Work

Non‐Immersive Virtual Reality Assessment of Functional Abilities in Older Adults is Less Influenced by Demographic Features than Traditional Cognitive Test Scores

Year: 2025

Type: article

Abstract: BACKGROUND: Cognitive tests are crucial for clinical assessment of dementia, but are biased by education, sex, and race. We examined bias in a new measure of everyday cognition called the Virtual Kitchen Challenge (VKC), which requires participants to select objects and sequence task steps on a computer touchscreen to achieve highly familiar task goals (e.g., make breakfast/lunch). We hypothesized that because of its practical nature and high familiarity, scores from the VKC will not show demographic biases relative to conventional cognitive scores. METHOD: 133 older adults (Mage = 73.40±6.53, range = 61-98; 75% healthy cognition, 57% White, 36% Black, 63% female; education: M = 16.03±2.73) completed the VKC and cognitive tests from the Modified Preclinical Alzheimer's Cognitive Composite (mPACC; MMSE, HVLT-D, BVMT-D, Trail Making Test-B, Animal Fluency). VKC scores (number of screen touches, percent of steps accomplished, time spent touching distractor objects, percent of time off-screen) and a self-report questionnaire (Functional Activity Questionnaire [FAQ]) were used to evaluate functional performance. Due to the small sample sizes of some racial groups, race analyses included only White and Black categories. RESULT: Better performance on the VKC measures was significantly associated with higher mPACC scores and lower FAQ rating (all p's < .05). Black older adults performed significantly worse on mPACC tests (MMSE: p = .021, d = -.36; HVLT-D: p < .001, d = -.52; BVMT-D: p = < .001, d=-.65; Trails B: p = .004, d = .44; Animal Fluency: p = .004, d=-.44) compared to White older adults. These two groups did not significantly differ on any VKC measure (total touches p = .230; % accomplished p = .921; time with distractor objects p = .269; % time off-screen p = .424). No significant sex differences were found in VKC or mPACC variables (all p's > .05). Education was significantly correlated with all mPACC scores (p's < .05) but not VKC (all p's > .05). CONCLUSION: The VKC is not influenced by demographic features compared to traditional cognitive testing, underlining its potential importance as a more accurate assessment tool for cognitive decline. For this reason, it should be considered for clinical staging or as an outcome in clinical trials.

Source: Alzheimer s & Dementia

DOI: 10.1002/alz70857_105197