Comparing Computer-Assisted and Teacher-Implemented Visual Matching Instruction for Children with ASD and/or Other DD.

Hu et al. (2020) compared two ways to teach visual matching to children with autism or developmental delay. One group used a tablet that tracked finger gestures. The other group used teacher-held flashcards.

The team ran short sessions every day and switched the teaching method back and forth. They counted how many prompts each child needed and how long each lesson took.

Both groups learned the matching skill equally well. The tablet group needed fewer hints and finished each lesson faster. Five weeks later the children still remembered the skill no matter which method they used.

Niland et al. (2026) got the same digital-vs-teacher pattern with a different task. They taught conditional discriminations using motion prompts on a screen and saw similar gains, showing the effect is not tied to one app.

Savaldi-Harussi et al. (2025) seems to disagree at first. Their Smart-Glove videos beat flashcards only for younger students with moderate delays, not older students with severe delays. The gap comes from population and task: Xiaoyi worked on simple matching with young ASD/DD kids, while Gat targeted global word reading in older students with more severe cognitive disability.

Older work backs the big idea. Craddock et al. (1994) already showed computers can teach adults with severe ID when practice is limited. Xiaoyi updates that story for young children and adds a direct teacher comparison.

If you run discrete trials in a classroom, you can swap some flashcard sets for a gesture-tracking app and save time without losing learning. Start with one student, one program, and track prompts and session length for a week. If the data look like Xiaoyi’s, you just gained minutes for other targets.