Are mindless robots less threatening? : The role of transparency about robots’ lack of human-like psychological capabilities

Sammanfattning: The use of social robots is often seen as a solution for handling future challenges such as caring for a growing population of elderly people. However, previous research has shown that robots can be perceived as threatening and for a successful implementation of robots in society we are dependent on the public’s acceptance of the technology. This thesis investigates whether transparency about robots’ lack of human-like psychological capabilities can reduce the perceived damage of robots on humans and human identity (henceforth “perceived damage”), and whether the effect of such transparency is moderated by the human-like appearance of the robot. Ninety-two study participants, randomly assigned to either a transparent or neutral condition, were presented with pictures and descriptions of four robots that varied in human-likeness. The capabilities of the robots were differently described in the two conditions, using either non-psychological (e.g., “programmed responses”) or psychological terminology (e.g., “respond in natural manner”). Participants subsequently filled in a scale measuring perceived damage. The results showed that the transparent condition was associated with lower perceived damage than the non-transparent condition for the most human-like robot. There was no significant interaction effect between transparency and human-like appearance. The findings suggest that transparency about robots' lack of human-like psychological capabilities can have a positive effect on perceived damage when the robot has a very human-like appearance. However, further research is needed to investigate whether transparency reduces perceived damage in real-world interactions with robots.