We present a study on the impact of unexpected robot behaviors on the perception of a robot by children and their subsequent engagement in a playful interaction based on a novel “domino” task. We propose an original analysis methodology which blends behavioral cues and reported phenomenological perceptions into a compound index. While we found only a limited recognition of the different misbehaviors of the robot that we attribute to the age of the child participants (4– 5 years old), interesting findings include a sustained engagement level, an unexpectedly low level of attribution of higher cognitive abilities and a negative correlation between anthropomorphic projections and actual behavioral engagement.
CITATION STYLE
Lemaignan, S., Fink, J., Mondada, F., & Dillenbourg, P. (2015). You’re doing it wrong! studying unexpected behaviors in child-robot interaction. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9388 LNCS, pp. 390–400). Springer Verlag. https://doi.org/10.1007/978-3-319-25554-5_39
Mendeley helps you to discover research relevant for your work.