Artificial Intelligence in Medical Education: Promoting Active Learning With a Customized Chatbot Tool

Area of Science:

• Medical Education Technology
• Artificial Intelligence in Healthcare
• Physician Assistant Education

Background:

• Artificial intelligence (AI) tools, like customized chatbots, offer potential to improve active learning in physician assistant (PA) education.
• This study assessed student perceptions and outcomes of integrating an AI chatbot for self-directed learning, retrieval practice, and personalized feedback in a clinical medicine course.

Purpose of the Study:

• To evaluate the impact of a course-specific AI chatbot on student learning, engagement, and faculty workload in PA education.
• To assess student perceptions of an AI chatbot designed to enhance self-directed learning and exam preparation.

Main Methods:

• A GPT-4o chatbot was developed, integrating lecture materials, medical texts, and PANCE-style questions for customized practice and remediation.
• Student utilization data and anonymous survey responses on usefulness and engagement were collected and analyzed.
• Quantitative data were analyzed descriptively, and qualitative survey responses underwent thematic analysis.

Main Results:

• 90% of students completed surveys, with 95% reporting improved exam preparation and information retention via the AI chatbot.
• Key beneficial features included practice questions and individualized feedback, with AI use increasing significantly throughout the course.
• Qualitative analysis revealed themes of enhanced active recall, self-assessment, and confidence; faculty reported reduced tutoring and quiz creation time.

Conclusions:

• Integrating a customized AI chatbot in PA education fostered active learning, self-efficacy, and AI literacy, while decreasing faculty workload.
• Early exposure to AI tools can improve student readiness for assessments and develop skills for responsible clinical technology use.