Nature-based and technology-assisted exercise for cognitive and mobility outcomes in older adults: a systematic review of randomized trials

Population aging and rising rates of mild cognitive impairment and mood symptoms have intensified interest in low-cost strategies to preserve brain health and independence. Physical activity is an established protective factor, but adherence to exercise programs remains suboptimal. Emerging work suggests that the exercise environment, including natural settings and immersive or interactive formats, may confer benefits beyond conventional indoor training. We systematically searched major electronic databases for randomized trials in older adults that compared nature-based or outdoor exercise, or immersive and interactive modalities (virtual reality or exergaming), with broadly dose-matched indoor or usual-exercise control conditions. Eligible studies enrolled older participants or reported stratified results for those aged at least 60 years or with mild cognitive impairment. Primary outcomes were brain-related measures spanning executive function and global cognition, mood and perceived restoration, dual-task gait, autonomic and endocrine stress markers, and indices of brain activation or connectivity. Risk of bias was assessed with the Cochrane risk of bias 2 tool for randomized trials (RoB 2), and findings were synthesized narratively. Twelve randomized trials met the inclusion criteria. Technology-assisted interventions that embedded cognitive demands in movement tasks produced small-to-moderate improvements in executive function, working memory, and dual-task gait, particularly in cognitively vulnerable older adults, compared with conventional exercise. Nature-based or outdoor sessions consistently enhanced affective valence, enjoyment, and perceived restoration relative to indoor or urban comparators, whereas effects on mobility, heart-rate variability, cortisol, and other physiological markers were smaller and heterogeneous. Functional near-infrared spectroscopy data suggested more efficient prefrontal recruitment during walking after interactive or enriched exercise. Overall risk of bias was low to some concerns, with small samples, short follow-up, and inconsistent exposure reporting limiting certainty. The context in which older adults exercise appears to be a modifiable component of brain-focused training. Immersive and interactive formats may enhance executive control and dual-task gait and mobility, while nature-based sessions reliably support affective and restorative responses that may facilitate adherence. Future trials should more rigorously match physical activity dose, standardize reporting of environmental and immersion characteristics, extend follow-up, and include core outcome batteries across cognitive, affective, physiological, and neuroimaging domains to identify context features that yield clinically meaningful benefits.