Do I still like myself? Human-robot collaboration entails emotional consequences

Abstract

Investigating emotional processes has been vital for understanding human-human interaction. Specifically, emotional concepts of oneself and interaction partners shape interaction style and are associated with mental health and cognitive performance. Whether these concepts are equally relevant in human-robot interaction (HRI) has not been investigated. Here, we measured emotional concepts before and after collaboration with a telepresent robot described as (a) able, (b) unable to experience emotions, or (c) autonomous without reference to emotions, compared to a (d) control condition without human-robot collaboration. Emotional concepts were measured with the affective His-Mine-Paradigm (aHMP) in which participants were asked to affectively evaluate pronoun-noun-pairs related to themselves (e.g., “my victory”) or the robot (e.g., “its victory”). Results indicated that (1) the aHMP can be validly used in HRI contexts, (2) emotional self-concept got less positive after interacting with “emotionless” robots, and emotional robot-concept got more positive after interacting with (3) “autonomous” or (4) “emotional” robots. Results suggest that beliefs about and interactions with telepresent robots can change emotional concepts which themselves are associated with well-being, performance, and interaction style. In sum, we report emotional consequences of HRI and argue that such consequences should receive more attention in future research and HRI design.