Number
223
Name
AI-Driven Self-Directed Learning Activities in Undergraduate Medical Education
Date & Time
Monday, June 8, 2026, 6:00 PM - 7:30 PM
Location Name
Oglethorpe Ballroom
Speakers
Authors
Anitha Shenoy, Texas A & M Health Science Center
Steven Maxwell, Texas A & M Health Science Center
Dustin Dubois, Texas A & M Health Science Center
Kristy Motte, Texas A & M Health Science Center
Bridget Lighthall, Texas A & M Health Science Center
Presentation Topic(s)
Curriculum
Description
PURPOSE:
This study examines the impact of a novel self-directed learning (SDL)
activity that integrates artificial intelligence (AI) into the first-year
medical curriculum within the foundational medical sciences (FOM II) course.
The activity aims to improve critical thinking, information-seeking skills,
ethical AI use, and evidence-based reasoning through USMLE/NBME style
AI-assisted multiple-choice question (MCQ) creation and peer/faculty feedback.
METHODS:
Students created MCQs on challenging course topics using AI, providing a
clear rationale and documenting their information-seeking process, including
database selection, source trustworthiness, and relevance assessment. Peers
evaluated question quality, fairness, and impartiality, while congruent with
the LCME definition of SDL, faculty assessed source credibility and information-seeking
skills using standardized rubrics. Students revised and resubmitted
questions, explicitly addressing feedback and reflecting on their SDL
engagement. Additionally, students were asked to reflect on their
self-directed learning experience, with these reflections submitted as part
of the activity and analyzed for themes related to SDL skills. The activity
contributes 1% to the overall course grade.
RESULTS:
Preliminary findings for 2029 cohort from Exam 1 show a decrease in student
failures from 10% to 7%, and an increase in B grades from 38% to 43%,
compared to the previous year’s (2028 cohort) examination results, suggesting
improved student understanding and learning outcomes. Additional findings
will discuss reflections of students on SDL and how the activity impacted
their approach to learning.
CONCLUSION:
Early results indicate that this AI-enhanced SDL activity may support
stronger learning outcomes, better critical appraisal skills, and more
responsible use of AI among medical students. Continued evaluation will help
determine the overall impact of this approach. Given its simplicity and
adaptability, this instructional strategy is highly transferable and can be
easily implemented across pre-clerkship, clerkship, and graduate medical
education settings. Future plans include adapting and expanding this model to
additional courses to broaden its educational reach.
This study examines the impact of a novel self-directed learning (SDL)
activity that integrates artificial intelligence (AI) into the first-year
medical curriculum within the foundational medical sciences (FOM II) course.
The activity aims to improve critical thinking, information-seeking skills,
ethical AI use, and evidence-based reasoning through USMLE/NBME style
AI-assisted multiple-choice question (MCQ) creation and peer/faculty feedback.
METHODS:
Students created MCQs on challenging course topics using AI, providing a
clear rationale and documenting their information-seeking process, including
database selection, source trustworthiness, and relevance assessment. Peers
evaluated question quality, fairness, and impartiality, while congruent with
the LCME definition of SDL, faculty assessed source credibility and information-seeking
skills using standardized rubrics. Students revised and resubmitted
questions, explicitly addressing feedback and reflecting on their SDL
engagement. Additionally, students were asked to reflect on their
self-directed learning experience, with these reflections submitted as part
of the activity and analyzed for themes related to SDL skills. The activity
contributes 1% to the overall course grade.
RESULTS:
Preliminary findings for 2029 cohort from Exam 1 show a decrease in student
failures from 10% to 7%, and an increase in B grades from 38% to 43%,
compared to the previous year’s (2028 cohort) examination results, suggesting
improved student understanding and learning outcomes. Additional findings
will discuss reflections of students on SDL and how the activity impacted
their approach to learning.
CONCLUSION:
Early results indicate that this AI-enhanced SDL activity may support
stronger learning outcomes, better critical appraisal skills, and more
responsible use of AI among medical students. Continued evaluation will help
determine the overall impact of this approach. Given its simplicity and
adaptability, this instructional strategy is highly transferable and can be
easily implemented across pre-clerkship, clerkship, and graduate medical
education settings. Future plans include adapting and expanding this model to
additional courses to broaden its educational reach.