Medical Student Ratings of AI and Human Concept Maps for Anatomy Education

74 medical students at Qatar University rated three concept maps—generated by a large language model, a large action model, and a human expert—on clarity, structure, and scientific accuracy. The data were analyzed using non-parametric and categorical tests. The supplementary material, a PDF file of 198.3 KB, was uploaded by figshare admin karger on May 7, 2026.

Use Cases

• Compare student preferences for AI-generated versus human-generated educational materials based on Likert-scale ratings.
• Analyze the perceived effectiveness of different AI models (LLMs vs. LAMs) in creating learning tools based on clarity and accuracy scores.
• Evaluate the correlation between concept map ratings and student-reported future use intentions.
• Benchmark AI-generated concept maps against expert benchmarks for scientific accuracy and engagement.

Strengths

• Includes ratings from 74 medical students, providing a specific sample size.
• Compares three distinct concept map sources: a large language model, a large action model, and a human expert.
• Likert-scale items measured specific attributes: clarity, structure, and scientific accuracy.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small (198.3 KB), indicating limited scope.

Provenance

Source

figshare

Collection Method

Cross-sectional study involving student evaluations.

Freshness

Last updated 2026-05-07 07:55:23; freshness should be verified.

Geography

Qatar University