Researchers at GAIPS Lab (University of Lisbon) in Portugal and CHILI Lab (Ecole Polytechnique Fédérale de Lausanne) in Switzerland have recently developed a self-directed system designed to help children in enhancing their handwriting skills. The research was published in a paper published in Springer’s International Journal of Social Robotics. The system involves the use of a social robot in individual learning sessions with children.
It is not surprising that for some children, handwriting can be a difficult skill to acquire, yet it is an essential stepping stone in their academic route. In fact, poor writing can adversely affect a child’s self-esteem, academic performance, and learning enthusiasm.
To learn the ropes of handwriting, a child is required to learn to coordinate motor, cognitive and perceptual skills, thus he/she might also need an extensive amount of practice. Therefore the children who find writing by hand mostly challenging are generally provided extra support from their teachers during one-to-one sessions.
“As the attainment of handwriting skills is a physical job and often needs physical interaction and assistance, modern technologies such as social robots can be used as an instrument to complement existing handwriting involvement methods for children,” said Shruti Chandra, one of the investigators who carried out the current study. “With this idea in mind, our research targets at investigating how a social robot can autonomously help children with the gaining of handwriting with the method where the children are the teachers who help the robot to improve its writing.”
The latest study carried out by the researchers at the GAIPS Lab is a section of a broader project named CoWriter. The main objective of the CoWriter project is to construct a robotic partner that children can teach handwriting to.
The basic idea behind this is that training a robot how to write by hand can help children to improve their own skill to write; a concept known as learning-by-teaching. Previous studies suggest that learning-by-teaching practices can be very beneficial in education, as they permit children to mirror on their own mistakes and improve, eventually improving their academic skills, self-esteem and motivation.
“We developed an autonomous educational method integrating a social humanoid robot (Nao by SoftRobotics) that offers a situation for children to improve their handwriting skills, Chandra explained. “The current method depends on the learning-by-teaching paradigm with the aim of answering a question: How would the learning capabilities of a tutee influence the learning of a tutor?”
Chandra and his colleagues designed a one-to-one setting that involves communication between a social robot and a child, in which the child takes on the character of an educator or tutor, evaluating the robot’s handwriting skills. This learning scenario is designed to increase children’s handwriting skills by asking them to teach someone else. The learning system offered by the researchers is entirely autonomous, so it does not require any extra work or administration during the child-robot contact.
“The interaction starts when the robot requests for help from the child in rectifying shapes of a few letters,” Chandra said. “First, the robot writes a distorted letter on the screen and asks the child to rewrite it using the computer screen. The child does correction by altering the shape of the letter through a slider and also provides an illustration of the letter on the screen.”
During the child-robot training session, the robot makes some of the most general handwriting mistakes witnessed in 4-8-year-old children, related to breaks, proportion and alignment. This allows the child who is instructing the robot to reflect on some of his/her own errors while correcting errors made by the robot.
To gauge the success of the teaching system they developed, Chandra and his colleagues started up two longitudinal studies in Portuguese elementary schools, each including four robot-child one-to-one tutoring sessions. They assessed each contributing child’s handwriting skills and his/her opinion of the social robot both before and after each session.
The children were allocated to one of three learning conditions: continuous learning, personalised learning or non-learning. This allowed the researchers to efficiently compare the effects of each of these learning conditions on the children’s learning and views of the robot.
“In the continuous-learning condition, the robot learns endlessly, within and between the meetings,” Chandra said. “Though, in the non-learning condition, the robot does not learn and constantly shows poor performance. Lastly, in the personalised learning condition, the robot accepts the child’s performance throughout the sittings, meaning if a child does well, the robot would do better or vice-versa.”
Before the commencement of each tutoring session, the children were asked to complete a pretest assessing their handwriting skills. Afterwards, they performed the combined writing activity, in which each child was asked to demonstrate the robot how to write by hand. Lastly, at the end of the task, the researchers asked the children questions concerning their insights about the interaction with the robot.
“Succeeding the four sessions with the robot, once every week, the participants did a post-test similar to the pretest to assess the effect of the four sessions on participant’s handwriting skills,” Chandra said. “We then matched the children’s pre- and post-test scores to gauge their learning gains within and between the conditions. Similarly, the children’s observations were analyzed and compared between and within the conditions.”
The researchers analyzed the statistics they collected and discovered that the handwriting skills of children in the personalized learning and continuous-learning conditions elevated significantly after the tutoring sessions. On the contrary, the handwriting of children in the non-learning condition did not improve substantially. In other words, coaching a robot with some handwriting capability appeared to help the children improve their own abilities, much more than tutoring a robot with no competency.
“We also found that the children’s opinions regarding some of the robot’s learning capabilities, such as writing ability and general performance, changed over time,” Chandra said. “After the last communication, more than 92% of the children in the primary study and all the children in the second study not only regarded themselves as a good teacher but also liked coaching the robot and wanted to teach it in the future.”
Apart from enhancing the children’s handwriting skills, the learning approach developed by the researchers also gives the impression of strengthening the children’s confidence in their abilities. Remarkably, the robot’s proficiency did not affect how confident in their abilities the children felt after the exercise, as they routinely considered themselves as proficient teachers regardless of whether the robot had improved or not.
“In general, we found that the children approved the system in both studies for some reasons,” Chandra said. “Firstly, we were able to execute two multi-session studies successfully in two different schools in Lisbon, Portugal and the children working with the robot through four to six weeks. Moreover, we found that most of the children desired to teach the robot in the future.”
In the future, the autonomous educational system introduced by Chandra and his associates could be used to teach children handwriting skills in an innovative and engaging manner. Furthermore, the results of initial assessments in schools suggest that it could help to improve both children’s handwriting and their confidence in their talents.
The researchers are now conducting another research for the CoWriter project that examines the use of a similar system to teach children with critical handwriting impairments. In this case, the children will be attending sessions with a therapist and the Nao social robot.
“In these other trainings, a therapist is present to help the child regarding the task-related questions and collaboration with the social robot, over numerous months in hospitals located in Lausanne, Paris, Switzerland, and France,” Chandra said. “We are intending to keep doing such studies with robots as these children need many sessions for progress. Our group at the CHILI Lab has also lately introduced an app called Tegami for making granular dysgraphia diagnoses in children so that they can obtain customized treatment for learning how to write.”