Long-term Social Interaction with an Expressive Robot

Kasap, Z. and Magnenat-Thalmann, N.

Abstract: This paper describes an expressive robotic learning companion for long-term interaction that can remember past exchanges with users and that can respond emotionally based on its initial personality. Previous work on long-term interaction with virtual characters and robots frequently reported that after the novelty effect disappeared, users’ interest into the interaction decreased with time. Our primary goal in this study was to develop a system that can still keep the attention of the users after the first interaction. We designed an experiment to measure the changes in social presence, task engagement and motivation level by considering memory and personality factors. Different from previous work on long-tem interaction, we found out that users’ interest in our system didn’t decrease with time and even increased from one interaction to the other. In addition, we looked at the effect of individual components in our system to the measurements. The results provide first evidence that existence of memory in a long-term interaction system can help to keep the attention of the users as time passes.

  booktitle = {Computer Graphics International},
  author = {Kasap, Z. and Magnenat-Thalmann, N.},
  title = {Long-term Social Interaction with an Expressive Robot},
  year = {2011},
  topic = {Social Robotics}