Agentic education · Learning science

Actionable companion · Motivation & engagement

From AI evidence to responsible practice: Impact of ChatGPT-assisted personalized learning on teaching acute abdomen to undergraduate medical students: A randomized crossover study

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

Read original paperSearch paper agentCC BY 3.0

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): Higher education. Health-professions education findings do not establish effectiveness in K–12 classrooms.

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

ABSTRACT Background &Objective: The application of Artificial Intelligence (AI) in medical education has emerged as a promising avenue for personalized learning experiences to the individual needs of medical students. This study investigated the impact of AI-driven personalized learning pathways on the academic performance of medical students in acute abdomen topic, comparing against traditional learning method. Methodology: This study used a randomized controlled crossover trial conducted from February 2024 to July 2024 among fourth year one hundred undergraduate medical students at Islamic International Medical College, Riphah International University, Pakistan. In this study, students enrolled in a general surgery course were randomly assigned to experimental group and control group following a pre-test. The experimental group engaged with AI-driven personalized learning pathways, powered by ChatGPT-4, the control group utilized conventional educational resources. Both groups completed a post-test to assess the effects of their respective learning interventions. Statistical analyses, including descriptive statistics, independent-samples t-tests and paired-samples t-tests were conducted using SPSS. Results: The pre-test scores of the experimental (M = 17.12, SD = 6.99) and control groups (M = 18.64, SD = 6.91) did not differ significantly (p = 0.277). Post-intervention, the experimental group showed a statistically significant improvement (M = 22.8, SD = 5.11) compared to the control group (M = 19.24, SD = 7.11), with a p-value of .005 and a moderate effect size (Cohen’s d = 0.58). This suggests that AI-driven personalized learning pathways had a positive and measurable impact on the students’ academic performance in the experimental group. Conclusion: AI-driven personalized learning pathways can enhance the academic performance of medical students, particularly in subject of surgery. Future research should explore the long-term effects of personalized AI-powered educational interventions.

What kind of evidence is this?

Supporting AI learning science · Randomized study reported

Stage: Higher education. 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

The study included fourth-year undergraduate medical students enrolled in the general surgery course, focusing on the “Acute abdomen” topic. Students with declared learning disabilities or significant mental health conditions were excluded from this initial efficacy study to reduce heterogeneity in learning support needs and because the intervention had not yet been evaluated within a framework that included tailored accommodations. These exclusions were based on self-report and institutional disclosure rather than formal diagnostic screening. Baseline technology-use survey showed no prior use of AI-assisted personalized learning platforms. Stratified block randomization was used to ensure balance in gender and baseline pre-test scores between groups.17 No formal washout period was introduced between phases because the intervention involved topic-specific knowledge acquisition, and learned educational content could not be practically or ethically withdrawn.

Limitations and uncertainty · source passage

It was conducted at a single institution on one focused surgical topic, which limits generalizability. Students with declared learning disabilities or significant mental health conditions were excluded, which may have introduced selection bias and reduced inclusiveness. No formal washout period was used between crossover phases; therefore, carryover effects cannot be excluded. In addition, the short duration of the intervention did not permit evaluation of long-term retention or sustained educational impact.

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.

  • 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. Choose one disciplinary task. Require a first attempt, a documented AI consultation, a critique of the output and an independent explanation. Align the comparison group and scoring rubric with the same objective. 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. Blind-score reasoning and accuracy on a new task without AI. Distinguish model examination scores, student perceptions and actual learner gains; only the last directly tests educational benefit. 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

Maria Ilyas; Rehan Ahmed Khan; Rahila Yasmeen; Zainab Kamal; Noor ul Ain. Impact of ChatGPT-assisted personalized learning on teaching acute abdomen to undergraduate medical students: A randomized crossover study. Pakistan Journal of Medical Sciences (2026-08-01). https://doi.org/10.12669/pjms.42.8.14827

Companion prepared 2026-11-01. 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 3.0 license. This guide has not undergone independent peer review.