Virtual Reality for the Assessment of Visual Capacities: A Systematic Review

Abstract

Objectives: Virtual reality (VR) enables controlled, interactive, and ecologically valid visual tasks, offering potential advantages over static clinical tests. This systematic review synthesized current evidence on the use of VR technologies for assessing visual capacities across ocular, neurological, and low-vision populations, with particular attention to technological approaches, clinical domains, methodological quality, safety, and future translational needs. Methods: Following PRISMA 2020, PubMed, Scopus, and Web of Science were searched from inception to 2 August 2025, using database-specific strategies for virtual reality and visual function. Three reviewers independently screened reports, extracted data, and appraised completed studies using design-appropriate NIH Study Quality Assessment Tools or Cochrane Risk of Bias 2 Owing to methodological heterogeneity, findings were synthesized narratively. The review was not registered and no protocol was prepared. Results: Of 364 records, 36 completed studies met the eligibility criteria and were included in the narrative synthesis (publication years 2014–2024); one additional randomized-trial protocol was catalogued separately as ongoing evidence. Most completed studies employed immersive head-mounted displays, while others used semi-immersive or desktop VR. Assessed domains included visual fields, visual acuity, contrast sensitivity, oculomotor and vergence function, postural control, functional navigation, spatial orientation, and obstacle avoidance. The most mature evidence concerned VR-based perimetry and functional visual assessment in glaucoma and low-vision populations. Reported adverse events were uncommon and mild, mainly transient cybersickness, dizziness, or visual discomfort. Among completed studies, methodological appraisal was favorable for 10, intermediate for 25, and unfavorable for 1. Overall, evidence supports feasibility, user acceptability, and convergent validity versus selected conventional tools, but substantial heterogeneity remains in hardware, calibration, exposure duration, outcome metrics, and follow-up. Conclusions: This systematic review highlights the growing role of VR in assessment, showing potential for controlled, ecologically oriented evaluations across domains like perimetry and functional vision. While promising, VR is currently a complementary framework rather than a replacement for established clinical tools, requiring stronger validation, repeatability testing, and calibration standardization for routine implementation.