Intelligent Tutoring Systems (ITSs) for Math still use traditional data input methods: computers’ keyboard and mouse. However, students usually solve math tasks using paper and pen. Therefore, the gap between the manner the students work and the requirements imposed by these typing-based systems expose students to an extraneous cognitive load, impairing their learning. Our study investigates the impact of the data input method on students’ learning and fluency in solving equations using step-based math ITSs. More specifically, we have considered the standard typing and handwriting input methods. We hypothesized that the students would be more fluent using their handwriting with online recognition to solve math equations than using the typing input method. This fluency indicates a reduction in cognitive load, freeing working memory for logical reasoning instead of interface preconditions, leading to improved learning. We have conducted an experiment with 55 seventh-grade students from a private school to validate the hypothesis, randomly assigned to control and experimental groups. Each group used one of the input methods on two different devices (desktop computers and tablets). Although students using handwriting solved more equations and were faster than students who typed their equations, we could not find statistically significant differences in the learning between students that used typing or handwriting. Additionally, we have found that the input method used in a not ideal device (e.g., handwriting with a computer’s mouse instead of using a touch screen device) can negatively affect the students’ performance.
The definition of effective pedagogical strategies for coaching and tutoring students according to their needs is one of the most important issues in Adaptive and Intelligent Educational Systems (AIES). The use of a Reinforcement Learning (RL) model allows the system to learn automatically how to teach to each student individually, only based on the acquired experience with other learners with similar characteristics, like a human tutor does. The application of this artificial intelligence technique, RL, avoids to define the teaching strategies by learning action policies that define what, when and how to teach. In this paper we study the performance of the RL model in a DataBase Design (DBD) AIES, where this performance is measured on number of students required to acquire efficient teaching strategies.