Interaction design for robotic avatars does avatar's aging cue affect the user's impressions of a robot?

Abstract

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. © 2013 Springer-Verlag Berlin Heidelberg.