As artificial intelligence systems increasingly make high-stakes recommendations and decisions automatically in many facets of our lives, the use of explainable artificial intelligence to inform stakeholders about the reasons behind such systems has been gaining much attention in a wide range of fields, including education. Also, in the field of education there has been a long history of research into self-explanation, where students explain the process of their answers. This has been recognized as a beneficial intervention to promote metacognitive skills, however, there is also unexplored potential to gain insight into the problems that learners experience due to inadequate prerequisite knowledge and skills that are required, or in the process of their application to the task at hand. While this aspect of self-explanation has been of interest to teachers, there is little research into the use of such information to inform educational AI systems. In this paper, we propose a system in which both students and the AI system explain to each other their reasons behind decisions that were made, such as: self-explanation of student cognition during the answering process, and explanation of recommendations based on internal mechanizes and other abstract representations of model algorithms.

In the realm of mathematics education, self-explanation stands as a crucial learning mechanism, allowing learners to articulate their comprehension of intricate mathematical concepts and strategies. As digital learning platforms grow in prominence, there are mounting opportunities to collect and utilize mathematical self-explanations. However, these opportunities are met with challenges in automated evaluation. Automatic scoring of mathematical self-explanations is crucial for preprocessing tasks, including the categorization of learner responses, identification of common misconceptions, and the creation of tailored feedback and model solutions. Nevertheless, this task is hindered by the dearth of ample sample sets. Our research introduces a semi-supervised technique using the large language model (LLM), specifically its Japanese variant, to enrich datasets for the automated scoring of mathematical self-explanations. We rigorously evaluated the quality of self-explanations across five datasets, ranging from human-evaluated originals to ones devoid of original content. Our results show that combining LLM-based explanations with mathematical material significantly improves the model’s accuracy. Interestingly, there is an optimal limit to how many synthetic self-explanation data can benefit the system. Exceeding this limit does not further improve outcomes. This study thus highlights the need for careful consideration when integrating synthetic data into solutions, especially within the mathematics discipline.

This research introduces the self-explanation-based automated feedback (SEAF) system, aimed at alleviating the teaching burden through real-time, automated feedback while aligning with SDG 4’s sustainability goals for quality education. The system specifically targets the enhancement of self-explanation, a proven but challenging cognitive strategy that bolsters both conceptual and procedural knowledge. Utilizing a triad of core feedback mechanisms—customized messages, quality assessments, and peer-generated exemplars—SEAF aims to fill the gap left by traditional and computer-aided self-explanation methods, which often require extensive preparation and may not provide effective scaffolding for all students. In a pilot study involving 50 junior high students, those with initially limited self-explanation skills showed significant improvement after using SEAF, achieving a moderate learning effect. A resounding 91.7% of participants acknowledged the system’s positive impact on their learning. SEAF’s automated capabilities serve dual purposes: they offer a more personalized and scalable approach to student learning while simultaneously reducing the educators’ workload related to feedback provision.

In the realm of mathematics education, self-explanation stands as a crucial learning mechanism, allowing learners to articulate their comprehension of intricate mathematical concepts and strategies. As digital learning platforms grow in prominence, there are mounting opportunities to collect and utilize mathematical self-explanations. However, these opportunities are met with challenges in automated evaluation. Automatic scoring of mathematical self-explanations is crucial for preprocessing tasks, including the categorization of learner responses, identification of common misconceptions, and the creation of tailored feedback and model solutions. Nevertheless, this task is hindered by the dearth of ample sample sets. Our research introduces a semi-supervised technique using the Language Learning Model (LLM), specifically its Japanese variant, to enrich datasets for the automated scoring of mathematical self-explanations. We rigorously evaluated the quality of self-explanations across five datasets, ranging from human-evaluated originals to ones devoid of original content. Our results show that combining LLM-based explanations with mathematical material significantly improves the model's accuracy. Interestingly, there's an optimal limit to how much synthetic self-explanation data can bene-fit the system. Exceeding this limit doesn't further improve outcomes. This study thus highlights the need for careful consideration when integrating synthetic data into solutions, especially within the mathematics discipline.

Self-explanation is increasingly recognized as a key factor in learning. Identifying learning impasses, which are significant educational challenges, is also crucial as they can lead to deeper learning experiences. This paper argues that integrating self-explanation with relevant datasets is essential for detecting learning impasses in online mathematics education. To test this idea, we created an evaluative framework using a rubric-based approach tailored for mathematical problem-solving. Our analysis combines various data types, including handwritten responses and digital self-explanations from 93 middle school students. Using hierarchical logistic regression, we examined feature groups such as Self-Explanation Quality, Handwriting Features, and Overall Level of Action. Models based solely on self-explanation achieved a 74.0% accuracy rate, while adding more features increased the final model’s accuracy to 80.06%. This improvement highlights the effectiveness of an integrated approach. The combined model, which merges generated handwriting features counts with self-explanation features, shows the importance of both qualitative and quantitative measures in identifying learning impasses. Our findings suggest that a comprehensive approach, leveraging detailed operational data and rich self-explanation content, can enhance the detection of learning challenges, providing insights for personalized education in online learning environments.