No abstract

This paper explores the relationship between the aging cue of a robotic avatar and the level of intelligence and safety perceived by the elderly as users. This initial study found that the avatar aging cue indeed, affects the elderly in their perception of the embodied robot, in terms not only of its intelligence but also of its safety: the elderly perceived the robot more intelligent and safer with older avatars.Due to the fact that the elderly perceived the aging cue of avatars as an effect of their expectation and interactions with the robot, the finding related to the avatar-user aging cue influences in the design of a series of attributions of the embodied robot. Therefore, the result of this study can provide interaction designers with a guideline in creating the visual appearance of an embodied agency in terms of its aging cue.

Human Computer/Robot Interaction has concerned about developing embodied computer/robot agents effective for their use in user interaction. In particular, the study on the dependency of the interaction design on the target users has been of a core theme to investigate. For instance, in the case of designing an embodied agent such as an avatar specifically to support the performance of a robot assistant to the elderly, the aspect of social interaction with the older adults should be of a serious concern. In this paper, we present a study that explores the relationship between the degree of aging cues (i.e., the visual features related to the age of embodied agents) and the level of perceived anthropomorphism, intelligence, safety and likeability by the older adults as customers. The study found that avatar aging cues affect the perception of the older adults in intelligence and safety: the older adults perceived the agent more intelligent with older avatars but safer with younger avatars. However, the aging cue seems not affecting the sense of anthropomorphism and likeability on users. An Interesting finding is the difference in the likability associated with the aging cue according to the gender of the older adults: the male participants tend to like older avatars while the female participants the younger ones. Since how the older adults perceive the aging cues of avatars could affect their expectation and trust on the assistant robots, thus, the findings related to the aging cue influence in the design of a series of attributions of the robots in terms of their roles and capabilities. Based on the results of this work, we can approach toward design considerations to help guide interaction designers in creating the visual appearance of an embodied agency as the robotic avatar interfaces for the elderly.