Date: May 2017
JJ Lee, C Breazeal, D DeSteno (2017). Role of Speaker Cues in Attention Inference. Frontiers in Robotics and AI (In Review).
Jin Joo Lee. A Bayesian Theory of Mind Approach to Nonverbal Communication for Human-Robot Interactions. Doctoral Thesis, Massachusetts Institute of Technology, 2017.
To develop robot learning companions that can engage children like a social peer, we first have to understand how children themselves are emotionally expressive to their friends.
Over a span of five weeks, I collected real world data of 5-6 year-old children narrating stories to their classmates. From this data, we can better understand the type of behavior children of this age express when storytelling. A big assumption that current interactive technologies make is that children behave emotionally similarly as adults. For example, adults typically nod during conversations to communicate that they are engaged and attentive, but only a small fraction of children understand the function of this nodding behavior. A robot’s social behavior has to be appropriately designed to work across all ages.
I also use this data to develop machine learning models that help robots understand and communicate in the same emotional language as children.