In a pilot with eight medical students, multimodal MedGame was perceived as more engaging and useful than text-only MedGame and original case material¶

Claim · All claims
Evidence · 1 study · causal r1 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Multimodal MedGame achieved the highest overall perception score, with a significant one-sided paired advantage over text-only MedGame (p = 0.0039). → Qian Wu 2026

Evidence¶

Qian Wu 2026¶

Qian Wu, Xinrong Zhou, Zizhan Ma, Kai Chen, Zheyao Gao, Xun Lin, Hongqiu Wu, Longfei Gou, Yixiao Liu, Ann Sin Nga Lau, Qi Dou. (2026). MedGame: Storytelling Gamification Empowered by Large Language Models for Medical Education. arXiv preprint. https://arxiv.org/abs/2607.21570 doi:10.48550/arxiv.2607.21570

q2 · i? · causal · r1

Paired pilot perception study with eight senior medical students, each rating the same five clinical cases under original case, text-only MedGame, and multimodal MedGame conditions on 1–5 Likert scales; student-level means were averaged over cases before testing. No effect size printed.

"multimodal MedGame achieves the highest overall perception score (4.19), followed by text-only MedGame (3.79) and original case material (3.19); the multimodal advantage over text-only is significant under a one-sided paired Wilcoxon signed-rank test over student-level means (p= 0.0039 )."

Discussion¶

Related Claims¶

• Without fine-tuning, commercial frontier models outperform open-source LLMs on MedGame Bench structural validity and content metrics — related
• Related-case overlap between MedGame Bench train and test splits shows no significant performance difference — related
• LLM-as-a-Judge scores correlate with blinded human expert judgments, more strongly for Medical Narrative Generation than Story Direction — related
• Task-specific fine-tuning substantially improves open-source LLMs on MedGame Bench and narrows the gap with commercial models — related