Toward automatic motivator selection for autism behavior intervention therapy.
A phone app can learn and update which reinforcers keep each child with autism engaged, saving staff guesswork.
01Research in Context
What this study did
Siyam et al. (2022) built a phone app that learns which toys or snacks a child with autism likes best.
The app uses Q-learning, a type of AI that gets smarter each time the child picks or refuses an item.
Staff tapped the screen after every trial; the system updated its list of top reinforcers on the spot.
What they found
The more the app was used, the better its suggestions matched what actually kept the child working.
Staff rated the tool easy to use and saw kids stay engaged longer when the app chose the reinforcer.
How this fits with other research
Gibson et al. (2021) reviewed 388 play-based studies for children aged 2-8. Their map includes any toy or game that boosts social skills. The new AI app lands inside that map, but it adds a self-updating twist: the toy list changes daily instead of staying fixed.
Naidoo et al. (2020) built a low-tech picture board so kids could point to dental tools they liked. Both projects give the child a voice, one with paper symbols and one with data. Together they show that choice—high or low tech—cuts problem behavior.
Li et al. (2015) found that kids with lower IQ and more tantrums refuse dental exams most often. The AI app could help here by quickly finding ultra-strong reinforcers for exactly those hard-to-motivate children, turning a likely “no” into a “yes.”
Why it matters
You no longer have to guess what will work today. Let the app run for a week, then use its top three items during teaching or dental desensitization. The list keeps evolving, so you stay one step ahead of satiation without extra prep time.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Run the app for ten trials at breakfast, then use the top item in your first work session.
02At a glance
03Original abstract
Children with autism spectrum disorder (ASD) usually show little interest in academic activities and may display disruptive behavior when presented with assignments. Research indicates that incorporating motivational variables during interventions results in improvements in behavior and academic performance. However, the impact of such motivational variables varies between children. In this paper, we aim to address the problem of selecting the right motivator for children with ASD using reinforcement learning by adapting to the most influential factors impacting the effectiveness of the contingent motivator used. We model the task of selecting a motivator as a Markov decision process problem. The states, actions and rewards design consider the factors that impact the effectiveness of a motivator based on applied behavior analysis as well as learners’ individual preferences. We use a Q-learning algorithm to solve the modeled problem. Our proposed solution is then implemented as a mobile application developed for special education plans coordination. To evaluate the motivator selection feature, we conduct a study involving a group of teachers and therapists and assess how the added feature aids the participants in their decision-making process of selecting a motivator. Preliminary results indicated that the motivator selection feature improved the usability of the mobile app. Analysis of the algorithm performance showed promising results and indicated improvement of the recommendations over time.
Universal Access in the Information Society, 2022 · doi:10.1007/s10209-022-00914-7