An Empirical Investigation of Care Service Robot Acceptance Using an Extended UTAUT–Godspeed Framework

• In Event: AI in Health Systems and Healthcare Innovation

Abstract

Care service robots offer a promising solution to growing care demands from population aging, yet user acceptance remains low despite strong functional performance, as existing technology acceptance models focus narrowly on instrumental utility and overlook how affective and social perceptions shape adoption in high-stakes care settings. This study investigates the mechanisms underlying care robot acceptance by extending the Unified Theory of Acceptance and Use of Technology (UTAUT) with anthropomorphism, animacy, likeability, and perceived safety, asking whether the original UTAUT model applies to care robots, whether anthropomorphism and animacy moderate the relationship between facilitating conditions and behavioral intention, and whether likeability and perceived safety predict intention and use behavior beyond functional factors. A cross-sectional survey was conducted across the Guangdong–Hong Kong–Macao Greater Bay Area (n=303), targeting medical decision-makers, family caregivers, and older adults, with a diverse sample covering ages 18–76+, varied education levels, living arrangements, and chronic disease status. Using validated multi-item five-point Likert scales, data were analyzed with Partial Least Squares Structural Equation Modeling (PLS-SEM) including bootstrapping with 5,000 resamples to test path significance and moderation effects. The results show that behavioral intention strongly predicts actual use (β=0.780). Likeability (β=0.361), perceived safety (β=0.209), social influence, performance expectancy, and facilitating conditions all positively affect behavioral intention, while effort expectancy is non-significant. Perceived safety also directly enhances use behavior (β=0.122). Notably, animacy positively moderates the facilitating conditions–intention relationship (β=0.127), whereas anthropomorphism negatively moderates it (β=-0.119)—a novel finding indicating that lifelike responsiveness facilitates acceptance but excessive humanlike features may hinder it in care contexts. These findings demonstrate that acceptance of care service robots depends not only on functional value but critically on affective evaluation and safety perceptions, especially for vulnerable aging populations. Designers should prioritize animacy over excessive anthropomorphism to avoid psychological resistance, and policymakers should support safety-assured, emotionally inclusive robot deployment. This study advances equity in aging by centering underrepresented users’ emotional and safety needs, guiding the development of trustworthy, person-centered care automation.

Authors

• Chenxi Ye, Southeast University
  Presenting Author

• Chaoran Wang
  Non-Presenting Co-Author

• SHI Shenghui shenghui hui, Southeast University
  Non-Presenting Co-Author