Teaching teens with autism to use everyday apps—like Maps, Weather, and Clock—to answer questions they don’t already know is a practical skill with real-world value. This study explored whether adolescents could learn to treat an iPad as a problem-solving tool, not just a device they know how to tap. For clinicians, the findings offer useful guidance on how to teach this kind of chained skill and where learners commonly get stuck.
What is the research question being asked and why does it matter?
The researchers wanted to know if teens with autism could learn to use common iPad apps to answer “hard” worksheet questions when they didn’t already know the answers. The apps were Maps (distance), Weather (temperature), and Clock (time).
This matters because many learners can answer questions they already know but get stuck when they need to find new information. In real life, people solve these “I don’t know” problems by using tools—like phones—instead of relying on adults.
The researchers also wanted to know if the skill would generalize. Here, that meant the teen could answer new questions with new cities, not just the ones taught. For one teen, it also meant answering out loud when the question was read to them.
For clinicians, the key issue isn’t “can they tap an app,” but “will they use the app as a problem-solving step when a question creates a need for information?”
What did the researchers do to answer that question?
They worked with two adolescents with ASD in a school setting. Both could do basic worksheet tasks, copy written text, and speak vocally.
The researchers created worksheets with three kinds of “problem” questions that couldn’t be answered from memory—because the city names always changed. They also added a few easy questions the students could already answer, to confirm they were willing and able to do worksheets without the iPad.
They measured the percent of problem questions answered correctly. The main goal was correct written answers; later, they also checked whether correct vocal answers happened when questions were read aloud.
During baseline, the iPad was available but no one taught the steps. Then they taught one app skill at a time using step-by-step prompt fading, running probes afterward to see if the student would now solve the worksheet questions.
Training didn’t start with “do the whole problem.” First, they taught parts of the chain: opening the right app and searching the city, saying a rule like “If it’s about time, use the clock app,” and underlining the key word in the question (time, temperature, distance).
For one student who still didn’t use the iPad during probes, they added extra steps—teaching her to follow her own rule (“use the iPad”) and to copy the answer from the screen to the worksheet. They also had to teach her to tell question types apart by mixing them together during training, because she confused which app matched which question.
How you can use this in your day-to-day clinical practice
If you want a learner to use apps to solve information problems, treat it like a chained skill that may need direct teaching. Don’t assume that “knowing how to open an app” will automatically turn into “using the app when a worksheet question is hard.”
In this study, both teens could learn the pieces, but only one started solving the worksheet problems after learning the keyword-and-rule step. The other could do the pieces during training but still didn’t use the app in probe sessions until extra links were taught—and, at times, prompts were added.
Start by defining the “problem” clearly: questions the learner can’t answer from memory, where the correct answer is available inside a tool. Use changing examples so the learner can’t memorize answers.
These researchers used different cities each time, keeping the establishing operation strong (the learner needed the app). You can do the same by rotating locations, dates, bus times, movie times, or store hours. If the learner can guess the answer, you won’t know if they’re truly using the tool.
Teach app navigation as its own skill first—but don’t stop there. The student needs to connect “this question is about X, so I should use Y app.”
In this study, the step that mattered for the faster learner was underlining the keyword in the question and then stating the rule. That combination may help some learners attend to the right feature of the question and reduce random app use. You can teach the learner to point to or underline “time,” “temperature,” or “distance,” then say a short rule, then act on it.
Plan for the possibility that self-talk doesn’t automatically control action. One learner could say the rule correctly but didn’t then use the iPad during probes.
The clinicians had to teach “listening to your own rule” as a separate step by prompting “Use the iPad” after the rule was stated. If you see a learner state the rule but then freeze, walk away, or do something else, don’t label it “noncompliance.” Treat it as a missing link in the chain, teach a clear next action, and fade that prompt.
Also plan for copying and response-format barriers. The study accepted answers without perfect formatting (no commas, units not required), which likely helped learners contact success sooner.
In clinic or school, loosen response requirements at first so the learner is practicing problem solving, not handwriting perfection. If the learner can find the answer but can’t write it, you may need to teach “copy from screen” as its own step—or allow alternatives like circling, matching, or typing. Only tighten formatting once the learner is reliably solving the problem.
Teach discrimination early by mixing problem types once the learner has some basic skill. One learner confused apps and showed weak discrimination until the researchers trained with time and temperature questions mixed on the same worksheet.
If you teach one app in isolation for too long, some learners may develop “always use this app” habits. A practical move: begin with two clearly different question types, mix them in short sets, and require the learner to check the keyword before choosing the app.
Use prompts carefully, and fade them on purpose. In this study, “rule prompts” (“Say your rule about time”) helped one learner perform during probes, and she sometimes kept using the rules even after prompts were removed.
Prompts like “What’s your plan?” or “Tell me your rule” can be helpful, but watch whether the learner becomes dependent on the adult prompt to start the chain. Your goal is for the worksheet question itself to cue the strategy.
Be cautious about generalization claims. This was a two-person study in one school with specific prerequisites (copying text, basic tacting, and city-name spelling/selection).
One teen showed strong generalization, including answering vocally and across apps. The other didn’t generalize well to vocal questions even after she could do the written worksheet problems. Don’t assume that teaching written problem solving will automatically transfer to spoken quizzes, classroom discussions, or community use. If you need vocal responding, teach and test it directly.
Finally, keep the goal meaningful and dignity-centered. The study used worksheets, which may not match real-life needs for every learner.
Consider teaching the same app-based problem-solving steps inside functional routines: checking weather to choose clothing, checking time zones before calling family, checking distance before planning travel, or checking store hours before a trip. Offer choice about which tool to use (app, voice assistant, asking a person), and focus on building independence without removing the learner’s right to ask for help when they want it.
Works Cited
Frampton, S. E., & Axe, J. B. (2025). A preliminary investigation into teaching adolescents with autism to use apps to solve problems. The Analysis of Verbal Behavior, 41, 26–39. https://doi.org/10.1007/s40616-024-00212-8
