Number
819
Name
Integrating Generative AI in Collaborative Learning: A Novel Approach to Neuroscience Education for Medical Students
Date & Time
Monday, June 8, 2026, 6:00 PM - 7:30 PM
Location Name
Oglethorpe Ballroom
Speakers
Authors
Amanda Chase, PhD, Xavier Ochsner College of Medicine
Katelyn Carnevale, PhD, Boston University Chobanian & Avedisian School of Medicine
Samiksha Prasad, PhD, Nova Southeastern University College of Allopathic Medicine
Presentation Topic(s)
Technology and Innovation
Description
Purpose
Traditional didactic approaches in medical education can fail to develop
competencies in AI literacy, collaborative learning, and clinical reasoning.
To address these gaps, we implemented a collaborative learning project that
enabled medical students to apply AI tools in the development of peer
learning resources.
Methods
We implemented a team-based project within a pre-clerkship neuroscience
course in which learners (n=61) worked in small groups to apply generative
artificial intelligence (AI) for development of peer learning resources. Each
group was assigned a learning objective in neuroanatomy, pathophysiology, or
clinical presentation of disease and, over one week, researched its topic and
used AI to develop (1) a novel teaching tool (mnemonic, diagram, or visual
aid) and (2) an NBME-style clinical reasoning question. Teams documented
their prompt engineering process, iteratively refined AI-generated materials,
verified accuracy, and created detailed rationales. The project culminated in
interactive presentations demonstrating their teaching tool, presenting their
NBME question, and explaining their AI evaluation process.
Results
Final team presentations (n=8) were evaluated qualitatively for content.
Common teaching tools included sketchy diagrams and mnemonics tailored to
neuroanatomy and disease mechanisms. Teams demonstrated iterative prompt
refinement, with an average of 4–6 revisions per resource to improve clarity
and accuracy. Accuracy verification was emphasized, and most groups reported
using board-prep resources to validate AI-generated content. Peer evaluations
and self-reflections were evaluated qualitatively across five domains.
Feedback indicated high engagement and perceived value of the activity.
Conclusions
This intervention demonstrates that integrating generative AI into
collaborative learning projects can effectively develop multiple competencies
simultaneously. The requirement to document and present the AI refinement
process promoted metacognitive reflection. This scalable model provides a
framework for other medical schools seeking to integrate AI literacy into
existing coursework to enhance competency-based outcomes.
Traditional didactic approaches in medical education can fail to develop
competencies in AI literacy, collaborative learning, and clinical reasoning.
To address these gaps, we implemented a collaborative learning project that
enabled medical students to apply AI tools in the development of peer
learning resources.
Methods
We implemented a team-based project within a pre-clerkship neuroscience
course in which learners (n=61) worked in small groups to apply generative
artificial intelligence (AI) for development of peer learning resources. Each
group was assigned a learning objective in neuroanatomy, pathophysiology, or
clinical presentation of disease and, over one week, researched its topic and
used AI to develop (1) a novel teaching tool (mnemonic, diagram, or visual
aid) and (2) an NBME-style clinical reasoning question. Teams documented
their prompt engineering process, iteratively refined AI-generated materials,
verified accuracy, and created detailed rationales. The project culminated in
interactive presentations demonstrating their teaching tool, presenting their
NBME question, and explaining their AI evaluation process.
Results
Final team presentations (n=8) were evaluated qualitatively for content.
Common teaching tools included sketchy diagrams and mnemonics tailored to
neuroanatomy and disease mechanisms. Teams demonstrated iterative prompt
refinement, with an average of 4–6 revisions per resource to improve clarity
and accuracy. Accuracy verification was emphasized, and most groups reported
using board-prep resources to validate AI-generated content. Peer evaluations
and self-reflections were evaluated qualitatively across five domains.
Feedback indicated high engagement and perceived value of the activity.
Conclusions
This intervention demonstrates that integrating generative AI into
collaborative learning projects can effectively develop multiple competencies
simultaneously. The requirement to document and present the AI refinement
process promoted metacognitive reflection. This scalable model provides a
framework for other medical schools seeking to integrate AI literacy into
existing coursework to enhance competency-based outcomes.