Ostracism research has shown that even being ostracized by a computer or observing ostracism hurts. Intersecting these two standpoints with human-robot interaction (HRI) research, we conducted three preregistered experiments to investigate observers' evaluations of vicarious robot ostracism. In Study 1 (N = 140), people observing an excluded (vs. included) robot in a Cyberball game attributed higher negative emotions and need-threat to the robot, experienced more negative emotions, perceived the humans more negatively, and perceived the human-robot group as less entitative. Study 2 (N = 172) replicated Study 1 and showed that participants' responses did not vary for an excluded human or robot. Study 3 (N = 238) replicated the previous findings using a social media ostracism paradigm and showed that not even the robot's anthropomorphic features influenced observers' responses. The research suggests that the ostracism detection system functions even for vicarious robot exclusion and uncovers anthropomorphism, human intragroup, and human-robot intergroup processes unfolding in HRIs.
Marinucci, M., Pagliaro, S., Teresi, M., Ballone, C., Riva, P. (2025). How low can you still go? Observing an ostracized robot is sufficient to hurt humans’ feelings. GROUP PROCESSES & INTERGROUP RELATIONS [10.1177/13684302241309578].
How low can you still go? Observing an ostracized robot is sufficient to hurt humans’ feelings
Marinucci M.
;Riva P.
2025
Abstract
Ostracism research has shown that even being ostracized by a computer or observing ostracism hurts. Intersecting these two standpoints with human-robot interaction (HRI) research, we conducted three preregistered experiments to investigate observers' evaluations of vicarious robot ostracism. In Study 1 (N = 140), people observing an excluded (vs. included) robot in a Cyberball game attributed higher negative emotions and need-threat to the robot, experienced more negative emotions, perceived the humans more negatively, and perceived the human-robot group as less entitative. Study 2 (N = 172) replicated Study 1 and showed that participants' responses did not vary for an excluded human or robot. Study 3 (N = 238) replicated the previous findings using a social media ostracism paradigm and showed that not even the robot's anthropomorphic features influenced observers' responses. The research suggests that the ostracism detection system functions even for vicarious robot exclusion and uncovers anthropomorphism, human intragroup, and human-robot intergroup processes unfolding in HRIs.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.