Current Research on Sociable Educational Robots

• In Event: 28.026 - Designing Robots for Children: Multisensory Interactions and Multimodal Assessments

Abstract

This presentation will address Q1. What is the current status of research and development on child/robot interaction? The presentation will introduce three lines of educational robot research for young children: 1) a robot as a personalized learning companion for early literacy, 2) a robot as a mediator for children’s collaboration, and 3) the role of a robot and a human facilitator to support ASD children’s emotional development.

The Personal Robots Group at the MIT Media Lab pioneered sociable robots for young children (https://www.media.mit.edu/groups/personal-robots/overview). The personalized peer-like sociable robots provide friendly companionship, collaboration, perspective-taking, social modeling, and emotional engagement. This research has found that (1) children are more engaged and become more relational toward robots as companions than they did to an avatar; (2) children retain more phrases and words used by the robot, as well as telling longer storylines with expressive robots than with non-expressive robots; (3) children demonstrate a growth mindset after interacting with a growth mindset robot; (4) children become more emotionally expressive with a robot than with a tablet; and (5) the emotional data extracted from 218 hours of time-synchronized multimodal interaction have improved the prediction of children’s word-reading skill.

Second, project IDEAL (http://www.createresearch.net/projects.html) is an inclusive design that uses a robot to mediate collaboration among children coming from different linguistic and cultural backgrounds, grounded in multicultural education and intercultural communication (Kim, et al., 2018). The robot-mediation strategies are threefold. The robot (1) consistently invites children into conversations, (2) provides opportunities for children to speak and engage in activities in either their native or a second language, and (3) always demonstrates empathy with children. This research to date shows that (1) children develop affectionate relationships and are very engaged with the robot, (2) children interact with the robot as they would with a friend, (3) children are very forgiving of the robot’s mistakes, and (4) children gradually learn to work with their peers, by taking turns and listening.

Third, another popular line of research deals with robotic intervention for children with Autism Spectrum Disorder (Shamsuddin, et al., 2012; Kim, et al., 2013), examining the inclusion of human facilitators in the children’s interactions with a robot. This examination is critical in that i) even the state-of-art robots are not robust enough to meet children’s varied needs, ii) humans have the basic psychological need for relatedness to other humans that robots cannot replace, and iii) the goal of the child-robot research is to increase children’s adaptability in the real world, so adding humans into the loop ultimately provides a seamless transition from child/robot interaction to human-to-human interaction, eventually leading to children’s well-being. In an example of applying a robot to treat stuttering, the learning activities and tutor’s facilitation becomes more important than the robot itself (http://www.design4hri.net/research-projects.html).

Authors

• Yanghee Kim, Northern Illinois University

• Lixiao Huang

• Mengxi Zhou, Indiana University