This thesis explores how autobiographical memory, cognition, and wellbeing can be supported through embodied, social, and person-centred activities in ageing and dementia. It examines the combined effects of walking and reminiscence across neural, behavioural, and community levels. Using a mixed-methods design, the research progresses in line with the Medical Research Council (MRC) framework for complex interventions, moving from intervention development and mechanistic investigation, through feasibility and piloting, to evaluation and implementation considerations. Throughout all phases, Patient and Public Involvement (PPI) was embedded via structured workshops and advisory meetings with people living with and without dementia, and relatives, who contributed directly to the co-design of walking routes, refinement of assessment procedures, ethical considerations, fMRI session design, and implementation planning, ensuring the intervention was dementia-inclusive, feasible, and responsive to lived experience. The first study used structural–functional magnetic resonance imaging to investigate autobiographical memory networks in older adults. Distinct activation patterns within posterior–medial and anterior–temporal regions varied across recent and remote memories, indicating flexible neural reorganisation with age and informing the development of autobiographical cueing strategies later used in the intervention. A development study (n = 2) was conducted with people living with dementia to refine the structure and delivery of a walking-and-reminiscence intervention. Pre–post descriptive changes were observed across autobiographical memory levels, with the most notable shifts in later-life and recent event recall, alongside smaller changes in personal semantic recall and selected cognitive and wellbeing measures, serving primarily to refine feasibility and delivery procedures rather than to test effectiveness in this small sample. A larger mixed-methods developmental study compared people living with dementia (n = 5) and cognitively healthy older adults (n = 5) following a four-week combined walking and reminiscence intervention. While most quantitative changes were modest and did not reach statistical significance, participants living with dementia demonstrated descriptive increases in episodic recall, particularly for late-adulthood and recent life epochs, whereas cognitively healthy participants showed relatively stable episodic performance alongside small increases in semantic autobiographical recall across selected life periods. Global cognition and wellbeing measures remained largely stable in both groups. Qualitative findings across groups highlighted strengthened emotional engagement, shared community narratives, and increasing coherence of personal storytelling over time. Building on this, a three-arm randomised intervention (N= 36; n = 12 per condition) disentangled walking-only, reminiscence-only, and combined conditions among community-dwelling older adults. Significant improvements in quality of life were observed over time, with the combined Walking + Reminiscence group reporting higher overall CASP-19 scores (Group p = .009; Time p = .010). Semantic autobiographical recall was consistently higher in the combined condition across early and middle adulthood epochs, while episodic recall remained stable across groups. Qualitatively, walking elicited sensory-rich and affectively positive memories, reminiscence fostered structured and culturally inspired narratives, and their combination integrated embodied retrieval with shared meaning and renewed narrative confidence. Finally, a survey involving health policymakers and older adults living with and without dementia (N = 65; 28 cognitively healthy older adults, 16 people living with dementia, 21 health policymakers) confirmed the perceived feasibility and acceptability value of the approach, while identifying practical barriers such as time, staffing, and integration within care systems and highlighting considerations for future implementation and scale-up. Together, these studies show that memory is an active and relational process strengthened through movement, meaning, and connection. The thesis concludes that walking and reminiscence are possible to implement in Ireland and internationally, with scope for cultural adaptation and tailoring to local needs, offering a simple, person-centred way to sustain cognitive and human connection in ageing and dementia.