triggered aggression; classroom distractions competed for attention; and co-occurring physical disabilities (motor challenges requiring swipe gestures) created additional barriers. Staff could administer the game without specialist training, a key advantage over traditional MCR therapy.
Relevance
This paper highlights overselectivity as an underaddressed challenge in the assistive technology community that directly impacts AAC use. For accessibility practitioners and AAC designers, the key message is that overselectivity should be considered as a potential co-occurring condition when designing icon-based communication systems — if a user cannot attend to multiple features of a symbol, no amount of interface simplification will solve the underlying learning barrier. The generalizable design features from Go Go Games — repetitive brief interactions, systematic scaffolding, child-preferred stimuli, corrective feedback, progress visualisation — provide a template for translating evidence-based behavioural therapies into accessible digital interventions. The study also demonstrates the challenge of intersectionality in assistive technology: children with overselectivity often have co-occurring physical, cognitive, and behavioural conditions that each create distinct design challenges. The small sample size and variable engagement levels are limitations, but the single-case design is appropriate for this population and the results demonstrate feasibility for larger studies.
Tags: augmentative and alternative communication · autism spectrum disorder · children · cognitive accessibility · game accessibility · special education · tablet · behavioral therapy · overselectivity