Original title: The Compensatory Trap: Activation Cost, Behavioral Persistence, and the Limits of Embedded Accessibility Features
Trystan Geoffre, Thierry Geoffre · September 2026
Educational software often includes built-in accessibility tools like text-to-speech or picture dictionaries. Creators assume that simply offering these tools is enough to help struggling learners. Researchers tested this idea by tracking how elementary students used embedded tools during reading exercises. They wanted to see if students actually recognize when they need help and turn tools on while their brains are working hard. They found that students used these tools in fewer than 9% of exercises. Even when students struggled, tool availability did not give an extra boost to performance. Instead of activating tools when they hit a tough spot, students mainly used low-effort tools out of habit. Difficult tools were linked to actual need, but students used tools less the longer they were in the program. Offering tools is not enough because kids need guidance to notice difficulties and choose to use available help.
You are staying in a hotel that provides a list of free items you can borrow: a hairdryer, an iron, a power adapter. You look at the list but rarely call to request them, even when you could use one. Over your stay, you rely on the same habits—drying your hair with a towel—and never try the actual tools. Here, the reading software's accessibility tools are the hotel items: available but rarely used, and students persist in old habits instead.
This matters because schools and educational platforms invest heavily in built-in accessibility features, assuming availability alone is sufficient. The study shows that students use them less than 9% of the time, and tool use does not improve performance during struggle. For neurodivergent students, who need more support, the tools provide proportionally less help, deepening inequity.
The study is limited to French reading exercises for primary students aged 8-11. It does not test the proposed Dynamic Accessibility Scaffolding. The unmonitored ecological phase may introduce confounding factors. Results may not generalize to other subjects, age groups, or languages. Additionally, the study does not explore why neurodivergent students receive proportionally less support despite similar tool use rates.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
The Compensatory Trap: Activation Cost, Behavioral Persistence, and the Limits of Embedded Accessibility Features — Trystan Geoffre, Thierry Geoffre — EdArXiv — doi:10.35542/osf.io/79jgh_v1
Abstract licence: CC BY
Explanation generated by Synthesis from the abstract — it is not the original paper. This is a preprint, so it hasn't necessarily been peer reviewed.