The Effects of Stakeholder Perceptions on the Use of Humanoid Robots in Care for Older Adults: Postinteraction Cross-Sectional Study (Preprint)

BACKGROUND Efficient use of humanoid social robots in the care for older adults requires precise knowledge of expectations in this area. There is little research in this field that includes the interaction of stakeholders with the robot. Even fewer studies have compared the perceptions of older people (as care recipients) and professional caregivers (representing those taking care of older adults in teams with robots). OBJECTIVE The aim of this study was to analyze whether specific aspects of the perceptions about humanoid robots influence attitudes after interacting with the robot and to compare the opinions of different stakeholders (older people and their professional caregivers) on this topic. We analyzed the potential impact of the differences in perception of the robot between stakeholder groups with respect to how the robot should be designed and tailored to fit the specific needs of future users. We also attempted to define areas where targeted educational activities could bring the attitudes of the two groups of stakeholders closer to each other. METHODS The studied group was a conveniently available sample of individuals who took part in the presentation of and interaction with a humanoid social robot. Among them, there were 48 community-dwelling older adults (aged ≥60 years), who were participants of day care units (which may signal the presence of self-care needs), and 53 professional caregivers. The participants were asked to express their views after an interaction with a humanoid social robot (TIAGo) using the Users’ Needs, Requirements and Abilities Questionnaire (UNRAQ) and the Godspeed Questionnaire Series (GQS). RESULTS Compared to the caregivers, older adults not only assessed the robot more positively with respect to its roles as a companion and assistant (<i>P</i>=.009 and <i>P</i>=.003, respectively) but also had higher scores on their need to increase their knowledge about the robot (<i>P</i>=.049). Regarding the robot’s functions, the greatest differences between groups were observed for the social aspects on the UNRAQ, including decreasing the sense of loneliness (<i>P</i>=.003) and accompanying the user in everyday activities (<i>P</i>=.005). As for the GQS, the mean scores of the Animacy, Likeability, and Perceived Intelligence scales were significantly higher for older participants than for caregivers (<i>P</i>=.04, <i>P</i>&lt;.001, and <i>P</i>&lt;.001, respectively). The only parameter for which the caregivers’ scores were higher than those of the older adults was the Artificial-Lifelike item from the Anthropomorphism scale of the GQS (<i>P</i>=.03). CONCLUSIONS The acceptance of the social functions of a humanoid robot is related to its perception in all analyzed aspects, whereas the expected usefulness of a care robot is not linked to aspects of anthropomorphism. Successful implementation of robots in the care for older people thus depends on considering not only the fears, needs, and requirements of various stakeholders but also on the perceptions of the robot. Given the differences between the stakeholders, targeted and properly structured educational and training activities for caregivers and prospective users may enable a seamless integration of robotic technologies in care provision.