Abstract
(1) Background: Immersive virtual reality (IVR) has emerged as a promising non-pharmacological intervention to promote relaxation and improve emotional well-being in this population. (2) Methods: This crossover study evaluated the effects of IVR on anxiety and psychological well-being in a sample of eight participants with mild dementia attending a day-care center. Participants underwent two conditions: an experimental condition involving relaxing nature-based VR scenarios (Nature Treks VR) and a control condition using personalized YouTube videos on a tablet. Each condition lasted 12 sessions. Assessments included heart rate (HR), the I-PANAS-SF, the reduced State-Trait Anxiety Inventory (STAI-r), behavioral observations, and a subjective response questionnaire. (3) Results: A significant reduction in HR over time was found during IVR exposure, suggesting a calming physiological effect not observed in the control condition. While changes in PANAS and STAI-r scores were not statistically significant, the PANAS score improvement in the experimental condition approached statistical significance (p = 0.054) and was just below the minimal clinically important difference (MCID), suggesting a potentially meaningful trend. Behavioral responses were higher during YouTube sessions, likely due to personalized content. All participants rated the IVR experience positively on the subjective questionnaire, indicating high acceptability, though social desirability bias cannot be excluded. (4) Conclusions: IVR appears to be a feasible and acceptable intervention for individuals with dementia, warranting further investigation.