Enhancing self-directed learning with custom GPT AI facilitation among medical students: A randomized controlled...

Area of Science:

• Medical Education
• Artificial Intelligence in Education
• Cognitive Science

Background:

• Self-directed learning is crucial for medical students.
• Traditional problem-based learning (PBL) can be enhanced with technological support.
• Assessing the efficacy of AI tools in medical training is an emerging area of research.

Purpose of the Study:

• To evaluate the impact of LearnGuide, a specialized ChatGPT tool, on medical students' self-directed learning.
• To determine if AI-assisted PBL training improves critical thinking and flow states compared to traditional methods.
• To assess the sustained effects of LearnGuide over a 14-week period.

Main Methods:

• A 14-week randomized controlled trial involving 103 medical students.
• Intervention group received PBL training with LearnGuide support; control group received identical training without AI.
• Primary outcomes included Self-Directed Learning Scale scores; secondary outcomes included Cornell Critical Thinking Test and Global Flow Scores.

Main Results:

• The LearnGuide group showed significant improvements in Self-Directed Learning Scale scores at 12 weeks, sustained at 14 weeks (p=0.01).
• Notable enhancements in Cornell Critical Thinking Test scores were observed at 12 weeks (p<0.001) and persisted.
• The intervention group also demonstrated superior Global Flow Scores from week 6 through week 12.

Conclusions:

• LearnGuide significantly boosts self-directed learning capabilities in medical students.
• AI-powered tools like LearnGuide demonstrably enhance critical thinking and engagement.
• The findings underscore the value of integrating AI into medical education for improved learning experiences and outcomes.