Eye tracking provides a marker of attention. In the educational context, such behavior can be harnessed to understand learning behaviors. However, a technology framework that captures and utilizes such multimodal indicators in educational activities is lacking. This paper presents LA-ReflecT, a platform integrating multimodal data for micro-learning activities. Teachers can author learning tasks and enable tracking eye fixation behaviors. A web camera-based eye-tracking function captures the gaze data while attempting the learning task. Learners can control the settings to stop or pause recording. We present data-driven services such as visualizing gaze attention heatmap and genetic algorithm-based group formation. A classroom study with 41 students illustrates using the proposed framework in an authentic context. Data collected is analyzed to answer an initial research question regarding the correlation between the heterogeneity of the click and gaze patterns in a learning task. The work is open for a demo.

This study investigates the transferability of Self-Direction behavior across different contexts and phases of learning using the GOAL system. Self-directed learning (SDL) is crucial for lifelong learning. It is significantly influenced by Self- Direction Skills (SDS), a meta-skill that is said to be transferable across different contexts, including the ability to identify learning needs, set goals, select strategies, and evaluate outcomes. Utilizing log data collected from Japanese junior high schools and analyzed using the iSAT system, we explored how Self-Direction behavior acquired in one context can be transferred to another and how these skills vary across the SDL phases. The results indicated that the Self-Direction behavior transferred between different activities and phases. In addition, the way of transfer is suggested to vary from phase and context. This study provides useful insights for the design and guidance of SDL support systems in educational programs. It suggests that it is important for educators to identify factors that facilitate the development and transfer of SDS.

This study looks at learning habits of temporal regularity in learning activities. Building such habits involves learners' regulation of their behaviors and requires learning strategies for time management, which is a cornerstone of self-regulated learning (SRL). Given the importance of habit-building in education, Learning Analytics (LA) techniques have been applied to various long-term supports by monitoring learners' habitual behaviors from the trace data. However, building a learning habit does not always mean the productive use of time. Scant supports attend to recommending learners by building which habit can improve their learning productivity. Hence, this study proposes recommendations for productive learning habit-building from learning logs. We focus on the context of English reading in a Japanese junior high school and design an algorithm to compute a recommended learning time slot. Furthermore, we collect learners' perceptions of their productivity and learning status at different times of the day. The comparison between self-report and log data presents that learners are not aware of their learning as the detection from their learning logs. This implies the potential of the proposed recommendations for facilitating learners to build productive learning habits. Specifically, our study can suggest an optimal time in learning plans and provide learners with a sustainable cue to automate learning behaviors from long-term perspectives. By building productive learning habits, learners can become more engaged in their studies as well as lead more balanced lives.

This study aims to investigate the relationship between students' assignment submission behavior and final grades in information literacy education using a large volume of learning logs stored on the LMS. A total of 12,516 freshman students participated in this study from the year 2018 to 2022, across the COVID-19 pandemic. The students were divided into high, medium, and low performance groups using k-means clustering. The results of the characteristics analysis show a significant early submission behavioral trend and a late submission behavioral trend in high and low performance groups, respectively. The on-demand class format during the COVID-19 pandemic resulted in more consistent early submission behavior for high performance students and late submission behavior for low performance students, respectively. The findings suggest that time management skill is a critical factor in both blended and online learning environments, affecting weekly submission behavior and final grades.

Self-directed learning (SDL) is considered a crucial skill for 21st-century learners, promoting personalized and responsive educational experiences. This study explores the untapped potential of self-reflection, particularly in e-learning environments. The research focuses on self-reflection notes, which contain strategies past students adopted when facing different situations or challenges. These notes can help current or future students facing similar situations. In this study, students take tests weekly and leave their self-reflection notes after tests. In these notes students recorded their feelings and issues, offering perspectives and insights that experts might overlook or misunderstand in some details, thus failing to provide appropriate assistance. Extracting and categorizing information from self-reflection notes is crucial to further utilize this data. Our research introduces a machine learning-based approach that effectively classifies these self-reflection notes such as cognitive, metacognitive, experiential, and irrelevant text. We compare the performance of BERT, based on the transformer architecture, with traditional machine learning classifiers such as Support Vector Machines (SVM) and Random Forests (RF). Additionally, we enhanced the BERT model by training it on synthetic data generated through GPT-4 and employing a hybrid loss combining Supervised Contrastive Learning (SCL) and Cross-Entropy (CE) to improve classification capabilities. Our results indicate that the BERT model, enhanced with advanced training techniques, outperforms traditional models in classifying learning strategies from self-reflection notes. This study not only advances the understanding of SDL in online learning environments but also demonstrates the potential of tailored machine-learning solutions to foster more effective and adaptive learning strategies.

This paper reports an innovative data capturing pipeline for learner modeling by integrating learning logs with physiological and environmental sensor data. The learning logs are collected from BookRoll, an ebook reader. Wearable device signals consist of the user’s affective state from a new version of the eye movement tracking device JINS Meme. An Omron sensor was used to collect environmental data like temperature, noise level, humidity, and luminescence. The architecture of the data collection and its potential are presented in this paper: As a pilot study participants did calculation and comprehension tasks in the ebook reader in two environmental conditions: one within a research lab and one at a cafe to simulate self-study environments. The data from the sensors were collected and synchronized to provide descriptive statistics.

Self-directed learning (SDL) is an important skill in the 21st century, while the understanding of its process in behavior has not been well explored. Analysis of the sequential behavior patterns in SDL and the relations with students’ academic performance could help to advance our understanding of SDL in theory and practice. In this study, we mined the behavioral sequences of self-directed extensive reading from students’ learning and self-directed behavioral logs using differential pattern mining technique. Furthermore, we built models to predict students’ academic performance using the conventional behavior frequency features and the behavior sequence features. Experimental results identified 14 sequential patterns of SDL behaviors in the high-performance student group. The prediction model revealed the importance of sequential patterns in SDL behavior, which was built with an acceptable AUC. These findings suggested that several SDL strategies in behavior contribute to students’ academic performance, such as analysis learning status before planning, planning before learning, monitoring after learning.

ENaCT is a framework for the design and analysis of critical thinking environments based on 4E cognition perspectives. In this paper, we describe a web-based critical thinking environment designed to implementœ the ENaCT framework. When users perform a critical thinking task in the environment their interaction logs are captured. We report on a pilot study with undergraduate participants and analyse how participants used the affordances in the environment as they performed the critical thinking task. One case of task-related behaviours (high activity) is elaborated to highlight the current possibilities of the system and discuss implications for redesign.

Self-direction skills in the context of learning can be supported with data in this digital era. This study analyzes the behaviors of learners during a self-directed reading and summarization task. Our work investigates an undergraduate course (n=72) where students worked on a reading and summarizing assignment while planning and monitoring the task in GOAL, a platform synthesizing learner's activity data from learning and physical activity contexts. This study focuses on the initial cohort analysis of the students' behavior based on the fine grain interaction data collected in the different systems using visual analytics techniques. Such a data-rich narrative of self-directed in-semester activities is not discussed yet in the literature to our knowledge. We discuss the implications of the trends that is found in our collected dataset for designing AI-support for self-direction skills with the GOAL platform and the scope of deeper analysis to further understand the process.

For the 21st century learner, developing self-direction skill is crucial for both academic activities and maintaining one's healthy lifestyle. While there are technology supports for specific self-regulated learning tasks and health monitoring, research is limited on how to support development of meta-skill of self-direction process itself. In our work, we focus on designing seamless technology infrastructure to foster self-directedness of learners. We consider learning and physical activities data as a context and DAPER (data collection-analyze-plan-execution monitoring-reflect), as a data-driven self-direction skill execution and acquisition model. We bridge Learning Analytics and Quantified-Self approaches to develop the GOAL (Goal Oriented Active Learner) system to support synchronize-visualize-analyze multisource data regarding learners' learning and physical activities. This paper proposes a measurement rubric as a basis of adaptive scaffolding for skill development during the process.

Students' learning behaviors in an online learning environment can be automatically recorded by learning systems. Such learning records provide new opportunities to model students' learning process. On the other hand, it has become more common to see students having wearable devices that assist in tracking their personal physical activities. These activity tracking can be integrated into a data-rich context for training students for developing their data-informed self-direction skills. We are building the GOAL (Goal Oriented Active Learner) system to support the development of self-direction skills using learning and health activity data. A key phase in any self-directed activity is goal setting and planning. This paper will introduce how to build a new model for self-planning and support the acquisition of planning skills in the GOAL system. We combine learners’ data from the self-directed activity and their interaction trace to build the model in the GOAL system. The modeling involves computing of trend value and degree of plan difficulty, then diagnosis of planning skills using a 5-point scoring criteria. An adaptive support is selected based on the computed score. The contribution of this work is modeling planning and promoting planning skills in a data-driven manner. Our approach grounds the theory of self-direction skills and enables learners to develop the skills in everyday life.

Current technology enables tracking of various learning and physical activities. User can use the data to analyze issues in the execution of those activities. Current work focuses on this analysis phase of data-informed self-directed activity cycle and proposes a measurement framework of the skill while learners work in a data-rich context. It is a paradigm shift to support and measure analysis skill from previous approaches which mostly rely on questionnaire-based measurements. In our approach, we emphasize the monitoring of learner’s analytical process and the automatic evaluation of the analysis results through system. Based on that, an automated measurement is carried out in the system to depict learner’s analysis skill and changes of skill. Additionally, we elaborate the framework in the context of the GOAL system which provides affordances of analysis based on physical and reading activity data.