The application of extended reality (XR) technology in education has been growing for the last two decades. XR offers immersive and interactive visualization experiences that can enhance learning by making it engaging. Recent technological advances have led to the availability of high-quality and affordable XR headsets. These advancements have spurred a wave of research focused on designing, implementing, and validating XR educational interventions. Limited literature focuses on the recent trends of XR within science, technology, engineering, and mathematics (STEM) education. Thus, this paper presents an umbrella review that explores the exploding field of XR and its transformative potential in STEM education. Using six online databases, the review zoomed in on 17 out of 1972 papers on XR for STEM education, published between 2020 and 2023, following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines. The results highlighted the types of XR technology applied (i.e., virtual reality and augmented reality), the specific STEM disciplines involved, the focus of each study reviewed, and the major findings from recent reviews. Overall, the educational benefits of using XR technology in STEM education are apparent: XR boosts student motivation, facilitates learning engagement, and improves skills, for example. However, using XR in education still has challenges that must be addressed, such as the physical discomfort of the learner wearing the XR headset and technical glitches. Besides revealing trends of using XR in STEM education, this umbrella review encourages reflection on current practices and suggests ways to apply XR to STEM education effectively.

Examining the effectiveness of machine learning techniques in analyzing engineering students’ decision-making processes through topic modeling during simulation-based design tasks is crucial for advancing educational methods and tools. Thus, this study presents a comparative analysis of different supervised and unsupervised machine learning techniques for topic modeling, along with human validation. Hence, this manuscript contributes by evaluating the effectiveness of these techniques in identifying nuanced topics within the argumentation framework and improving computational methods for assessing students’ abilities and performance levels based on their informed decisions. This study examined the decision-making processes of engineering students as they participated in a simulation-based design challenge. During this task, students were prompted to use an argumentation framework to articulate their claims, evidence, and reasoning, by recording their informed design decisions in a design journal. This study combined qualitative and computational methods to analyze the students’ design journals and ensured the accuracy of the findings through the researchers’ review and interpretations of the results. Different machine learning models, including random forest, SVM, and K-nearest neighbors (KNNs), were tested for multilabel regression, using preprocessing techniques such as TF-IDF, GloVe, and BERT embeddings. Additionally, hyperparameter optimization and model interpretability were explored, along with models like RNNs with LSTM, XGBoost, and LightGBM. The results demonstrate that both supervised and unsupervised machine learning models effectively identified nuanced topics within the argumentation framework used during the design challenge of designing a zero-energy home for a Midwestern city using a CAD/CAE simulation platform. Notably, XGBoost exhibited superior predictive accuracy in estimating topic proportions, highlighting its potential for broader application in engineering education.

Analyzing the integration of scientific knowledge to guide decision-making processes supporting design challenges in engineering education is critical. However, effectively engaging in informed design decision-making processes is challenging, particularly in the context of online education. Simulation-based learning can bring authentic design practices to online education, but effective guidance and scaffolding must be provided to learners. Therefore, this research investigates the implications of integrating simulation-based learning with an argumentation framework to guide students in incorporating scientific knowledge into their design decisions, particularly in designing energy-efficient housing. This study took place during online learning due to the COVID-19 pandemic and was implemented within a first-year engineering technology undergraduate course. It aimed to analyze students’ decision-making processes when designing a zero-energy home for a Midwestern city using Aladdin, an integrated CAD/CAE platform that can be used to design a structure and simulate its function within a single system. This study investigates how students informed their decision-making processes in design for energy-efficient homes and the recurring trends in students’ designs related to economic decision making and energy science. The overall results show how cost constraints significantly influenced students’ observation and argumentation processes during their design challenge, highlighting the pivotal role of economic considerations in shaping their decision making. Moreover, the findings underscore the importance of holistic approaches in providing insights into teaching strategies for online learning, particularly in navigating the intersection of scientific and economic factors in design challenges.

This study examines the use of extended reality (XR) in helping students with conceptual comprehension of artificial intelligence (AI) concepts, specifically neural networks (NNs) and handwritten digit recognition. Using a multi‐methods approach, this study assesses student performance and understanding of such concepts. Student participants ( N = 29) engaged in an XR environment designed to teach NNs and completed in‐lesson assessments consisting of multiple‐choice questions and open‐ended questions. Quantitative data were analyzed using the k ‐means clustering method to classify performance levels based on the accuracy of the answers. The elbow approach determined the number of clusters, and the average silhouette score showed the cluster quality after clustering. Qualitative data underwent thematic analysis to identify challenges in handwritten digit recognition. Results showed that the accuracy of the students' responses ranged from 17% to 100% and could be classified into three groups, and that factors like handwriting clarity, digit placement, and writing style significantly impacted the accuracy of handwritten digit recognition. The findings suggest the potential of using XR for supporting learning and engagement in studying AI concepts. Future research is encouraged to apply XR across various education levels and explore broader AI concepts. This study contributes to the literature on applying XR in computer science education by providing insights into how XR can enhance conceptual comprehension of complex AI concepts like NNs.

This paper is an extended version of the best paper from the HUCAPP 2025 conference. The abstract nature of Artificial Intelligence (AI) concepts presents an educational challenge. This paper compares an immersive extended reality (XR) environment to a traditional desktop setup for teaching Neural Networks and Handwritten Digit Recognition. We analyzed differences in student engagement, user experience, and learning outcomes between the two modalities. In this comparative study, 56 participants learned about AI concepts using either an XR headset (Meta Quest 3) or a desktop computer. We collected data on usability, satisfaction, immersion, and likelihood to recommend using the System Usability Scale (SUS), User Satisfaction Questionnaire (USQ), Immersion Presence Questionnaire (IPQ), and Net Promoter Score (NPS) questionnaires. Learning outcomes were assessed via multiple-choice questions administered during the lesson. The XR group reported significantly higher engagement, immersion, satisfaction, and likelihood to recommend the system. However, this increased engagement did not translate to superior learning outcomes; performance on in-lesson questions was comparable between the two groups. XR users also identified challenges, including physical discomfort and unfamiliarity with the technology. The findings suggest that XR can increase student motivation in AI education, but this does not automatically lead to better learning performance. Usability challenges and the novelty of the technology may hinder knowledge absorption. Future work should focus on scaffolding strategies to mitigate these issues and better leverage XR’s educational potential by personalizing the learning experience.