The integration of immersive Virtual Reality (I-VR) technology in education has emerged as a promising approach for enhancing learning experiences. There is a handful of research done to study the impact of I-VR on learning outcomes, comparison of learning using I-VR and other traditional learning methods, and the impact of values such as haptic sensation, and verbal and non-verbal cues on the learning outcomes. However, there is a dearth of research on understanding how learning is happening from the perspective of the behavior of the learners in the Virtual Reality Learning Environment (VRLE). To address this gap, we developed an Interaction Behavioral Data (IBD) logging mechanism to log all the interaction traces that constitute the behavior of the learners in a Virtual Reality Learning Environment (VRLE). We deployed the IBD logging mechanism in a VRLE used to learn electromagnetic induction concepts and conducted a study with 30 undergraduate computer science students. We extract the learners' actions from the logged data and contextualize them based on the action features such as duration (Long and Short), and frequency of occurrence (First and Repeated occurrence). In this paper, we investigate the actions extracted from logged interaction trace data to understand the behaviors that lead to high and low performance in the VRLE. Using Epistemic Network Analysis (ENA), we identify differences in prominent actions and co-occurring actions between high and low performers. Additionally, we apply Differential Sequence Mining (DSM) to uncover significant action patterns, involving multiple actions, that are differentially frequent between these two groups. Our findings demonstrate that high performers engage in structured, iterative patterns of experimentation and evaluation, while low performers exhibit less focused exploration patterns. The insights gained from ENA and DSM highlight the behavioral variations between high and low performers in the VRLE, providing valuable information for enhancing learning experiences in VRLEs. These insights gained can be further utilized by the VR content developers to develop adaptive VR learning content by providing personalized scaffolding leading to the enhancement in the learning process via I-VR.

Temporality in Self-Regulated Learning (SRL) has two perspectives: one as a passage of time and the other as an ordered sequence of events. Each of these conceptions is distinct and requires independent considerations. Only a single analytic method is not sufficient in adequately capturing both these facets of temporality. Yet, most research uses a single method in temporally-focused SRL research, and those that use multiple methods do not address both aspects of temporality. We propose CTAM4SRL, a consolidated temporal analytic method which combines advanced data visualisation, network analysis and pattern mining to capture both facets of temporality. We employ CTAM4SRL in a cohort of 36 learners engaged in a reading-writing activity. Using CTAM4SRL, we were able to provide a rich temporal explanation of the interplay of the self-regulatory processes of the learners. We were further able to identify differences in SRL behaviours in high and low performers in terms of their approach to learning comprising deep and surface strategies. High performers were able to more selectively and strategically combine deep and surface learning strategies when compared to low scorers– a behaviour which was only hypothesised in SRL literature previously, but now has empirical support provided by our consolidated analytic method.

Online learning and MOOCs have become increasingly popular in recent years, and the trend will continue, given the technology boom. There is a dire need to observe learners' behavior in these online courses, similar to what instructors do in a face-to-face classroom. Learners’ strategies and activities become crucial to understanding their behavior. One major challenge in online courses is predicting and preventing dropout behavior. While several studies have tried to perform such analysis, there is still a shortage of studies that employ different data streams to understand and predict the drop rates. Moreover, studies rarely use a fully online team-based collaborative environment as their context. Thus, the current study employs an online longitudinal problem-based learning (PBL) collaborative robotics competition as the testbed. Through methodological triangulation, the study aims to predict dropout behavior via the contributions of Discourse discussion forum ‘activities’ of participating teams, along with a self-reported Online Learning Strategies Questionnaire (OSLQ). The study also uses Qualitative interviews to enhance the ground truth and results. The OSLQ data is collected from more than 4000 participants. Furthermore, the study seeks to establish the reliability of OSLQ to advance research within online environments. Various Machine Learning algorithms are applied to analyze the data. The findings demonstrate the reliability of OSLQ with our substantial sample size and reveal promising results for predicting the dropout rate in online competition. Overall, the study contributes to online learning by addressing the need to understand and predict dropout behavior in online courses. The study’s methodological triangulation, involving qualitative interviews, provides insights into such contexts' unique dynamics and challenges by utilizing a fully online team-based collaborative environment.

VR is increasingly being utilized in various domains, including education, due to its unique characteristics. Research in this area often relies on physiological sensors, eye-trackers, VR device orientation, human observers, and pre-test and post-test to collect data for quantitative studies and on questionnaires, surveys, and interviews to collect data for qualitative studies in VR Learning. However, there is a dearth of reliable data sources for studying learner behavior in VRLE, and minimal efforts have been made to automatically collect behavioral data in this context. Furthermore, there is a lack of studies that investigate learning processes through the lens of learners' dynamic interaction behavior in VRLE. To address these gaps, we have developed a real-time data collection mechanism that automatically logs learners' interaction behavior in VRLE, including timestamps. This mechanism was deployed in a room-scale VRLE called MaroonVR, and a study was conducted involving undergraduate engineering students. The main objectives of this paper are to identify differences in interaction behavior between high and low performers and to develop an optimal predictor model to predict the learning outcome using learners' interaction behavior in VRLE. Furthermore, we propose that the study's findings can be utilized to model learners' behavior in VR and to provide scaffolding and adaptive personalized VR learning content.

Embodied Cognition theory asserts that the physical body, its environment, and the interplay between them hold a pivotal role in the process of embodied learning. In comparison to traditional and computer-based learning methods, Virtual Reality (VR) enhances embodied learning, primarily owing to its capacity to offer a sense of situatedness and engage learners through physical interactions. Furthermore, the incorporation of haptic sensations in VR-based learning introduces tactile sensory memory, complementing the existing auditory and visual sensory dimensions. Despite the evident advantages of VR in facilitating embodied learning, the current body of literature has yet to delve into the dynamic behaviors exhibited by learners within Virtual Reality Learning Environments (VRLEs) during embodied interactions, and their inherent relationship with the embodiment phenomenon. In this paper, we take a significant step by integrating the Interaction Behavioral Data (IBD) collection mechanism, a development from our previous work, into a VRLE. This integration is achieved by adopting a structured framework for embodied learning activities within a VR context. Through an empirical study involving 14 participants, we meticulously logged their interaction traces using the IBD logger. Subsequently, we effectively extracted the diverse embodied learning activities carried out by these participants within the VRLE, from the interaction trace data collected in the IBD logger. Our endeavor aims to establish meaningful connections between these embodied interaction activities and the overarching concept of embodiment. These correlations, once identified, will serve as invaluable insights for VR content developers, offering clear guidelines for the design and creation of VRLEs that optimally facilitate enhanced learning experiences. Ultimately, this knowledge transfer will not only empower instructors and learners to harness VR as a potent educational tool within their regular teaching and learning practices but also foster the seamless integration of VR into mainstream education.

Artificial intelligence in education (AIED) is the use of artificial intelligence techniques and tools to enhance and support learning and teaching processes. However, AIED also raises ethical challenges related to transparency, which is the ability to understand how AIED systems make decisions that influence educational outcomes. This paper reviews the literature on transparency in AIED, and presents a set of learner centric guidelines for ensuring transparency. The paper uses the Transparency Index Framework by Chaudhry et al. (2022) as a guiding tool, and discusses the best practices for providing information, allowing oversight, and respecting rights and choices in AIED. The paper also illustrates these practices with examples of ethically and transparently designed AIED systems. The paper concludes by highlighting the trade-offs and precautions involved in ensuring transparency in AIED, and suggests future research directions. This paper contributes to the ethical discourse on transparency in AIED by providing a comprehensive overview, identifying the best practices and challenges, and proposing the learner centric guidelines.

Adaptive intelligent educational systems are gaining popularity, offering personalized learning experiences to students based on their individual needs and styles. One crucial feature of such systems is real-time personalized feedback. However, identifying real-time learning processes impacting student performance remains challenging due to data volume constraints. Current research often relies on labor-intensive human observation, which is time-consuming and not scalable. To efficiently collect real-time data, an observation tool is essential. Qualitative/Mixed Method research explores participant experiences in education, social science, and healthcare, utilizing methods like focus groups and observations. However, these methods can be labor-intensive, particularly in maintaining observation time intervals. Existing tools lack comprehensive support for education-focused focus groups and observations. To address these issues, this paper introduces the Data Logging and Organizational Tool (DLOT), a flexible tool designed for qualitative studies with human observers. DLOT offers customizable time intervals, cross-platform compatibility, and data saving and sharing options. The tool empowers observers to log timestamped data and is available on GitHub. The DLOT was validated through two studies. The first study predicted students' affective states using real-time annotations collected via DLOT, observing 30 students in each class. The second study created multimodal datasets in a computer-enabled learning environment, observing 38 students individually. A successful usability test was conducted, offering a potential solution to challenges in real-time learning process identification and labor-intensive qualitative research observation.

When learners collaborate on complex problems and open-ended tasks, the mechanism of negotiation plays a crucial role in establishing a common understanding and achieving a shared goal among them. Research has shown that negotiation improves problem-solving processes, making it an essential skill to be developed among learners. In this study, we propose a method for automating the identification of negotiation in learners' discourse during collaboration. We leverage language models like BERT, RoBERTa, and GPT2 along with traditional machine learning models like logistic regression to detect utterances of negotiation in learners' discourse while they collaboratively solve engineering estimation problem in an Open-Ended Learning Environment (OELE) called Modeling Based Estimation Learning Environment (MEttLE). Our findings suggest that our approach can accurately identify negotiation utterances with a high accuracy of 0.924 and 0.781 kappa value with a relatively smaller training set. Our method is the first step in real-time detection of negotiation, thereby enabling educators to design scaffolds and environments to help learners engage in effective negotiations.

The rapid rise of digital learning platforms has ushered in an era of educational transformation. While these platforms offer the advantage of scalability, they often fall short in facilitating meaningful interaction, which is pivotal for effective learning. Addressing this concern, our study introduces PyGuru 2.0, an innovative online learning environment for Python programming that aligns with the ICAP framework with an advanced conversational agent. We further investigate the interactions between students and a chatbot, employing a qualitative approach to comprehensively explore the diverse ways in which students interact with the chatbot. The interaction categories encompass a wide spectrum, including code assistance, error resolution, and conceptual explanation. In future, we plan to further elaborate on this coding scheme and see its impact on students’ learning outcomes.

Virtual Reality (VR) has the potential to improve learning in the education domain due to its characteristics such as multi-sensory stimuli, immersion, interaction, first-person perspectives, etc. In spite of these advantages, the literature analysis carried out has revealed that there is no literature that explored the learning processes happening in a VR environment. The learning processes can be analyzed using the data collected from the interaction behavior happening in a VR learning environment (VRLE). The existing data collection mechanisms in VR such as questionnaires, surveys, physiological sensors, and human observers do not provide data related to the interaction behavior of the learners happening in VR head-mounted displays (HMD). Hence, in this paper, we discuss a mechanism developed to log the interaction data happening in VRLE automatically in real-time and store them in a locally accessible location. The logged interaction data can be further processed to extract features and study the learning processes in VRLE.

Software conceptual design knowledge and skills are essential for graduating computer engineering students. Several approaches have been used to teach software conceptual design through various frameworks and learning environments. Our brief literature survey found that none of these systems analysed learner engagement with design elements using learning analytics. This study aims to analyse the learner engagement with design elements as they create software designs. We analysed log data of N=37 students in Think & Link, a technology-enhanced learning environment (TELE), based on the FBS design framework. The differences in student engagement are compared for two categories of learners, high and low, based on the learning gain in the pre and post-test performance. Differences and similarities in logs of actions were identified for high and low learners. Process Models were used to identify differences in the sequence of engagement with various design elements. We find high learners to identify significant design elements and create links among them. The analysis also provided insights on points for scaffolding to help low learners create meaningful designs.

In the past, unraveling learner interaction data in TELE was a challenge. However, the advent of LA has helped in uncovering latent information in log data to scaffold learning. This paper focuses on learner interaction in VeriSIM, a TELE, to teach software design diagrams. The learners’ performance in the system is used to categorize them into three groups, namely, "full scorers", "partial scorers", and "give uppers". Our analysis found that the full scorers spend a significantly higher duration per action than the give-uppers in an introductory challenge presented in the learning environment. Further analysis unravels the strategies used by consistent and inconsistent learners, and it was observed that the learner interaction strategies evolve with increasing difficulty levels as they navigate through the challenges.

In developing countries like India, proficiency in English is a desired skill since it boosts job prospects and improves economic condition. However, students' proficiency level remains inadequate, even though English is taught in schools. Mobile language learning apps like Hello English addresses this gap by providing learners with interactive activities, collaborative learning opportunities and game-based challenges to develop their English proficiency. In this context, the current paper explores three research questions that examines gender-based variation in motivation, engagement and interaction behavior of school students who use Hello English. Students' motivation level and their subsequent engagement with the app are measured through two separate questionnaires. Gender difference in students' interaction behavior is analyzed using their interaction data with the app. We found that female students are highly motivated, prefers to interact with the app more often and spend more time on app compared to male students.

The relationship between learners' cognitive and affective states has become a topic of increased interest, especially because it is an important component of self-regulated learning (SRL) processes. This paper studies sixth grade students' SRL processes as they work in Betty's Brain, an agent-based open-ended learning environment (OELE). In this environment, students learn science topics by building causal models. Our analyses combine observational data on student affect to log files of students' interactions within the OELE. Preliminary analyses show that two relatively infrequent affective states, boredom and delight, show especially marked differences among high and low performing students. Further analysis shows that many of these differences occur after receiving feedback from the virtual agents in the Betty's Brain environment. We discuss the implications of these differences and how they can be used to construct adaptive personalized scaffolds.

Researchers have long recognized the importance of using technology to support students' collaboration in learning and problem solving tasks. Recently, there has been a lot of research in capturing and characterizing student discourse and how they regulate each other when they perform learning tasks in pairs or in small groups. In this paper, our goal is to dive a little deeper into how students collaborate, and the learning behaviors they exhibit when working in pairs on a learning by modeling task, while also teaching a virtual agent in the Betty's Brain system. We report the results of a quasi-experimental study, where students were divided into two groups: one group worked in pairs and the other group worked individually. The results illustrate that students in the collaborative group built more correct causal maps than those working individually, and their pre-post test results show significantly higher learning gains in the science content. A differential sequence mining algorithm applied to their action sequences captured in log files showed differences in the learning behaviors between the two groups. The differences imply that the collaborative groups were better at debugging their evolving causal maps than the students who worked individually.

In Intelligent Tutoring Systems, affect-based computing is an important research area. Common approaches to deal with the affective state identification are based on input data from external sensors such as eye-tracker and EEG, as well as methods based on mining of ITS log data. Sensor based methods are viable in laboratory settings but they are tough to implement in real-world scenario which might cater to a large number of students. In our research, we create a mathematical model of frustration based on its theoretical definition. We identify the variables in the model by applying the theoretical definition of frustration to the ITS log data. This approach is different from existing data mining techniques, which use correlation analysis with labeled data. We apply our model to Mindspark, a commercial maths Intelligent Tutoring System, used by several thousand students. We validate our model with human observations of frustration.