A Virtual Reality-Based Puzzle Game for Visuospatial Cognitive Training in Older Adults with Non-Amnestic Mild Cognitive Impairment: Protocol for a Single-Arm Feasibility Study (Preprint)

BACKGROUND Mental rotation, the ability to mentally transform visuospatial representations, supports everyday spatial behaviors (e.g., navigation) and can be vulnerable in later life. Older adults with mild cognitive impairment (MCI) often show greater difficulties in visuospatial processing than cognitively unimpaired peers, including lower accuracy and higher variability in mental rotation tasks. Because MCI represents a prodromal stage associated with elevated risk for subsequent dementia, the critical period occurring before rapid cognitive decline to dementia, it may be an important window for interventions that target specific cognitive vulnerabilities. In non-amnestic forms of MCI (MCI-NA), visuospatial and/or executive deficits can be prominent, and longitudinal outcomes are heterogeneous, varying in part by underlying neuropathology. Accordingly, interventions that are explicitly designed to engage visuospatial processes relevant to MCI-NA may be a useful, deficit-targeted approach to evaluate in feasibility studies and to inform future controlled trials for cognitive training programs aiming to prolong daily functioning and reduce suffering. OBJECTIVE This study aims to evaluate the feasibility and acceptability of the Virtual Reality-Visuospatial Cognitive Training (VR-VCT) program in older adults with MCI-NA and to estimate preliminary within-subject changes in visuospatial cognition to inform a future randomized trial. Participants (n=40) will meet eligibility criteria consistent with commonly used definitions of MCI-NA, including subjective cognitive concerns, preserved basic activities of daily living, absence of dementia, and objective impairment on standardized measures emphasizing visuospatial and/or executive functioning. This study aims to: (1) quantify the feasibility and acceptability of VR-VCT in older adults with MCI-NA, and (2) estimate preliminary within-subject change on visuospatial cognitive outcomes following VR-VCT. METHODS 40 Participants with MCI-NA will be enrolled in a structured VR Cubism program using the Meta Quest 3 VR headset. The intervention will involve three 30-minute sessions per week for 12 weeks, with tasks progressing in difficulty over time. Cognitive and visuospatial outcomes will be assessed at baseline (T0), immediately post-intervention (T1), and at follow-up (T2; 12 weeks post-intervention) to evaluate whether observed changes are maintained. Global cognition will be assessed using the Montreal Cognitive Assessment (MoCA). Visuospatial construction will be assessed using the Wechsler Adult Intelligence Scale (WAIS) Block Design subtest, and mental rotation will be assessed using the Vandenberg Mental Rotation Test (VMRT). Changes in performance across time points will be analyzed using repeated-measures models (e.g., linear mixed models) to estimate within-subject change, with effect sizes and confidence intervals reported to inform future controlled trials. RESULTS Participants will be recruited from local assisted living facilities, memory care settings, and community outreach programs. This study has been approved by the University of Utah School of Medicine Institutional Review Board. The data collection will be started in March 2026. Data analysis is anticipated to be concluded by August 2026. CONCLUSIONS The findings will inform the study design, outcome measurement, and power calculations of a future randomized controlled trial. If feasible and acceptable, VR-VCT may represent a scalable, engaging, and deficit-targeted intervention approach with the potential to support visuospatial cognitive functioning during a critical window prior to dementia onset.