Medical Science Educator Education

Non-Interactive Virtual Avatar–Delivered Structured Instruction Versus Self-Directed Virtual Reality Anatomy Learning: A Comparative Study Among Medical Students

Šlebodová, Miriama, Hvizdošová, Natália, Majerník, Jaroslav, Hodorová, Ingrid, Tkáč, Michal, Šandor, Filip, Mäsiar, Aleš, Havránek, Tomáš, Brngál, Tomáš
Springer Nature · 2026-12-01 · DOI 10.1007/s40670-026-02927-w

Abstract

Instructional design may influence learning outcomes within the same immersive VR environment. Avatar-delivered structured instruction was associated with higher immediate knowledge scores than self-directed VR exploration. The effects of the non-interactive avatar and structured instructional features could not be distinguished. Introduction Virtual reality (VR) is increasingly used in anatomy education to support spatial visualisation and interactive learning. However, evidence on the most effective instructional approaches within VR environments remains limited. This study aimed to compare knowledge acquisition and student perceptions between avatar-guided structured instruction and self-directed VR anatomy learning. Methods This prospective comparative study included 186 first-year medical students from the Faculty of Medicine, Pavol Jozef Šafárik University in Košice. Thirteen pre-existing practical class groups were randomly assigned to a structured instruction group administered by non-interactive virtual avatar (hereafter avatar-guided group) (six groups; n = 91) or self-directed learning (seven groups; n = 95). Both groups studied the urinary system using the Human Anatomy VR for Institutions application and had access to the same anatomical content. The avatar-guided group completed a predefined, educator-prepared lesson delivered by a non-interactive virtual avatar using synchronised narration, visual highlighting, and instructional prompts, whereas the self-directed group explored the content independently. Knowledge acquisition was assessed using a 10-item multiple-choice test, and student perceptions were evaluated using a five-point Likert-scale questionnaire. Knowledge-test scores were compared using Welch’s independent-samples t-test, while questionnaire items were analysed using the Mann–Whitney U test. Results The students in the avatar-guided group achieved higher knowledge-test scores than those of the self-directed group (7.57 ± 1.80 vs. 6.36 ± 2.27; p < 0.001; Cohen’s d = 0.59). The responses to the questionnaire were obtained from 181 students. The avatar-guided group reported more positive perceptions of learning effectiveness, confidence in identifying anatomical structures, spatial understanding, and clarity of content (all p ≤ 0.003; effect sizes r = 0.22 to 0.31). Conclusion Structured instruction delivered by a non-interactive virtual avatar was associated with higher immediate knowledge-test scores and more positive student perceptions than self-directed exploration of the same VR content without a prescribed sequence, narration, visual highlighting, or instructional prompts. Because these features formed a combined intervention, their individual contributions cannot be determined.

Read at publisher →