What the study found
The study found that a personalized exercise-selection method based on contextual Thompson sampling, a way of choosing actions using current learner information and past performance, recommended exercises associated with greater skill improvement. It also adapted effectively to differences across learners.
Why the authors say this matters
The authors conclude that the framework enables personalized practice at scale. They also say it highlights exercises with consistently strong learning value and helps instructors identify learners who may benefit from additional support.
What the researchers tested
The researchers introduced a method that generates personalized sequences of exercises by selecting, at each step, the exercise most likely to advance a learner’s understanding of a targeted skill. Learning progress was measured as the change in estimated skill level before and after each exercise, and the method was tested using data from an online mathematics tutoring platform.
What worked and what didn't
Using the tutoring-platform data, the approach recommended exercises associated with greater skill improvement. The abstract also reports that it adapted effectively to differences across learners. It does not describe outcomes that did not work well.
What to keep in mind
The summary is based only on the abstract, so details about study design, comparison methods, and limitations are not described. The results are reported for an online mathematics tutoring platform, so the scope in the available summary is limited to that setting.
Key points
- The study tested a contextual Thompson sampling approach for choosing exercises.
- The method used learner information and past performance to guide exercise selection.
- Exercises recommended by the method were associated with greater skill improvement.
- The approach adapted effectively to differences across learners.
- The authors say the framework can support personalized practice at scale.
Disclosure
- Research title:
- Contextual Thompson sampling improved learner skill gains
- Authors:
- Lukas De Kerpel, Arthur Thuy, Dries F. Benoit
- Institutions:
- Ghent University, Ghent University, Ghent University, Ghent University Hospital, Ghent University Hospital, Ghent University Hospital, HOGENT University of Applied Sciences and Arts, HOGENT University of Applied Sciences and Arts, HOGENT University of Applied Sciences and Arts
- Publication date:
- 2026-03-10
- OpenAlex record:
- View
Get the weekly research newsletter
Stay current with scholarly research without reading academic papers — one filtered digest, every Friday.