Agentic education · Learning science

Actionable companion · Retention & transfer

Learning with agency: Implementation and Educational Impact of a Story-Centered Curriculum Using a Large Language Model: A Class on Internal Disorders for Physiotherapy Students

AI-assisted editorial implementation guide, separate from the original research. Proposed actions are not claims or recommendations attributed to the source authors. No independent peer review of this guide is asserted.

Read original paperSearch paper agentCC BY 4.0

Evidence and context

Title and abstract screening; stage labels describe reported settings, not verified numerical ages. Design labels require full-text appraisal. Reported stage(s): Age / stage not established. Verify the participant population and task before transferring the findings.

Original authors’ abstract, reproduced under the linked source license:

Background Story-centered curricula (SCC) can effectively enhance clinical reasoning and learner engagement; however, developing high-fidelity scenarios is resource-intensive. We developed an SCC using a large language model (LLM; ChatGPT, GPT-4) to generate longitudinal respiratory cases for physiotherapy students and evaluated its educational impact. Methodology This single-institution, quasi-experimental study used a non-randomized historical control design. Physiotherapy students in a 15-session course were divided into an SCC group, which completed eight LLM-generated narrative sessions, and a control group, which received traditional case-based sessions without a continuous storyline. The primary outcome was the change in score on a 30-item, 300-point proficiency test administered before and after the intervention. Secondary outcomes included five Likert-scale items evaluating learner experience. Group comparisons used appropriate parametric or non-parametric tests, and multivariable linear regression adjusted for age, sex, program, and pre-test score. Results The final sample consisted of 169 participants, with 92 in the SCC group and 77 in the control group. Baseline characteristics and pre-test median scores were 90 with an interquartile range (IQR) of 70 to 100 in the SCC group and 90 with an IQR of 70 to 110 in the control group, showing no significant differences between groups. The SCC group demonstrated a greater improvement in test performance, with a median change of 105 and an IQR of 78 to 140, compared with a median change of 80 and an IQR of 50 to 110 in the control group (p < 0.001). The SCC group also achieved higher post-test scores, with a median of 195 and an IQR of 160 to 223, compared with a median of 180 and an IQR of 150 to 200 in the control group (p = 0.013). Positive questionnaire responses (scores of 4 or 5) exceeded 90% across all domains, including immersion 87 (94.6%) and learning retention 88 (95.7%). Participation in the SCC program remained an independent predictor of post-test performance, with a regression coefficient of 23.87 and a 95% confidence interval of 11.32 to 36.42 (p < 0.001). Conclusions An SCC utilizing an LLM is an innovative educational approach that effectively balances improved learning outcomes with efficient scenario development, offering significant potential to advance simulation-based education in physiotherapy.

What kind of evidence is this?

Supporting AI learning science · Randomized study reported

Stage: Age / stage not established. Screening labels describe reported context and study-design language. They are not a risk-of-bias rating or proof of effectiveness for an age band.

Study methods · source passage

This quasi-experimental study employed a non-randomized controlled trial design. The participants were physiotherapy students enrolled in a course on internal disorders at a single institution. The study included two cohorts, namely, the SCC group, comprising students from the 2025 academic year who received an intervention based on the SCC, and the control group, comprising students from the 2024 academic year who received conventional instruction. The exclusion criteria were as follows: (1) failure to provide informed consent, (2) failure to complete both pre- and post-course proficiency tests, and (3) withdrawal from the program during the study period, including leave of absence or dropout. No formal sample size calculation was performed because the aim was to evaluate the educational effectiveness of the SCC intervention across student populations.

Limitations and uncertainty · source passage

Nonetheless, certain limitations of this study must be acknowledged. This was a single-institution, non-randomized historical control study and, therefore, is susceptible to confounding factors, such as cohort differences, student background, faculty experience, and influences outside the curriculum. However, even after controlling for potential confounders through multiple regression analysis, significant differences in Δ proficiency test scores remained, which is noteworthy. The primary outcomes were limited to short-term knowledge acquisition. Long-term retention, clinical performance, and patient-related outcomes were not assessed, indicating the need for longitudinal studies.

Design the system before the pilot

Use this study to inform a learning mechanism or implementation question. It is not direct evidence that autonomous agents improve learning. Keep the proposed agent behavior separate from what the source actually tested.

  • Include a delayed assessment and an unfamiliar transfer problem without AI. Immediate supported performance cannot establish durable learning.
  • Measure engagement separately from achievement. Check whether novelty-driven participation persists and whether it translates to independent performance.
  • Use common rubrics, blind scoring where feasible and a fresh unaided reasoning task. Document permissible assistance and avoid treating AI-detector scores as proof of misconduct.

A 90-day district pilot

  1. Days 1–15 · Appraise. Have the curriculum lead and a teacher read the methods and limitations. Record the study design, sample, comparator, outcomes and evidence gaps. Define a local question that the study can help inform.
  2. Days 16–30 · Design. Treat the paper as background evidence until its population is verified in the full text. Do not assign an age-specific intervention or extend adult results to children. Name an accountable owner and set the baseline, success threshold and budget before implementation.
  3. Days 31–60 · Test. Start with a small voluntary cohort, preserve an accessible non-AI route and require teacher review. Use synthetic or de-identified examples; approve data handling and tool terms through district processes.
  4. Days 61–90 · Decide. Extract the participant ages or education stage, recruitment, task and comparator before selecting a local pilot. If these cannot be established, retain the paper for design discussion rather than age-band efficacy claims. Compare with baseline, document uncertainty and stop or revise if learning worsens, access is unequal or critical errors persist.

The board decision brief

Report the learning problem, evidence quality, participating population, total cost including review time, measured benefit, errors, access differences and remaining uncertainties. Make a dated recommendation to stop, revise or expand. A positive perception survey alone does not establish learning improvement.

Attribution and reuse

Shota Okuno; Kenta Kawamitsu; Tamotsu Yamaguchi. Implementation and Educational Impact of a Story-Centered Curriculum Using a Large Language Model: A Class on Internal Disorders for Physiotherapy Students. Cureus (2025-10-12). https://doi.org/10.7759/cureus.94390

Companion prepared 2026-10-30. Original authorship and DOI are preserved. The abstract is an attributed source extract; the implementation plan is new AI-assisted editorial material, offered under the same CC BY 4.0 license. This guide has not undergone independent peer review.