Number
820
Name
Integrating Socratic Pedagogy in an Evidence-Based AI Tutor
Date & Time
Sunday, June 7, 2026, 5:30 PM - 7:00 PM
Location Name
Oglethorpe Ballroom
Speakers
Authors
Anusha Aiyar, Medical College of Georgia
Henry Moon, Medical College of Georgia
Presentation Topic(s)
Technology and Innovation
Description
PURPOSE
Effective interpretation of complex medical topics, such as pulmonary
physiology, is challenging for pre-clerkship students due to limited
classroom exposure and static explanations provided by standard tools. The
Socratic method, characterized by structured questioning that encourages
learners to articulate reasoning and examine assumptions, is a proven
pedagogical strategy. This study systematically reviews its application in
pre-clerkship education and examines its integration into our Pulmonary
Function Test (PFT) learning assistant, an augmented GPT designed to enhance
evidence-based reasoning.
METHODS
Following PRISMA 2020 guidelines, we systematically searched for studies on
Socratic questioning, guided inquiry, or structured dialogue in pre-clerkship
education. Of 71 records screened, 38 full texts were reviewed and 29 met
inclusion criteria. Narrative synthesis identified common instructional
themes. Our PFT learning assistant applies a Retrieval-Augmented Generation
framework to combine curated scientific literature with adaptive Socratic
questioning to reinforce understanding and promote critical reasoning.
RESULTS
Five major themes emerged: (1) Socratic questioning enhances critical
reasoning and internalization of learning; (2) AI-supported Socratic methods
provide personalized feedback and assess understanding; (3) Socratic
strategies promote active participation and self-directed learning; (4)
Socratic questioning reinforces classroom instruction; and (5) AI-mediated
Socratic tools show promise for scalable, individualized learning. Effective
questioning commonly elicited prior knowledge, probed assumptions, scaffolded
conceptual links, and encouraged reflection.
CONCLUSION
Integrating Socratic principles into a RAG-augmented GPT model produced a
learning tool that fosters deeper reasoning, conceptual understanding, and
engagement. The adaptive questioning helped learners identify gaps, clarify
assumptions, and reinforce classroom instruction while mitigating faculty
time constraints and variability in teaching quality. Although depth is
prioritized over breadth, the tool complements existing curricula and offers
a scalable approach to pre-clerkship medical education. Future work should
evaluate its impact on learner outcomes and expand its application across
broader topics.
Effective interpretation of complex medical topics, such as pulmonary
physiology, is challenging for pre-clerkship students due to limited
classroom exposure and static explanations provided by standard tools. The
Socratic method, characterized by structured questioning that encourages
learners to articulate reasoning and examine assumptions, is a proven
pedagogical strategy. This study systematically reviews its application in
pre-clerkship education and examines its integration into our Pulmonary
Function Test (PFT) learning assistant, an augmented GPT designed to enhance
evidence-based reasoning.
METHODS
Following PRISMA 2020 guidelines, we systematically searched for studies on
Socratic questioning, guided inquiry, or structured dialogue in pre-clerkship
education. Of 71 records screened, 38 full texts were reviewed and 29 met
inclusion criteria. Narrative synthesis identified common instructional
themes. Our PFT learning assistant applies a Retrieval-Augmented Generation
framework to combine curated scientific literature with adaptive Socratic
questioning to reinforce understanding and promote critical reasoning.
RESULTS
Five major themes emerged: (1) Socratic questioning enhances critical
reasoning and internalization of learning; (2) AI-supported Socratic methods
provide personalized feedback and assess understanding; (3) Socratic
strategies promote active participation and self-directed learning; (4)
Socratic questioning reinforces classroom instruction; and (5) AI-mediated
Socratic tools show promise for scalable, individualized learning. Effective
questioning commonly elicited prior knowledge, probed assumptions, scaffolded
conceptual links, and encouraged reflection.
CONCLUSION
Integrating Socratic principles into a RAG-augmented GPT model produced a
learning tool that fosters deeper reasoning, conceptual understanding, and
engagement. The adaptive questioning helped learners identify gaps, clarify
assumptions, and reinforce classroom instruction while mitigating faculty
time constraints and variability in teaching quality. Although depth is
prioritized over breadth, the tool complements existing curricula and offers
a scalable approach to pre-clerkship medical education. Future work should
evaluate its impact on learner outcomes and expand its application across
broader topics.
Presentation Tag(s)
Student Presentation