Actionable companion · Feedback & scaffolding
Learning with agency: GPT-4’s capabilities for formative and summative assessments in Norwegian medicine exams—an intrinsic case study in the early phase of intervention
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.
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:
The growing integration of artificial intelligence (AI) in education has paved the way for innovative assessment methods. This study explores the capabilities of GPT-4, which is a large language model (LLM), on a medicine exam and for formative and summative assessments in Norwegian educational settings. This research builds on our previous work to explore how AI, specifically GPT-4, can enhance assessment practices by evaluating its performance on a full-scale medical multiple-choice exam. Prior studies have revealed that LLM’s can have certain potential in medical education but have not specifically examined how GPT-4 can enhance formative and summative assessments in medical education. Therefore, my study contributes to filling gaps in the current knowledge by examining GPT-4’s capabilities for formative and summative assessment in medical education in Norway. For this purpose, a case study design was employed, and the primary data sources were 110 exam questions, 10 blinded exam questions, and 2 patient cases within medicine. The findings from this intrinsic case study revealed that GPT-4 performed well on the summative assessment, with a robust handling of the Norwegian medical language. Further, GPT-4 demonstrated a reliable evaluation of comprehensive student exams, such as patient cases, and, thus, aligned closely with human assessments. The findings suggest that GPT-4 can improve formative assessment by providing timely, personalized feedback to support student learning. This study highlights the importance of both an empirical and theoretical understanding of the gap between traditional assessment methods and educational practices and AI-enhanced approaches—particularly the importance of the ability of chain-of-thought prompting, how AI can scaffold tutoring, and assessment practices. However, continuous refinement and human oversight remain crucial to ensure the effective and responsible integration of LLM’s like GPT-4 into educational settings.
What kind of evidence is this?
Supporting AI learning science · Design requires full-text appraisal
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
The growing integration of artificial intelligence (AI) in education has paved the way for innovative assessment methods. This study explores the capabilities of GPT-4, which is a large language model (LLM), on a medicine exam and for formative and summative assessments in Norwegian educational settings. This research builds on our previous work to explore how AI, specifically GPT-4, can enhance assessment practices by evaluating its performance on a full-scale medical multiple-choice exam. Prior studies have revealed that LLM’s can have certain potential in medical education but have not specifically examined how GPT-4 can enhance formative and summative assessments in medical education. Therefore, my study contributes to filling gaps in the current knowledge by examining GPT-4’s capabilities for formative and summative assessment in medical education in Norway. For this purpose, a case study design was employed, and the primary data sources were 110 exam questions, 10 blinded exam questions, and 2 patient cases within medicine. The findings from this intrinsic case study revealed that GPT-4 performed well on the summative assessment, with a robust handling of the Norwegian medical language. Further, GPT-4 demonstrated a reliable evaluation of comprehensive student exams, such as patient cases, and, thus, aligned closely with human assessments. The findings suggest that GPT-4 can improve formative assessment by providing timely, personalized feedback to support student learning. This study highlights the importance of both an empirical and theoretical understanding of the gap between traditional assessment methods and educational practices and AI-enhanced approaches—particularly the importance of the ability of chain-of-thought prompting, how AI can scaffold tutoring, and assessment practices. However, continuous refinement and human oversight remain crucial to ensure the effective and responsible integration of LLM’s like GPT-4 into educational settings.
Limitations and uncertainty · source passage
This study used GPT-4 with a default temperature setting of 0.7, allowing for a certain amount of variability in responses. Only a single iteration per question was performed, and the results were accepted as final. Future studies could use a temperature setting of 1.0 and multiple iterations to better evaluate consistency and provide a more robust statistical analysis of GPT-4’s accuracy in summative assessments.
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.
- Give feedback on the learner’s reasoning, then fade hints. Record whether the learner can complete the next task without that scaffold.
- 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
- 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.
- 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.
- 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.
- 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
Rune Johan Krumsvik. GPT-4’s capabilities for formative and summative assessments in Norwegian medicine exams—an intrinsic case study in the early phase of intervention. Frontiers in Medicine (2025-04-10). https://doi.org/10.3389/fmed.2025.1441747
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.