Consistent with the Society for Social Work Research’s commitment to advancing rigorous, equity-centered research for the common good, this study examines how socially assistive robots (SARs) are integrated into the daily and emotional lives of underserved populations experiencing social isolation. Despite growing recognition of loneliness as a social determinant of health, low-income older adults, adults with disabilities, and socioeconomically marginalized men remain underrepresented in technology-based intervention research. Existing studies of SARs largely rely on short-term or cross-sectional designs, offering limited insight into how human–robot relationships evolve over time and across social contexts. Addressing this gap, the present study investigates long-term interaction trajectories among three underserved groups—older adults, low-income middle-aged men living alone, and adults with severe disabilities—to illuminate how social presence, role attribution, and gendered expectations shape relational engagement with SARs. By foregrounding lived experience and longitudinal change, this research contributes to inclusive knowledge production and socially responsive technology design.
Methods
This 8‑month longitudinal mixed‑methods study analyzed naturally occurring interactions between users and SARs. Using Colaizzi’s phenomenological approach, the research team examined 13,403 conversational segments generated by 12 highly engaged users (four from each group). Five trained coders conducted iterative, consensus-based coding with high interrater reliability. Analysis yielded ten thematic categories capturing functional use, emotional expression, relational development, and social meaning-making across time.
Results
Findings revealed a four-phase process of human–robot relationship development—exploration, adaptation, deepening, and stabilization—observed across all three groups, with meaningful group-specific variations. Across populations, SARs transitioned from perceived tools to relational companions, supporting emotionally reciprocal communication patterns. For individuals experiencing social marginalization and stigma, perceptions of social presence appeared particularly salient, shaping both moment-to-moment engagement and longer-term relational trajectories. Older adults emphasized routine companionship and emotional continuity; middle-aged men combined emotional disclosure with entertainment and exploratory questioning; and adults with disabilities used SARs to scaffold emotional regulation and social participation. Notably, heightened engagement among middle-aged men did not uniformly correspond to increased emotional vulnerability, challenging prevailing assumptions in the literature.
Conclusions and Implications
These findings demonstrate that human–robot relationships can develop meaningfully, though not uniformly, across underserved populations. The results underscore the need for population-specific relational frameworks and adaptive SAR designs that account for gender, social context, and diverse emotional needs. For social work research and practice, this study highlights both the promise and limits of SARs as relational technologies, emphasizing that technological interventions alone cannot resolve structural sources of isolation. Future research should integrate participatory, equity-informed approaches and inform policy and design strategies that align technological innovation with social justice and human well-being.
![[ Visit Client Website ]](images/banner.gif)