Scaffolded AI-Supported Problem-Based Learning for Medical Interns: Exploratory Retrospectively Registered Randomized Controlled Evaluation With a Voluntary Feasibility Follow-Up
Zhengqi Zhao, Baijing Wu, Shuyuan Tian, Haigen Hu, Pengyi Hao · 2026
Read Scaffolded AI-Supported Problem-Based Learning for Medical Interns: Exploratory Retrospectively Registered Randomized Controlled Evaluation With a Voluntary Feasibility Follow-Up by Zhengqi Zhao, Baijing Wu, Shuyuan Tian, Haigen Hu, Pengyi Hao (2026) 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-13541230.txt.
Description
Zhengqi Zhao and four co-authors build an AI environment that tutors medical interns through cases instead of answering for them, in this 2026 study. Problem-based learning teaches clinical reasoning well but needs skilled facilitators, so it scales badly. Most education AI is an answer machine, which skips the reasoning the method exists to build.
Their system, MAPLE-CR, uses generative agents as scaffolding: prompts that push reasoning, simulated tutor and peer roles for the social side, and structured workflows with feedback for self-regulation. In a randomized evaluation with fifty-two interns, the scaffolded group beat case-matched self-study on posttests by a clear margin and reported high satisfaction and self-efficacy. Eighteen volunteers then used it weekly for four more weeks, showing repeated use was feasible with steady within-session gains. Participation in discussion lagged other ratings, a target for improvement.
Catalog record
| CREATOR | Zhengqi Zhao, Baijing Wu, Shuyuan Tian, Haigen Hu, Pengyi Hao |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 58 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |