Number
214
Name
Retrieval-Based Learning in Medical Education: Effects of Spaced Repetition and Question-Based Practice on CBSE Scores
Date & Time
Monday, June 8, 2026, 6:00 PM - 7:30 PM
Location Name
Oglethorpe Ballroom
Speakers
Authors
Arthur Yu, Medical College of Georgia
Yash Patel, Medical College of Georgia
Shilpa Brown, Medical College of Georgia
Lynne Coule, Medical College of Georgia
Wanda Jirau-Rosaly, Medical College of Georgia
Nicole Rockich-Winston, Medical College of Georgia
Presentation Topic(s)
Curriculum
Description
PURPOSE
This study examined how medical student use of Anki, a spaced-repetition
flashcard platform, and practice questions influences NBME Comprehensive
Basic Science Exam (CBSE) scores. A secondary objective was to evaluate
whether total Anki card review volume correlates with CBSE performance. Since
USMLE Step 1 transitioned to pass/fail, identifying effective self-directed
learning approaches has become increasingly important in medical curricula.
METHODS
Second-year medical students at the Medical College of Georgia completed an
anonymous survey reporting their Anki use, number of practice questions
completed, and CBSE scores after one year of preclerkship instruction.
Students reported perceived test effort, allowing exclusion of low-effort
responses. Mean CBSE scores were compared between Anki users and non-users
using a two-tailed t-test with unequal variances. Linear regression assessed
the association between review volume and CBSE performance. Differences in
CBSE performance across levels of practice-question volume were evaluated
using one-way ANOVA.
RESULTS
Sixty-eight students were included in the final analysis. Students who used
Anki (n = 53) had higher mean CBSE scores (56.52%) than non-users (n = 15;
51.66%), though this difference was not statistically significant (p = 0.08).
Among Anki users, total reviews over the prior year were significantly
associated with higher CBSE performance (F(1,53) = 5.04). Practice-question
volume was significantly associated with higher CBSE scores (p = 0.0006).
CONCLUSION
Although Anki use did not significantly predict higher CBSE scores, Anki
users showed higher average performance, and greater Anki review volume was
significantly correlated with higher scores. Completing more practice
questions was strongly associated with higher CBSE performance. These
findings suggest that consistent engagement with retrieval-based learning
tools may support stronger USMLE Step 1 preparedness within accelerated
curricula and the pass/fail environment. Limitations of this
single-institution study include response bias and reliance on self-reported
measures.
This study examined how medical student use of Anki, a spaced-repetition
flashcard platform, and practice questions influences NBME Comprehensive
Basic Science Exam (CBSE) scores. A secondary objective was to evaluate
whether total Anki card review volume correlates with CBSE performance. Since
USMLE Step 1 transitioned to pass/fail, identifying effective self-directed
learning approaches has become increasingly important in medical curricula.
METHODS
Second-year medical students at the Medical College of Georgia completed an
anonymous survey reporting their Anki use, number of practice questions
completed, and CBSE scores after one year of preclerkship instruction.
Students reported perceived test effort, allowing exclusion of low-effort
responses. Mean CBSE scores were compared between Anki users and non-users
using a two-tailed t-test with unequal variances. Linear regression assessed
the association between review volume and CBSE performance. Differences in
CBSE performance across levels of practice-question volume were evaluated
using one-way ANOVA.
RESULTS
Sixty-eight students were included in the final analysis. Students who used
Anki (n = 53) had higher mean CBSE scores (56.52%) than non-users (n = 15;
51.66%), though this difference was not statistically significant (p = 0.08).
Among Anki users, total reviews over the prior year were significantly
associated with higher CBSE performance (F(1,53) = 5.04). Practice-question
volume was significantly associated with higher CBSE scores (p = 0.0006).
CONCLUSION
Although Anki use did not significantly predict higher CBSE scores, Anki
users showed higher average performance, and greater Anki review volume was
significantly correlated with higher scores. Completing more practice
questions was strongly associated with higher CBSE performance. These
findings suggest that consistent engagement with retrieval-based learning
tools may support stronger USMLE Step 1 preparedness within accelerated
curricula and the pass/fail environment. Limitations of this
single-institution study include response bias and reliance on self-reported
measures.
Presentation Tag(s)
Student Presentation