While recent advances in robotics are improving the value of digital tutoring experiences, there are still some challenges to overcome, especially when it comes to students asking questions.
This study hypothesized that designing adaptive help strategies would improve learners' approaches to asking for help and would ultimately help them learn better. It compared students’ performances with two robot tutoring groups; one had on-demand help from the robot and the other used adaptive supporting strategies. Two adaptive support strategies were used in the adaptive robot groups: in the case that learners requested consecutive hints without trying the problem in between, the robot would ask learners to try to solve the problem before asking further assistance; in the case when learners did not ask for help after failing to solve problems in two consecutive times, the robot would provide them with the next hint that they haven’t yet requested.
Findings reveal that adaptive help strategies effectively reduced hint requests from those students who wanted to game the system through clicking hints continuously; these designs also forced learners to spend more effort on problems. For those who were not used to seeking help, these strategies encouraged them to get the necessary support. Most importantly, students in the adaptive strategies robot tutoring group improved their test scores, while the on-demand help robot tutoring group members did not.
As robots increasingly play various roles in teaching and learning activities, we need a clearer understanding of how they can effectively facilitate learning. This study shows that simple adaptive help reminders can effectively enhance learning. More research on interactions between robot tutors and learners will help learners to get the most out of these one-on-one sessions.
Ramachandran, A., Litoiu, A., & Scassellati, B. (2016, March). Shaping Productive Help-Seeking Behavior During Robot-Child Tutoring Interactions. In The Eleventh ACM/IEEE International Conference on Human Robot Interaction (pp. 247-254). IEEE Press.Image: Robot via Flickr