EdTech & Innovation
Avatar-guided VR anatomy instruction yields higher scores in medical student trial
Šlebodová et alMedical Science Educator

In a prospective comparative study evaluating instructional approaches within virtual reality (VR) anatomy education, structured teaching delivered by a non-interactive virtual avatar demonstrated superior short-term knowledge acquisition and student satisfaction compared with unguided exploration.
The trial enrolled 186 first-year medical students across 13 practical class groups at Pavol Jozef Šafárik University in Košice. Groups were randomly allocated to either an avatar-guided structured instruction cohort (six groups; n = 91) or a self-directed learning cohort (seven groups; n = 95). Both cohorts utilized the Human Anatomy VR for Institutions application to study the urinary system using identical anatomical content. The avatar-guided cohort engaged with a predefined lesson featuring synchronized audio narration, visual structure highlighting, and sequential instructional prompts delivered by a non-interactive avatar, while the control cohort explored the virtual environment independently.
Post-intervention assessment via a 10-item multiple-choice test revealed that the avatar-guided cohort achieved significantly higher knowledge scores than the self-directed cohort (7.57 ± 1.80 vs. 6.36 ± 2.27, respectively; p < 0.001; Cohen’s d = 0.59). Additionally, analysis of five-point Likert-scale feedback from 181 responding students demonstrated that the avatar-guided group reported significantly superior ratings for perceived learning effectiveness, confidence in identifying anatomical structures, spatial comprehension, and overall content clarity (all p ≤ 0.003; effect sizes r = 0.22 to 0.31).
A noted study limitation is that the avatar intervention combined synchronized narration, visual highlighting, and instructional sequencing simultaneously, preventing determination of the relative contributions of each individual component.
Key Takeaways
- Students receiving avatar-guided structured instruction achieved significantly higher knowledge-test scores than those utilizing self-directed virtual reality learning (mean 7.57 ± 1.80 vs 6.36 ± 2.27; p < 0.001; Cohen's d = 0.59).
- Avatar-guided virtual instruction resulted in significantly higher student ratings for learning effectiveness, anatomical identification confidence, spatial understanding, and content clarity (all p ≤ 0.003; effect sizes r = 0.22 to 0.31).
Read the original sourceMedical Science Educator
