Enhancing Emergency Medical Response Education through Generative AI-Powered Game-Based Learning: A Retrospective...

Area of Science:

• Medical Education
• Artificial Intelligence in Education
• Emergency Medical Response Training

Background:

• Traditional lecture-based learning (LBL) is inadequate for developing critical decision-making skills in high-stakes fields like Emergency Medical Response (EMR).
• Game-based learning (GBL) offers immersive training but often lacks real-time expert feedback.
• Integrating generative Artificial Intelligence (AI) as an intelligent tutor within GBL addresses this gap.

Purpose of the Study:

• To compare the effectiveness of LBL, GBL, and AI-GBL on medical students' knowledge acquisition and retention.
• To evaluate the impact of these learning modalities on student motivation and cognitive load.
• To assess the utility of AI-GBL in an EMR context.

Main Methods:

• A retrospective comparative study involved 86 medical students across three cohorts (2022-2024).
• Each cohort experienced one learning modality: LBL (n=29), GBL (n=28), or AI-GBL (n=29).
• Knowledge was measured using pre-test, post-test, and final-test scores; student feedback assessed motivation, cognitive load, and technology acceptance.

Main Results:

• Both GBL and AI-GBL significantly improved immediate knowledge acquisition compared to LBL.
• AI-GBL demonstrated superior delayed knowledge retention over both GBL and LBL.
• AI-GBL resulted in significantly lower cognitive load and was perceived as more useful than GBL.

Conclusions:

• AI-enhanced GBL is a promising model for EMR training, significantly improving knowledge acquisition and retention.
• This approach effectively reduces cognitive load, enhancing learning efficiency.
• AI-GBL shows potential for developing complex medical competencies in high-stakes environments.