Abstract
Introduction: Dementia is increasingly prevalent globally. Existing questionnaire-based cognitive assessment tools may not comprehensively assess cognitive function and real-time task-performance across all cognitive domains. CAVIRE (Cognitive Assessment by VIrtual REality), a fully immersive virtual reality system incorporating automated audio-visual instructions and a scoring matrix was developed to assess the six cognitive domains, with potential to maintain consistency in execution of the testing environment and possibly time-saving in busy primary care practice. Aims: This is a feasibility study to compare the completion times of the questionnaire-based Montreal Cognitive Assessment (MoCA) and the CAVIRE in cognitively-healthy Asian adults aged between 35 and 74 years, overall, and in and across each 10-year age group (35-44; 45-54; 55-64; 65-74). Methods: A total of 100 participants with a MoCA score of 26 or more were recruited equally into the four 10-year age groups at a primary care clinic in Singapore. Completion time for the MoCA assessment for each participant was recorded. They were assessed using the CAVIRE, comprising 13 segments featuring common everyday activities assessing all six cognitive domains, and the completion time was also recorded through the embedded automated scoring and timing framework. Results: Completion time for CAVIRE as compared to MoCA was significantly ($p < 0.01$) shorter, overall (mean difference: 74.9 seconds) and in each age group. Younger, vs. older, participants completed both the MoCA and CAVIRE tasks in a shorter time. There was a greater variability in the completion time for the MoCA, most markedly in the oldest group, whereas completion time was less variable for the CAVIRE tasks in all age groups, with most consistency in the 45-54 year-age group. Conclusion: We demonstrate almost equivalent completion times for a VR and a questionnaire-based cognition assessment, with inter-age group variation in VR completion time synonymous to that in conventional screening methods. The CAVIRE has the potential to be an alternative screening modality for cognition in the primary care setting.
Introduction
Dementia prevalence is expected to triple globally by 2050, making early detection critical. However, current screening tools like the Montreal Cognitive Assessment (MoCA) and Mini-Mental State Examination (MMSE) face significant bottlenecks. They are labor-intensive, requiring trained healthcare professionals for administration and manual scoring. Furthermore, they are often literacy-dependent and abstract, utilizing "paper-and-pencil" tasks that may not accurately reflect a person's real-world functional abilities.
This study explores whether a fully immersive virtual reality (VR) system, CAVIRE (Cognitive Assessment by VIrtual REality), can provide a faster, more objective, and "ecologically valid" alternative for cognitive screening in high-pressure primary care environments.
Methods
The researchers recruited 100 cognitively healthy Asian adults (MoCA score $\ge$ 26) in Singapore, balanced across four 10-year age brackets (35–74 years). The study compared the completion times of the traditional MoCA against the CAVIRE system.
The CAVIRE System
CAVIRE utilizes an HTC VIVE Pro headset and Leap Motion controller for hand-tracking, allowing natural gesture-based interaction without handheld controllers. The assessment consists of 13 segments simulating daily activities, such as preparing breakfast or calculating change at a stall. These segments are designed to map onto the six major cognitive domains.
The system features automated audio-visual instructions and an embedded scoring/timing framework, eliminating the need for a human administrator to run the test.
Results
The study found that CAVIRE was significantly faster than MoCA across all age groups, saving an average of 74.9 seconds per patient ($p < 0.01$). Mean completion times were approximately 515 seconds for MoCA and 440 seconds for CAVIRE.
Timing Variability and Age
A key finding was the reduced variability in VR completion times compared to MoCA. While MoCA completion times varied widely, particularly in the oldest age group (65-74 years), CAVIRE times remained more consistent across all participants.
The data also revealed a significant "inflection point" around age 50. Younger participants (35–54) were consistently fast, while completion times increased sharply in the 55–64 and 65–74 groups for both assessments.
Discussion and Conclusion
The CAVIRE system demonstrates high feasibility for primary care screening. It offers "ecological validity" by testing functional cognition through simulated daily tasks rather than abstract exercises. The automation of the system serves as a force multiplier, potentially allowing non-clinical staff to oversee screenings and standardizing the testing environment.
While this feasibility study focused on cognitively healthy individuals, the results suggest that VR-based assessments can match or exceed the efficiency of traditional methods. Future work will involve testing CAVIRE on patients with Mild Cognitive Impairment (MCI) and dementia to further validate its time-saving benefits and diagnostic accuracy. Additionally, VR could eventually be used for longitudinal tracking—detecting subtle changes in a person's "virtual" daily routine over years. In conclusion, CAVIRE has the potential to be an effective alternative screening modality for cognition in the primary care setting.
Summary updated