Special Sessions

Yet this transformation is outpacing the governance structures needed to keep it accountable, equitable, and pedagogically sound. As institutions embed AI into OER workflows, they increasingly face unresolved questions: Which regulatory regime applies when an institution deploys, rather than develops, AI tools (e.g., under the EU AI Code of Practice)? How should provenance, authorship, and machine-generated content be marked and disclosed in resources intended for open reuse? What competencies do faculty need to co-create AI-supported OER responsibly, and how can adoption, usability, and pedagogical acceptance be measured rather than assumed? This Special Session responds directly to the EDUCON 2027 theme of engineering digital education ecosystems that harness AI responsibly.

It brings together governance frameworks, empirical toolkits, and institutional experience from ongoing initiatives in Europe and Latin America, including work on lifecycle models for AI-OER governance, policy toolkits for decision-makers, and co-creation protocols currently being piloted with faculty, to surface a shared, evidence-based agenda for the field. The session welcomes contributions examining governance frameworks, ethical and regulatory compliance, curation criteria for AI tools used in OER production, faculty development models, and empirical evaluations of AI-supported OER adoption. Particular emphasis is placed on knowledge commons governance and Responsible Research and Innovation (RRI) approaches that can travel across institutional and regional contexts, including under-resourced and Global South settings, where governance gaps and access gaps often compound one another.

By convening researchers, policymakers, and practitioners working at the intersection of AI governance and open education, the session aims to consolidate a research agenda that is both academically rigorous and operationally actionable: one that gives engineering education institutions concrete criteria, instruments, and reference models for adopting AI in OER production without sacrificing openness, equity, or educational quality. The session will close with a moderated discussion on open research questions and opportunities for cross-institutional collaboration.

Contributors

• Virginia Rodés Paragarino, Spain
• Leonardo David Glasserman Morales, Mexico
• Inés Álvarez-Icaza Longoria, Mexico
• Rasikh Tariq, Mexico
• Regina Motz, Uruguay
• Benito Javier Martínez-Briones, Mexico
• Nicia Guillén-Yparrea, Mexico

However, much of this information remains descriptive rather than actionable. Although learning analytics can identify patterns of engagement, achievement and academic risk, educators often lack sufficient explanations to determine which interventions are appropriate and whether these interventions improve learning. This Special Session examines how explainable and human-centred artificial intelligence (AI) can transform learning analytics into meaningful pedagogical action in computing and engineering education.

Alignment with EDUCON 2027: The session directly supports EDUCON 2027's theme, Engineering Digital Education Ecosystems: Harnessing AI, XR, Cybersecurity, and Emerging Technologies for the Future of Education. It focuses particularly on intelligent learning ecosystems, responsible AI, data-driven educational decision-making and innovative assessment. By connecting educational data, interpretable AI models, adaptive interventions and human oversight, the session addresses how interconnected digital systems can support personalised, inclusive and effective engineering education. Particular attention will be given to the development of engineering competencies, including computational thinking, problem-solving, design, practical skills and collaborative learning.

Rationale for a Distinct Special Session: While existing AI-in-education research frequently concentrates on predictive accuracy, automated content generation or individual learning technologies, comparatively less attention is devoted to how analytical insights translate into explainable, context-sensitive educational interventions and demonstrable learning improvements. This session addresses that connection through three interrelated dimensions: understanding learner behaviour, explaining AI-supported recommendations and evaluating the resulting pedagogical actions. Its distinctive contribution lies in integrating these dimensions within authentic engineering education environments, rather than examining them as isolated technological developments. This focus also creates a forum for discussing learner agency, educator trust, algorithmic fairness, privacy and the responsible governance of educational data.

Expected Research Contributions and Impact: This session welcomes empirical investigations, design-based research, comparative evaluations, longitudinal studies and evidence-informed educational case studies. Contributions may explore explainable learner modelling, intelligent tutoring, adaptive feedback, multimodal learning analytics, human-AI collaboration and AI-assisted competency assessment. Particular emphasis will be placed on evaluating educational effectiveness, the interpretability of AI recommendations, teacher and learner experiences, and transferability across institutions. By connecting technical innovation with educational evidence, the session aims to establish practical design principles, identify methodological challenges and stimulate interdisciplinary collaboration among engineering educators, Computer Science researchers and educational technology specialists.

Contributors

• Dr. Muhammad Aamir
• Mr. Arslan Amjad
• Dr. Aamir Anwar, United Kingdom
• Dr. Waqar Asif
• Dr. Haleem Farman
• Dr. Ijaz Ul Haq
• Muhammad Irfan Khalid
• Dr. Muhammad Zakir Khan
• Dr. Shahzad Latif
• Dr. Ali Akbar Forouzesh Nejad, United Kingdom
• Dr. Ikram Ur Rehman, UAE
• Dr. Syed Muhammad Usman

This raises an increasingly important educational question: how does AI mediation change the cognitive and metacognitive activity undertaken by learners, and with what consequences for learning? In engineering education, where students engage extensively in disciplinary problem solving, design, modelling, coding, and decision-making, this question has particular relevance. Recent empirical studies indicate that improved GenAI-supported task performance does not necessarily correspond to equivalent gains in learning, retention, transfer, or metacognitive engagement.

This distinction directs attention beyond the quality of AI-supported outputs towards the cognitive processes through which learning occurs. Recent theoretical work conceptualises this problem more explicitly, for example by proposing that GenAI can make the locus of generative activity less transparent when activities previously performed by learners are generated, prompted, or supported externally. The distribution of cognitive activity between learner and AI showcases in arguing that the educational consequences of delegation depend on whether the supported activity is peripheral to, or constitutive of, the intended learning process. Metacognition adds a further dimension: learners must monitor and regulate not only their own understanding and strategies, but also how and when AI support is sought, interpreted, evaluated, and acted upon.

This special session focuses on cognitive and metacognitive processes in AI-mediated learning in engineering education. It invites theoretical, empirical, methodological, and design-oriented research examining how different forms of AI support influence learners' reasoning, problem solving, explanation, evaluation, monitoring, and metacognitive control; how cognitive activity is distributed across learner and AI; and how these processes relate to learning, retention, and transfer. Particular interest is given to studies that move beyond AI use, perceptions, or output quality to provide evidence about the cognitive and metacognitive activity learners engage in under different forms of human–AI interaction.

Conributors

• Nam Aghaee
• Richard Davis
• Iosif Gidiotis
• Partick Hanwell
• Stefan Hrastinski
• Malin Jansson
• Thashmee Karunaratne
• Christina Keller
• Elizabeth Keller
• Marcus Lithander
• Anna Overgaard Markman
• Bente Nørgaard
• Jalal Nouri
• Arnold Pears
• Stefan Stenbom

Pre-university STEM outreach initiatives are essential for connecting formal education with real-world applications, inspiring young people to pursue STEMrelated pathways, and promoting technological literacy. Educators, researchers, engineers, volunteers, and professional organizations worldwide are developing innovative approaches to make STEM education more accessible, engaging, and impactful.

Building on the success of our Special Session at EDUCON 2026, which attracted 17 paper submissions and featured five accepted papers of outstanding quality, we propose to continue this initiative at EDUCON 2027. The strong interest and high quality of contributions demonstrated the relevance of this topic and the importance of providing a dedicated forum for the international pre-university STEM outreach community.

This year's Special Session aims to further strengthen this growing community by encouraging the exchange of research findings, practical experiences, and evidence-based approaches to STEM outreach. Particular attention will be given to emerging technologies, including artificial intelligence, innovative pedagogical strategies, diversity and inclusion, and methods for assessing the long-term impact of outreach initiatives.

Proposed by the IEEE Educational Activities Pre-University Education Coordinating Committee (PECC), this Special Session seeks to bring together educators, researchers, engineers, and policymakers to share successful practices, address emerging challenges, and explore opportunities for international collaboration. Inspired by the momentum established at EDUCON 2026, our broader ambition is to foster a sustainable international community of practice within IEEE, connecting individuals and organizations committed to advancing preuniversity STEM outreach. By encouraging continued knowledge exchange, collaborative research, and partnerships beyond the conference itself, we aim to strengthen the long-term impact, inclusiveness, and effectiveness of STEM outreach initiatives worldwide.

Contributors

• Stamatis Dragoumanos
• Lorena Garcia
• Debra Gulick
• Martina Holenko
• Tomislav Jagušt
• Joelyn de Lima
• Prashant R. Nair
• Navaneethakrishnan Ramanathan
• Tina Vrieler

In all these cases, the usefulness of the information depends not only on the measured value itself, but also on the reliability of the underlying data. Measurement uncertainty, sensor performance, artefacts, missing or inconsistent data, and processing choices may all affect the quality of the information available for subsequent interpretation and decision-making.

This Special Session focuses on methods and applications for assessing measurement and data reliability in biomedical and clinical contexts and on how this information can support reliable decision-making. The aim is to bring together contributions that address the evaluation and propagation of measurement uncertainty, the assessment of data quality, reliability, and confidence, and the identification of unreliable, degraded, incomplete, or heterogeneous measurements.

Particular attention will be given to how information on uncertainty, quality, and confidence may affect decision criteria, classifications, risk estimates, alarms, diagnoses, and other clinically relevant outcomes. Particular attention will also be devoted to AI-assisted measurement systems, addressing their metrological characterization and validation, uncertainty quantification, robustness, and trustworthiness. The session will explore how uncertainties arising from both measurement processes and AI models affect the reliability of estimated quantities and subsequent clinical decisions, including challenges related to model generalization and performance under changing measurement conditions.

The session welcomes methodological, experimental, and application-oriented contributions across physiological monitoring, laboratory and point-of-care testing, wearable systems, and medical imaging.. Relevant approaches may include statistical and probabilistic methods, signal and data processing, sensor and data fusion, risk-based methods, as well as artificial intelligence and machine learning techniques

Contributors

• Prof. Marco Carratù, Italy
• Prof. Claudio De Capua, Italy
• Dr. Mariacarla Valeria Lugarà, Italy
• Dr. Filippo Ruffa, Italy

This Special Session aims to explore the latest advancements in electromyography measurement systems, focusing on the entire data chain, from hardware design to signal processing and interpretation. The session will address innovative sensing technologies (including surface and high-density EMG systems), and the design of sensor networks. In addition, the emphasis will be on multimodal approaches that integrate and compare sEMG data with other data types (e.g. inertial measurement units (IMUs), kinematic and kinetic sensors, etc.) to enhance the robustness, accuracy and reproducibility of the measurements. Key topics will cover advanced signal processing methodologies for artifact mitigation, features extraction, onset and contraction burst detection, and multisensory data fusion, in both static and dynamic tasks. Furthermore, the session will emphasize data analysis approaches ranging from classical methods to artificial intelligence and machine learning techniques for signal processing and muscle pattern recognition. This session will promote interdisciplinary discussion on EMG measurement systems and their translation into clinical, biomechanical, and ergonomic practices.

Contributors

• Dr. Roberto Billardello, Italy
• Prof. Gianni Cerro, Italy
• Dr. Carissimo Chiara, GB
• Serena Pizzocaro, Italy
• Dr. Cecilia Provenzale, Italy