From Social Robotics to Ecological Cognitive Care: An Enaction-Based Umbrella Review on Neurocognitive Disorders

Background: As ageing populations grow, the prevalence of dementia and pre-dementia conditions is rising. Emerging approaches to neurorehabilitation emphasize not only performance-based outcomes but also holistic, experiential, and person-centred aspects of care. The extended mind thesis further highlights the potential role of external tools in supporting impaired cognitive functions. Within this ecological and experiential perspective, Social Assistive Robotics (SAR) may offer a multidimensional approach to address cognitive, emotional, and social needs in neurocognitive disorders. Objective: To synthesize current evidence on the effects of robotic interventions within an enactive framework integrating mind, body, environment, and technology. Methods: A systematic search was conducted in PubMed, Ovid Medline, Scopus, ScienceDirect, Springer, Wiley, IEEE Xplore, ACM Digital Library, and the Cochrane Library. Due to heterogeneity among included studies, an umbrella review was performed using vote-counting by direction of effect as a non-quantitative synthesis method. Methodological rigour followed JBI and Cochrane guidelines. Results: Sixteen reviews were included. The strongest and most consistent benefits emerged for affective outcomes, particularly emotional response and social interaction p = 0.007 (two-sided). Conversely, outcomes related to cognition, anxiety, agitation, depression, and quality of life showed mixed or non-significant effects, while neuropsychiatric symptoms demonstrated no benefit. Conclusions: Discrepancies across reviews seem driven by methodological limitations in primary studies, limiting interpretability. The strength of this umbrella review lies in identifying systematic gaps that can guide future research. With stronger evidence, integrating SAR into experiential neurorehabilitation may offer a promising avenue for holistic, ecologically grounded care that extends beyond traditional task-based performance. Trial Registration: PROSPERO CRD420251165419.