That Robot Touch that Means so Much: On the Psychological Effects of Human-Robot Touch

摘要

Robots are unique compared to other forms of artificial intelligence because they own a physical body that allows them to get into physical contact with humans. There is a huge amount of literature and empirical work that demonstrates the importance of touch for the development of man as well as other species as monkeys. Especially, the positive impact of interpersonal touch on health, bonding and attachment has been frequently observed. Furthermore, positive evaluative consequences of interpersonal touch on the toucher, i.e., the initiator of touch, and the surroundings has been revealed, as well as an influence on compliance and helping. Taken together it seems as if touch is a powerful way of nonverbal communication that is capable to increase the well-being of the receiver, and the initiator of touch, too. Against the background of earlier work in the realm of human-computer interaction that demonstrated that people equate interactions with interactive media as computers or robots with interpersonal interactions (media equation: Nass & Moon, 2000; Reeves & Nass, 1996; see also Krämer & Hoffmann, 2016, for an updated overview), it is assumed that touch from a robot can result in comparable positive impacts on human well- being. As a prerequisite for the examination of the physiological and psychological effects of human-robot touch, basic research was first necessary to structure the field of human- robot touch in comparison to interpersonal touch. For that purpose, four empirical studies were conducted that untangled the perception of human-robot touch referring to research on interpersonal touch, to test whether the media equation also applies to a fundamental interpersonal phenomenon such as touch. Therefore, the perception of different forms of human-robot touch with different robots (Study 1 & 2) in different interaction contexts (Study 3) were first considered in the most controllable way by means of the observation and evaluation of photographs. In addition, the underlying appraisal process that decides upon the final reaction to robot touch was theoretically and practically considered based on expectancy violations theory (Burgoon, 1983; Burgoon & Hale, 1988) that originates from the field of interpersonal communication. Finally, actual touch initiated by a robot to participants in a laboratory experiment were regarded and analyzed with respect to the impact of robot touch on participants’ subjective evaluation, as well as on observable reactions as laughing, compliance and helping (Study 4). In conclusion, influencing variables that determine the desirability of human-robot touch were revealed, namely characteristics of the robot (morphology and size), characteristics of the person (e.g., attitudes towards touch, attitudes towards robots, touch expectancies), the intimacy and direction of touch, i.e., who touches whom, and the meaning assigned to human-robot touch regarding relational interpretations of closeness and affection. Finally, positive emotional reactions to actual robot-initiated touch, and heightened compliance have been observed in a laboratory experiment.