Location
Hilton Hawaiian Village, Honolulu, Hawaii
Event Website
https://hicss.hawaii.edu/
Start Date
3-1-2024 12:00 AM
End Date
6-1-2024 12:00 AM
Description
As robots are becoming increasingly intelligent and autonomous, situations will increase where robots make decisions that cause side-effects. To investigate humans’ intentionality judgments in these situations, this study examines the Knobe effect in human-robot interactions. The Knobe effect describes the phenomenon that the perceived goodness or badness of the side effect of actions asymmetrically influences people’s intentionality attributions. Examining three different agents (i.e., human, humanoid robot, android robot), we found a Knobe effect for the human agent and the android robot, but not for the humanoid robot. The results suggest that as robots become more human-like, the Knobe effect becomes more relevant. Furthermore, a mediation analysis shows that existing explanatory approaches of the Knobe effect cannot mediate the effect of the nature of the side effect on intentionality judgments in the android robot. This work provides important insights into the debate about robots as intentional agents.
Recommended Citation
Kegel, Mona and Ghanem, Laid, "You Did That on Purpose! An Investigation of the Knobe Effect in Human-Robot Interactions" (2024). Hawaii International Conference on System Sciences 2024 (HICSS-57). 3.
https://aisel.aisnet.org/hicss-57/cl/social_and_service_robots/3
You Did That on Purpose! An Investigation of the Knobe Effect in Human-Robot Interactions
Hilton Hawaiian Village, Honolulu, Hawaii
As robots are becoming increasingly intelligent and autonomous, situations will increase where robots make decisions that cause side-effects. To investigate humans’ intentionality judgments in these situations, this study examines the Knobe effect in human-robot interactions. The Knobe effect describes the phenomenon that the perceived goodness or badness of the side effect of actions asymmetrically influences people’s intentionality attributions. Examining three different agents (i.e., human, humanoid robot, android robot), we found a Knobe effect for the human agent and the android robot, but not for the humanoid robot. The results suggest that as robots become more human-like, the Knobe effect becomes more relevant. Furthermore, a mediation analysis shows that existing explanatory approaches of the Knobe effect cannot mediate the effect of the nature of the side effect on intentionality judgments in the android robot. This work provides important insights into the debate about robots as intentional agents.
https://aisel.aisnet.org/hicss-57/cl/social_and_service_robots/3