JMIR Publications

ABSTRACT

Background:

Dementia and mild cognitive impairment (MCI) are progressive neurocognitive disorders associated with a decline in cognitive abilities, subsequently affecting daily functioning and quality of life. Pharmacological treatments have limited efficacy, highlighting the need for effective non-pharmacological interventions. Advances in digital technologies, including virtual reality (VR), extended/mixed reality (XR/MR), web-based platforms, and AI-assisted tools, offer promising opportunities for personalized, engaging, and scalable cognitive interventions.

Objective:

This systematic review aimed to (1) identify the characteristics of technology-assisted cognitive interventions for people with dementia (PwD) and MCI, (2) evaluate their effectiveness on cognitive, functional, and psychological outcomes, such as mood, depressive and anxiety symptoms, motivation, engagement, and quality of life, and (3) explore factors that may influence intervention outcomes, including intervention modality, cognitive stage, personalization, and integration of cognitive-physical components.

Methods:

A systematic search was conducted across five digital libraries (ACM, PubMed, IEEE Xplore, Sage, ScienceDirect), selected to ensure broad coverage across computer science, engineering, and health-related disciplines, for studies published between 2018 and 2025 , a period chosen to capture recent advances in rapidly evolving digital technologies and ensure the relevance of findings to current research and practice. Nineteen studies meeting the inclusion criteria were analyzed using a mixed quantitative and qualitative approach. Data were extracted on study characteristics, participant demographics, intervention modalities, cognitive domains targeted, duration, personalization/adaptive features, and outcomes. Quantitative synthesis examined effect sizes and statistical trends, while qualitative synthesis addressed non-cognitive outcomes, ecological validity, feasibility, and user experience. Cross-study comparisons identified patterns and potential moderators of intervention efficacy.

Results:

Among the 19 included studies, interventions employed VR (n=8), XR/MR (n=3), web-based or tablet-based platforms (n=5), and other innovative technologies (n=3). Most studies targeted memory (n=10), attention (n=7), and executive functions (n=6), with several addressing activities of daily living (n=4) and emotional well-being (n=3). Sixteen studies reported significant improvements in at least one cognitive or functional outcome, with immersive VR and combined cognitive-physical interventions showing the strongest evidence. Non-cognitive benefits included enhanced mood, motivation, engagement, and quality of life. Personalized and adaptive features were associated with greater user adherence and efficacy. Trends suggested that cognitive stage, intervention modality, and multimodal integration may influence outcomes, although systematic quantitative analyses of moderators were limited.

Conclusions:

Technology-assisted interventions, particularly immersive and personalized approaches, are effective in improving cognitive, functional, and psychological outcomes in people with dementia and MCI. VR-based and multimodal cognitive-physical programs appear especially promising. Future research should focus on larger, rigorously-designed trials with standardized outcome measures, longitudinal follow-ups, and formal evaluation of moderators to optimize intervention personalization and scalability. These findings support the integration of digital interventions as viable components of cognitive rehabilitation and healthy aging strategies.