Truace
Subscribe

Curriculum-Aligned Conversational AI to Support Pathology Self-Directed Learning in Preclinical Medical Students: A Crossover Randomized Pilot Study

Background Artificial intelligence (AI) is reshaping the landscape of medical education, yet the widespread adoption of general-purpose AI tools in student learning raises substantial concerns about curricular alignment, information accuracy, and the absence of pedagogical scaffolding. Objective This crossover pilot study evaluated a curriculum-aligned custom GPT that was developed to support AI-scaffolded self-directed learning (SDL), preparation that remains student-initiated but is supported by structured Soc…

By the Truvace desk1 min read1 sourcePeer-reviewedTRV-2026-2014· cite & verify
Curriculum-Aligned Conversational AI to Support Pathology Self-Directed Learning in Preclinical Medical Students: A Crossover Randomized Pilot Study
"Cytomegalovirus infection - Case 301" by Pulmonary Pathology is licensed under CC BY-SA 2.0. To view a copy of this license, visit https://creativecommons.org/licenses/by-sa/2.0/.

In brief

Researchers at the University of South Carolina's Floyd School of Medicine tested a custom GPT built around the course curriculum and guided by Socratic prompting. Forty-four second-year students prepared for two pathology small group discussions, using the tool for one case and ordinary resources for the other.

The results suggest that structured, curriculum-matched AI may help students prepare more effectively than unguided study. The evidence is preliminary, though: this was a small, single-course pilot, and knowledge retention later on showed no significant difference in the primary analysis.

Main points

  1. The study used a crossover design in which each student prepared one case with the custom GPT and one with conventional resources.
  2. The two cases were Cushing's disease and prostate cancer, discussed in faculty-led pathology small groups.
  3. SDL rubric scores averaged 10.3 with the custom GPT versus 9.4 out of 12 without it, a mean difference of 0.9.
  4. Delayed knowledge retention did not differ significantly in the primary analysis.

The gain

Second-year medical students who prepared with a curriculum-aligned custom GPT scored higher on faculty-assessed self-directed learning documentation than with conventional resources.

The rundown

The tool was designed to address concerns that general-purpose AI can stray from the curriculum, give inaccurate information, and lack pedagogical scaffolding. It used structured Socratic prompting while keeping preparation student-initiated.

Faculty rated students' self-directed learning documentation forms and their small group discussion performance using standardized rubrics. Participation was voluntary and took place at a single medical school.

The authors call for larger, multisite studies of curriculum-aligned AI tools. They also recommend prospective registration for future trials, since this pilot was not registered.

Sources

  1. Peer-reviewedJMIR Medical Education2026-10-13

The debate