Evaluation of Few-Shot AI-Generated Feedback on Case Reports in Physical Therapy Education: Mixed Methods Study
Hisaya Sudo, Yoko Noborimoto, Jun Takahashi · 2025
Read Evaluation of Few-Shot AI-Generated Feedback on Case Reports in Physical Therapy Education: Mixed Methods Study by Hisaya Sudo, Yoko Noborimoto, Jun Takahashi (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12811036.txt.
Description
Hisaya Sudo and two co-authors tested AI feedback on student case reports in a 2025 mixed methods study. Writing comments on every report strains faculty time, and AI could help, but students may distrust machine comments. The twist tested was few-shot prompting: the same AI model, Gemini 2.5 Pro, wrote feedback either cold or after seeing nine teacher-written examples. Thirty-five final-year physical therapy students in Japan compared both versions on their own reports.
Three quarters preferred the example-guided feedback for overall usefulness. A questionnaire showed no overall score gap, though specific scales favored the guided version. Focus groups added texture, and the team also tracked how views shifted once students learned which text came from the AI. Example-guided feedback won trust more easily.
Catalog record
| CREATOR | Hisaya Sudo, Yoko Noborimoto, Jun Takahashi |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 37 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |