The 22nd annual ACM Interaction Design and Children (IDC) Conference was held in Chicago last week and featured some interesting and exciting design experiments of how children interact with robots. In this article, we discuss some interesting topics around studies involving the use of robots for in-home learning and helping children with slightly impaired learning abilities.
Robots as an aid to learning at home
There were multiple studies presented by researchers at the People and Robots Laboratory at University of Wisconsin-Madison which explored how the Misty II robot (Priced in the $2500 range) can be effectively used by as a learning aid by children. One of the studies focused on how the robot can prompt parents to have conversations with children around math topics. A study was performed on 9 families with children between the ages of 4 and 5. The robot helped the parent and child in a reading and a structured play activity. As an example, in the reading activity, the parent read out a book to the child. Each page of the book was marked with a tag, and when this tag would be shown to the Misty II robot, the robot would utter a predefined prompt designed to promote a conversation between the parent and the child. The studied showed that parents agreed that the social robot’s prompts of math talk facilitated valuable discussions on the topic with their child, which they would not have had otherwise. The parents also indicated a strong preference for having the robot independently engage the child when the parent is busy so that the child is occupied and doesn’t get bored. A brief demonstration of the exercise is available in the a demo video.
We found this idea very interesting and important because the tag based prompting can be easily designed with smaller robots such as the Anki Vector. The study also shows the value of using robots to prompt human conversations. If you are keen to perform a similar experiment in your home setting with the Anki robot, please reach out.
Another study from researchers at the People and Robots Laboratory explored social learning companion robots which can help and support student’s while they do their homework. This is an online experiment where the researchers worked with 10 middle school children and their parents. The study involved students and teachers prepare a interactive homework guidance tool for a science article on fossils. As the students read the article, they annotated it with comments on what could support a good homework and learning experience. The teachers did the same exercise from their point of view. Once the students were done, they were shown the teachers comments and asked if they wanted to edit their own comments. Then, they were asked to translate these comments into speech, expressions, and emotions that they would like to see in a Misty II robot to be used as a homework companion. At the end, the students were asked to share results with their parents. The researchers spoke to both the students and the parents to gather inputs regarding the exercise.
The results showed that the students first preferred the robot to have clarity of speech. Their next priority was to design great emotional expressions to support the speech, such as, designing facial expressions, head, arm and body movements, or changes in the chest light. The students believed that the emotional expressions such as showing interest or anticipation would be great motivating factors to do homework. Most parents too thought that the reactions from the robot would make a fun and engaging atmosphere to facilitate learning. Another finding from the study is that most children wanted the robot to be a friend at home and act as a personal companion instead of a shared resource. The study concludes with a set of design guidelines for manufacturers of educational robots along the lines of the above observations.
What do we think?
Both studies make important contributions in showing ways in which robots can be used as a learning tool at home. We see a vast potential for exploration and commercialization of some of these ideas. It is also interesting to see that many ideas can be easily translated to smaller robots, such as the Anki Vector, which would be much more affordable in family settings compared to the Misty robot. We have all witnessed how online apps such as those from Byju’s have flooded the Education Technology (EdTech) market. Robots definitely have a great potential in education if the rough edges can be sorted out and a product with mass appeal and ease of use be made.