Abstract
BACKGROUND: As global populations age, healthcare and social systems face mounting pressure to provide effective support for older adults. Social robots have emerged as promising tools to enhance companionship, cognitive engagement, and daily assistance. However, fear of robots among older adults remains a critical barrier to adoption. OBJECTIVE: This scoping review examined how fear manifests in human-robot interaction (HRI), what factors contribute to these reactions, and how they influence technology acceptance. METHODS: A systematic search of six major databases (PubMed, Scopus, IEEE Xplore, ACM Digital Library, PsycINFO, and Web of Science) identified studies published between January 2014 and March 2025. Following PRISMA-ScR guidelines, 49 studies were included, encompassing 6,670 older participants across 16 countries. RESULTS: Thematic synthesis revealed seven main fear categories: privacy and autonomy concerns, trust and reliability issues, emotional and ethical discomfort, usability challenges, fear of dependence, unfamiliarity with technology, and the Uncanny Valley effect. Fear levels were shaped by robot design, cultural background, prior technology experience, and contextual factors such as care settings. Mitigation strategies, including co-design with older adults, gradual exposure, transparent system behavior, and emotionally congruent interaction, were associated with improved acceptance. CONCLUSIONS: This review uniquely maps fear typologies to robot functions and intervention strategies, offering a framework to guide emotionally adaptive and culturally sensitive robot design. Addressing emotional barriers is essential for the ethical and effective integration of social robots into eldercare. Future research should prioritize longitudinal, cross-cultural studies and standardized fear measurement tools to advance evidence-based HRI implementation.