Virtual reality anatomy learning shows no clear knowledge edge in Zambian trial

Standalone virtual reality has been promoted as a way to let medical students explore three-dimensional anatomy interactively, but rigorous comparative evidence from resource-constrained medical-education settings has remained scarce. A new four-arm randomized exploratory study in Zambia set out to test how a head-mounted VR anatomy session stacked up against three other learning resources when first- and second-year Bachelor of Science in Clinical Sciences students had a single self-paced session of up to 30 minutes. The results, published open access in Advances in Simulation, temper expectations: VR did not outperform the alternatives on immediate knowledge, and its acceptance advantages were limited once statistical adjustments were applied.

The trial enrolled 67 students and assigned them using Excel-generated random group numbers to one of four conditions: standalone VR running the 3D Organon platform, the AMBOSS clinical learning platform, curated textbook materials, or self-directed internet search. There was no allocation concealment. Knowledge was measured with a 10-item pre-post test, and the authors were candid that this instrument was unvalidated. Immediate acceptance was measured with an adapted Technology Acceptance Model questionnaire.

Because not every enrolled participant completed every measurement, the primary knowledge analysis included 59 students: 12 in the VR group, 17 using AMBOSS, 16 with the textbook, and 14 in the internet-search group. The primary between-group knowledge analysis used baseline-adjusted analysis of covariance with HC3-robust standard errors. Before the study, the team reviewed whether the tested concepts could be located in the three fixed-content resources. That review did not establish equivalence in depth, organisation, or accessibility, and could not be applied to the self-directed internet group.

The headline statistical result was a null finding. The primary analysis did not detect an overall difference in baseline-adjusted post-test scores between the four modalities, F(3,54) = 1.945, p = .133. None of the six pairwise comparisons reached significance after Holm adjustment for multiple testing. Within groups, score increases crossed the adjusted significance threshold for AMBOSS, the textbook, and the web-resource conditions, but not for VR, where the adjusted p value was .053, just shy of the threshold. This finding does not establish equivalence.

Acceptance data told a similarly restrained story. Analyses covered 55 to 60 participants per construct, and after adjustment across six domains of the Technology Acceptance Model questionnaire, only one domain differed significantly between modalities: Attitude, with H(3) = 11.801, ε² = 0.17, adjusted p = .049. AMBOSS was rated more favourably than the self-directed internet search (adjusted p = .019, Cliff’s δ = −0.60). Notably, VR did not emerge as the standout in any adjusted quantitative acceptance comparison. Two constructs, perceived ease of use and perceived enjoyment, showed low internal consistency in the questionnaire data, which the authors say limits how much weight their results can carry.

Participants contributed 211 open-text comments, and 64 people supplied at least one. Brief responses underwent qualitative content analysis. Some VR participants valued visualisation, novelty, and enjoyment but reported navigation and content-access difficulties. AMBOSS participants frequently valued organisation and accessibility, while experiences with textbook and web resources were mixed.

Taken together, the authors conclude that this small exploratory study did not provide conclusive evidence of differences in immediate knowledge outcomes between the four modalities, and that quantitative acceptance differences were also limited after adjustment for multiple comparisons. Rather than crowning a winner, the trial’s value lies in what it exposes about study design and implementation in low-resource contexts: content accessibility, curricular alignment, interface familiarization, and exposure time all surfaced as priorities that larger, prospectively designed studies should address, ideally using validated instruments and tracking longer-term outcomes beyond a single session.

Subject of Research: Comparing virtual reality with other learning resources for anatomy education among Zambian clinical sciences students

Article Title: Beyond knowledge gains – learner experience and technology acceptance of virtual reality for anatomy education in Zambia: a randomized exploratory study

Article References: Barteit, S., El Ahmadie, N., Nieder, J., & Andreadis, P. (2026). Beyond knowledge gains – learner experience and technology acceptance of virtual reality for anatomy education in Zambia: a randomized exploratory study. Advances in Simulation, 11(1), Article 81. https://doi.org/10.1186/s41077-026-00493-0

Image Credits: AI Generated

DOI: 10.1186/s41077-026-00493-0

Keywords: virtual reality, anatomy education, Zambia, randomized study, technology acceptance, medical education, 3D Organon, AMBOSS, knowledge outcomes, learning resources, exploratory study, clinical sciences students