If you also believe that children are our future, know that robots will be right alongside them. New research offers a glimpse at a that future where robots could become an important social and educational tool for kids, by showing that children are more motivated to read when they’ve got helper bot encouraging them.
Published Wednesday in the journal Science Robotics, the study the study uses “Minnie,” described as “a learning-companion robot” who helps early adolescent children engage in reading over a two-week period during summer break. As part of their research, 24 kids aged 10-12 read a book either by themselves and then complete a “paper-based” reading activity, or read to 13.5-inch tall Minnie instead. Their positive or negative experiences with reading were compared by Joseph Michaelis and Bilge Mutlu, researchers at the University of Wisconsin-Madison.
Mutlu tells Inverse part of his and Michaelis’ research is about “finding unique opportunities in education where the robot can provide additional benefit, such as serving as a reading companion to improve reading interest and motivation.” Michaelis agreed, and tells Inverse: “Our vision is for a social robot to be a supplement to educational activities or learning environments.”
Interestingly, Minnie the robot has the ability to execute some human-like qualities, according to the report:
“Minnie was programmed to respond with thoughtful comments on the book of choice, make eye contact with the child, avert its gaze when speaking to appear more thoughtful, and make idle movements to appear more lifelike.”
“As the child reads, AprilTags will appear roughly every three to six pages of each book, and the child is instructed to scan these tags when they encounter them. The tag ID is linked to a specific comment, written by our research team, related to what is currently happening in the book (DE4). When a scan is received by Minnie, it blinks, makes a distinct beep, and delivers the comment via text-to-speech,” write the researchers.
In a rather fascinating outcome, although the subjects involved described both reading activities as “positive experiences,” compared to the control group, fewer kids who read to the robot described the reading activities in negative terms.
In possibly the most important takeaway from the study, the children said Minnie “helped them better understand what was happening in the book,” and made them want to read more often than the kids in the control group.
In particular, they indicated that Minnie encouraged them to spend more time working on sounding out unfamiliar words. It appears that reading to Minnie instead of alone motivated the children to read and “make sure what you’re saying is correct,” one subject noted.
Additionally, because of the comments Minnie could make, the kids believed Minnie had a personality and emotions, and could develop “a deepening social connection with the robot.”
Another paper on Minnie, published last year in CHI ‘17 Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems, also used Minnie as a way to explore families’ habits and views about reading. In that study, Michaelis and Mutlu worked with eight families, and looked at how children perceived reading with Minnie. Those findings also indicated that reading with the learning companion provided a “way to socially engage with reading.” “Since Minnie was designed to behave socially,” the authors wrote in2017, “this behavior seemed to help children find social companionship while they read with it.” Companionship and connection seem to be key with this reading robot.
The results of the study might remind one of programs where kids read to dogs in shelters. The dogs provided a nonjudgmental reading partner for kids developing literacy skills, just like a robot could, but they can’t react the way Minnie can.
The study reports that before and after reading activities, Minnie would tell the kids she “enjoys reading with them, learns from their reading, and looks forward to reading with them again.”
Indeed, Minnie’s presence seemed especially important in at least one instance, when a subject said reading with Minnie was “better than normal reading,” and that with Minnie, “I have someone to read to that doesn’t interrupt.”
“Educational use of robots is a rapidly expanding area of study and has the potential to provide powerful and personalized support to learners from a wide variety of backgrounds, ages, abilities, and interests,” the authors concluded.
Michaelis tells Inverse, “I can see our approach being applied in or out of a classroom whenever someone would benefit from learning with a social partner, but a human partner isn’t available. I can also see the robot serve as the facilitator within a group of kids to help guide their conversation, and act as a peer resource for the group.”
But Mutlu cautioned that “We have be careful in identifying and designing for these opportunities, as we do not want robots to replace natural ways of learning or available resources, such as interaction with a parent or an educator.” He also tells Inverse, “in principle, a robot like Minnie can be used by adult learners as well, although its behaviors and support for reading must be designed specifically for the target user group.”
It should be noted that “there was a measurable reduction in time spent reading over time” in the study, and the authors suggested that some technical aspects of the robot that the kids found “frustrating” may have kept the children from reading more often as time went on. The researchers observed in a paper last year that a wider breadth of study is needed on Minnie — longer amounts of time, different settings, different ages of children, and so on— but are encouraged by their work .