In this paper, we present and describe a Fraction Generator tool, which is one of the components of our mathematical musician digital learning platform, Play My Math. Teachers and students can use the Fraction Generator tool to discuss, practice, and reinforce conceptual and procedural knowledge of fractions. To that end, the tool offers customizable settings that randomly render exercises in six different categories (i.e., identifying fractions, identifying notations, comparison of fractions, arithmetic applied to fractions, least common multiple, and quantity of a fraction). Additionally, we offer an extra module called the Worksheet Demonstration that teachers can use to guide the students when working with paper-based worksheets. The rationale for the Fraction Generator came from feedback given by teachers during a study we conducted in Mexico in 2024. For designing our tool, we considered principles of technology-enhanced learning and tangible learning interfaces. Finally, to further develop our Fraction Generator initial design, we analyzed qualitative data coming from UK teachers’ feedback as part of an intervention we conducted in the UK primary education context at the beginning of 2025.

We present and describe one of the components of our mathematical musician platform, Play My Math, called the Course Creator Tool. Teachers can use this tool to integrate music in their pedagogy and create cross-curricular (i.e., music and mathematics) lesson plans. The current version of our Course Creator Tool works with eight modules (i.e., Description Block, Multimedia Block, Music Lab, Beat Maker, Fraction Lab, Music Generator, Fraction Generator suite, and Worksheet Demonstrator) that the teachers can freely organize and customize for their lessons. We have also created templates they can copy and modify to cover 17 subtopics of fractions. The need for a Course Creator Tool was derived from feedback given by teachers during a study we did in Mexico during the first quarter of 2024, and from UK teachers’ feedback coming from an intervention done during the first quarter of 2025. By closely collaborating with teachers in our design-based research iterations, we aim to increase the educational value of the Play My Math platform, empower mathematics teachers’ practice, facilitate musical integration into mathematics pedagogy, and ultimately progress towards a mathematical musician curriculum for K–12 education.

This study explores how participants of massive open online courses (MOOCs) perceive value creation within online learning environments. Drawing on the value creation framework (VCF), we developed and empirically validated a questionnaire, which was completed by 1227 learners enrolled in MOOCs offered by TU Delft. The aim was to provide deeper insight into participants’ experiences and the perceived impact of MOOCs on their personal and professional development. More specifically, this research explores the immediate, potential, applied, realized, and transformative value creation cycles. Our findings reveal significant insights into the multifaceted impacts of study behavior on learners’ perceptions. Participants reported benefits such as skill acquisition, professional development, and enhanced confidence while highlighting areas needing improvement, such as practical application opportunities and course relevance. This study highlights the importance of aligning MOOC content with learner needs and providing ongoing support to maximize the educational value that online courses can offer. These insights contribute to understanding educational value in the postdigital age, advocating for the development and support of MOOCs to foster continued personal and professional growth.

Computational Thinking (CT), particularly abstraction, is essential in engineering education, enabling students to break down complex systems into manageable parts. Abstraction helps learners focus on key elements of a problem, ignoring extraneous details. The PGK framework, suggested by Per-renet, Groote, and Kaasenbrood, defines abstraction across four cognitive levels: problem, algorithm, program, and execution. At a higher education institution that focuses on engineering education, we assessed students' abstraction skills using sorting algorithms, chosen for their foundational role and suitability for testing such skills. Our study focused on two areas: (1) the performance of computer science (CS) and non-CS students on algorithmic abstraction tasks, and (2) how factors like demographics, training, programming proficiency, and self-assessed abstraction mastery correlate with task performance. Results showed that all students, especially non-CS majors (including Engineering), need stronger skills at the algorithm, program (coding algorithms), and execution (code functionality) levels. Many non-CS students overestimated their abilities, highlighting a gap in mastery. Students with programming experience performed better, underscoring the importance of hands-on training. These findings suggest interventions for non-CS students are needed to gain experience in programming and to bridge the gap between perceived and actual skills. Future research should focus on discipline-specific curricula and long-term studies to ensure that all students develop the essential CT skills for the digital era.

Goal setting is the first and driving stage of the self-regulated learning cycle. Studies have shown that supporting goal setting is an effective means of improving academic performance among higher education students. However, doing so can be complex and resource intensive. In this study, a goal-setting conversational agent was designed and deployed to support higher education students in setting academic goals. Across 5-weeks, we tested the effects of goal-setting prompts (guided vs. unguided) and adaptive feedback (with vs. without) when delivered via a goal-setting conversational agent. We explored the effects of these supports (i.e., guidance and feedback) on students' 1) goal quality and 2) goal attainment. Findings showed that guidance and feedback combined had the largest positive effect on goal quality. They also revealed that guidance alone produced initially high-quality goals which decreased in quality overtime, whereas feedback had a delayed but cumulative effect on quality across multiple goal setting iterations. However, neither guidance nor feedback had significant effects on goal attainment, and there was no significant relationship between goal quality and attainment. This study provides insights into how a goal-setting conversational agent and adaptive feedback can be used to support the academic goal setting process for higher education students.

Learner behaviours often provide critical clues about learners' cognitive processes. However, the capacity of human intelligence to comprehend and intervene in learners' cognitive processes is often constrained by the subjective nature of human evaluation and the challenges of maintaining consistency and scalability. The recent widespread AI technology has been applied to learning analytics (LA), aiming at a more accurate, consistent and scalable understanding of learning to compensate for challenges that human intelligence faces. However, machine intelligence has been criticized for lacking contextual understanding and difficulties dealing with complex human emotions and social cues. In this work, we aim to understand learners' internal cognitive processes based on the external behavioural cues of learners in a digital reading context, using a hybrid intelligence (HI) approach, bridging human and machine intelligence. Based on the behavioural frameworks and the insights from human experts, we scope specific behavioural cues that are known to be relevant to learners' attention regulation, which is highly relevant for learners' cognitive processes. We utilize the public WEDAR dataset with 30 subjects' video data, behaviour annotation and pre–post tests on multiple choice and summarization tasks. We apply the explainable AI (XAI) approach to train the machine learning model so that human evaluators can also understand which behavioural features were essential for predicting the usage of the cognitive processes (ie, higher-order thinking skills [HOTS] and lower-order thinking skills [LOTS]) of learners, providing insights for the next-round feature engineering and intervention design. The result indicates that the dominant use of attention regulation behaviours is a reliable indicator of low use of LOTS with 79.33% prediction accuracy, while reading speed is a valuable indicator for predicting the overall usage of HOTS and LOTS, ranging from 60.66% to 78.66% accuracy, highly surpassing random guess of 33.33%. Our study demonstrates how various combinations of behavioural features supported by HI can inform learners' cognitive processes accurately and interpretably, integrating human and machine intelligence. Practitioner notes What is already known about this topic Human attention is a cognitive process that allows us to choose and concentrate on relevant information, which leads to successful learning. In affective computing, certain behavioural cues (eg, attention regulation behaviours) are used to indicate learners' attentional states during learning. What this paper adds Attention regulation behaviours during digital reading can work as predictors of different levels of cognitive processes (ie, the utilization of higher-order thinking skills [HOTS] and lower-order thinking skills [LOTS]), leveraged by computer vision and machine learning. By developing an explainable AI model, we can predict learners' cognitive processes, which often cannot be achieved by human observations, while understanding behavioural components that lead to such machine decisions is critical. It can provide valuable machine-driven insights into the relationship between humans' external and internal states in learning. Based on the frameworks spanning cognitive AI, psychology and education, expert knowledge can contribute to initial feature selection and engineering for the hybrid intelligence (HI) model development and next-round intervention design. Implications for practice and/or policy Human and machine intelligence form an iterative cycle to build a HI to understand and intervene in learners' cognitive processes in digital reading, balancing each other's strengths and weaknesses in decision-making. It can eventually inform automated feedback loops in widespread e-learning, a new education norm since the COVID-19 pandemic. Our framework also has the potential to be extended to other scenarios with digital reading, providing concrete examples of where human intelligence and machine intelligence can contribute to building a HI. It represents more systematic supports that apply to real-life practices.

The increasingly digital landscape of higher education has highlighted the importance of self-regulated learning in digital learning environments. To support this, academic goal setting is frequently used to enhance self-regulated learning in order to improve academic performance. Although many studies have explored the implementation of goal setting activities as behavioral modifiers, the implementation of goal setting across these studies is varied, and there is little consensus on the components which should be included and reported when studying goal setting activities. To provide an overview of the current state of the field, a systematic review was carried out examining studies which implemented academic goal setting activities within higher education over the last 14 years (2010–2024) to determine for whom, in what contexts, and how goal setting has been implemented. The results from the 60 included studies reveal a wide array of goal setting implementations covering many countries and academic disciplines. Overall, these implementations are highly heterogeneous, with large differences between studies in how goal setting is carried out. However, results also show a strong trend toward partial digitalization, with most studies using technology to deliver their goal setting activities, but very few adopting technologies for any further enhancements or support. Overall, the review reveals a focus on non-experimental studies exploring the content of student goals, with only a small selection testing the effect of goal setting in experimental studies. Based on these results we suggest future work focuses on testing the effect of goal setting, especially focusing on the interplay between the design of the activities and individual student needs, as well as further investigation of how emerging educational technologies can be used to scale and enhance goal setting activities.

Gamification has emerged as a promising strategy to enhance student engagement and learning outcomes in computer science education. This study uses Wenger's Value Creation Framework to evaluate and design the gamification elements in the Answers platform, a Professional Learning Network (PLN) developed at TU Delft. Using a mixed-methods approach with 372 participants, this research examines the platform's impact on learning, motivation, and social interaction. Findings indicate that the platform significantly enhances academic engagement and applied value, as students actively use it for knowledge acquisition and problem-solving. However, social connectivity remains limited, as reflected in lower scores for relatedness and potential value. Qualitative insights reveal that students primarily engage with the platform for academic support rather than networking or peer collaboration. This study contributes to e-learning practice by offering design recommendations to integrate collaborative learning elements better and foster social interaction within gamified learning environments. Additionally, it advances theoretical discussions on gamified PLNs by illustrating how Wenger's framework can be operationalized to assess value creation in digital learning networks. The findings highlight the need for a more holistic approach to gamification that extends beyond point-based rewards to include community-driven engagement mechanisms. By addressing these gaps, this research provides actionable insights for educators, platform designers, and policymakers, supporting the development of more effective gamified learning environments that balance motivation, collaboration, and engagement in online education.

Aircraft maintenance training is shifting from time-based and theory-based toward Competency-Based Training and Assessment (CBTA), as promoted by the International Civil Aviation Organization (ICAO). This transition highlights the importance of non-technical, transversal competencies (TVCs), yet their assessments remain challenging. This study explores how TVCs can be assessed effectively in the context of aircraft maintenance by establishing the design requirements and methods using focus groups. Results indicate that programmatic approaches were preferred over traditional methods, particularly the use of personal competency portfolios integrating self, peer, and instructor assessments. Qualitative rubrics defining performance standards were identified as critical to ensure objectivity, supported by instructor training in evaluation and calibration. These findings provide practical guidance for embedding TVCs within aircraft maintenance training and assessment.

Artificial Intelligence (AI) is reshaping education, learning, and instruction, yet current research in this area is fragmented, often tool-specific, and dominated by short-term perspectives. This article develops a broader research agenda for AI and Education (AI&ED), bringing together Artificial Intelligence in Education (AIED) and AI literacy within an educational ecology framing. Using a collective writing methodology, an expert panel of eleven internationally recognised scholars from various disciplines within computer and learning sciences contributed ten standalone reflections on the challenges, opportunities, and transformations of AI&ED. Two additional leading scholars provided critical commentaries to strengthen the analysis. A thematic analysis of the contributions identifies five main challenges (learning and instructional practices and curricula, access and ethics, assessment and evaluation, research capacity, and stakeholder readiness), five areas of opportunity (enhanced pedagogies, innovation in design and research, support for learning processes, critical skills, and hybrid knowledge), and four transformational themes (AI technologies and the design of education, human-AI interplay, lifelong learning, and organisation of AI&ED research). The article proposes an educational ecology research agenda across macro (policy, research ecosystem, society), meso (curricula, institutions, leadership), and micro (instructors, learners, learning processes) levels. We argue for a future-oriented, critical, and inter- or multidisciplinary approach that recognises AI as a socio-technical assemblage and sustains educational values such as equity, democracy, and human dignity in postdigital societies.

Generative AI offers potential for educational support, but often lacks pedagogical grounding and awareness of the student’s learning context. Furthermore, researching student interactions with these tools within authentic learning environments remains challenging. To address this, we present JELAI, an open-source platform architecture designed to integrate fine-grained Learning Analytics (LA) with Large Language Model (LLM)-based tutoring directly within a Jupyter Notebook environment. JELAI employs a modular, containerized design featuring JupyterLab extensions for telemetry and chat, alongside a central middleware handling LA processing and context-aware LLM prompt enrichment. This architecture enables the capture of integrated code interaction and chat data, facilitating real-time, context-sensitive AI scaffolding and research into student behaviour. We describe the system’s design, implementation, and demonstrate its feasibility through system performance benchmarks and two proof-of-concept use cases illustrating its capabilities for logging multi-modal data, analysing help-seeking patterns, and supporting A/B testing of AI configurations. JELAI’s primary contribution is its technical framework, providing a flexible tool for researchers and educators to develop, deploy, and study LA-informed AI tutoring within the widely used Jupyter ecosystem.

Self-regulated learning (SRL) is essential for academic success in higher education, yet many students struggle to effectively regulate their own learning behaviours. While goal-setting interventions can help students set high-quality goals as the foundation for their learning behaviours, additional supports are needed to help transition from goal setting into effective goal striving. This study examines the impact of monitoring and reflection supports, delivered through a scripted conversational agent, on students’ goal attainment, SRL skills, and academic performance. In this study, 84 undergraduate students were randomly assigned to one of four experimental conditions: Control, Monitoring Only, Reflection Only, or Monitoring & Reflection. Over a five-week intervention, participants engaged in weekly goal-setting activities, with additional monitoring and/or reflection prompts depending on their assigned condition. Results showed that participants in the Monitoring Only and Monitoring & Reflection conditions reported significantly higher goal attainment than those in the Reflection Only and Control groups, suggesting that monitoring plays a critical role in reinforcing goal-directed behaviour. While SRL skills improved across all conditions, no significant differences were found between groups, indicating that consistent goal setting alone may support SRL development. There were no significant effects of the intervention on academic performance. These findings highlight the immediate effectiveness of progress-monitoring activities and suggest that reflection may require longer intervention periods to have significant effects. This study supports the use of conversational agents for delivering scalable SRL support and provides insights into the design of multi-phase SRL supports in digital education.

Computational Thinking (CT) is considered a core 21st century digital skill. The aspect of assessment is crucial and knowing what, who, when, how, and where to assess is important for assessment design. In this study, we conducted an umbrella review to gain insights regarding CT assessment in higher education. In total, we analyzed 11 reviews, focusing on: (1) bibliographical and methodological characteristics of the reviews; (2) aspects relevant of assessment design, including a) assessed constructs, b) applied assessment methodologies, and c) assessment contexts. Our findings suggest an increased attention on this topic. However, hardly any reviews reasoned the selection of their review methodology, and most of the reviews did not thoroughly examine existing reviews. Regarding assessment design aspects, most reviews did not confine their scope to higher education; however, findings on interventions and educational settings show commonalities. We identified 120 unique assessed constructs and around 10 types of assessment methods. Though a combined use of distinct assessment methods is suggested in reviews, guidelines for appropriate assessment design are yet to be constructed. Based on the findings, we argue that it is necessary to explore different combinations of assessment design in various contexts to construct assessment guidelines.

Computational Thinking (CT) is vital in today's digital era, especially in Engineering Education. While no official policy or teaching framework on CT education has been established in the Netherlands, a Western European country, there have been various initiatives for the integration of CT into the curriculum. Recognizing the crucial role of teachers in CT integration, we surveyed the perceptions and intentions of teachers in tertiary education in the Netherlands. Our survey encompassed two aspects: (1) teachers' perceptions of CT, and (2) their intentions to integrate CT into pedagogical activities. 38 teachers, mostly in Engineering Education, from across the Netherlands completed the questionnaire based on the UTAUT framework. Regarding CT perceptions, our investigation reveals that teachers possess an inadequate understanding of the relationship between CT and Computer Science, have limited training experiences in CT, and hold differing opinions on when and which constructs of CT should be integrated into different domains. Concerning teachers' intentions to integrate CT, the results exhibited a strong positive correlation between performance expectancy, attitude towards CT, and behavioral intention to implement CT in learning activities. To foster the integration of CT in tertiary education, our findings suggest the need for further development of higher education teacher training programs focused on CT and its relation to CS. Additionally, there is a call for further exploration of how to enhance teachers' performance expectancy and effort expectancy.

Evaluation of educational innovation at the course level is dominated by frameworks that are linear, built on simple models, lack grounding in theory, and are generally not flexible enough to provide actionable recommendations for improvements. Moreover, the guidelines provided by existing literature are difficult to transfer to other contexts as most of these are aimed at specific innovations only. Therefore, there is a need for a new approach that is flexible enough to enable adaptation to different contexts, and that draws on the strengths of existing evaluation traditions. This paper forms a starting point for a larger research initiative for developing a comprehensive innovation evaluation framework. The context of the paper is a project based within the four universities of technology in The Netherlands (4TU) dedicated to strengthening collaboration in engineering education and research needed to take on global societal problems. The potential reach of our framework has implications for the wider field. This paper describes the results of an exploratory literature review to draw on the strengths of various evaluation theories to develop a new approach to evaluating educational innovation in courses. Value, Methods, and Use are the main components identified for this approach. Secondly, we developed a workshop to pilot this approach. Lastly, the workshop was presented to educational researchers, engineering educators, and educational advisors. We share lessons learned from the workshop and conclude with descriptions of future research to refine the framework to prepare its application in real innovation initiatives.

Agent-based training systems can enhance people's social skills. The effective development of these systems needs a comprehensive architecture that outlines their components and relationships. Such an architecture can pinpoint improvement areas and future outlooks. This paper presents ARTES: a general architecture illustrating how components of agent-based social training systems work together. We studied existing systems and architectures for training and tutoring to design ARTES and identify its essential components and interaction characteristics. ARTES comprises two core components: the agent simulation of social situations, and educational elements to provide guided learning. We link ARTES's crucial components to four primary learning theories (behaviourism, cognitivism, social cognitive theory, and constructivism) to illustrate the role of agent simulation and tutoring elements in establishing desired learning outcomes. Furthermore, we map ARTES's components against eight architectures, 43 systems and three tools to indicate the components' relevance, completeness, generalisation, and deployment potential across contexts. In addition to ARTES, the paper also contributes by identifying future improvements and research directions, such as the agent's thinking, tutoring methods, knowledge transfer, and ethical implications. We believe ARTES can help bridge the gap between virtual human simulations and impactful educational learning, offering training system developers desirable features like understandability and adaptability.

While courses are periodically updated and improved with the integration of new teaching methods and technologies, we still lack a systematic, research- informed method for evaluating educational innovations. Existing evaluations often lack contextual transparency to enable transferability to other courses and are mostly limited to student surveys. To take on this problem, we developed a framework for evaluating innovation in courses. The accompanying poster will present elements from this framework. The results presented in the accompanying poster are based on two literature reviews and a workshop. The framework prescribes an iterative process: (1) analysis of the innovation and its context of implementation, and (2) development of the evaluation plan. To apply the framework, both a formal (consultancy format) and an informal method (workshop) have been developed. Educators can use the framework to take a more scholarly approach to evaluating educational innovations for better decision-making, and to make teaching achievements more visible.

Even though the field of Computer Science (CS) affects different aspects of society, several groups of society are underrepresented, including women and nonbinary people. Children might have different learning opportunities in CS due to their project preferences. Girls are likelier to work on stories and simple programs in Scratch, whereas boys tend to create games and more complex programs. We explore whether preferences and program implementation differ between genders within a story, game and visual adventure in Hedy, a gradual textual programming language. We analysed 14,233 programs within five Hedy levels created by 2,819 users who turned 10 to 14 in 2023. We found that boys, girls and nonbinary children worked most on the game adventure. Within the individual adventures, gender differences occur in all three adventures in the most elaborate Hedy level analysed. However, for some levels, no gender differences were found. Thus, programming assignments can be created in which children of different genders work on similar programs in terms of size and number of (unique) commands used.

Despite extensive studies from the software engineering community on how naming practices influence programming behavior, the topic receives little attention in education. Prior work indicated little agreement on good naming because it depends on many factors. Students are told that "naming is important'' and "should be meaningful,'' yet its practical implementation is rarely discussed and feedback is lacking. The current work presents a dialogic teaching approach focused on teaching a critical reflection on naming practices through five activity types: (A) perceptions and experiences, (B) create names, (C) evaluate through ranking, (D) compare codes, and (E) locate a mistake. We developed, ran, and analyzed a one-hour workshop, that we present here and share our experiences, leading to recommendations for teachers. Our contribution is twofold: (1) we provide a set of (adaptable) activities and exercises for supporting deliberate naming practices, thereby assisting teachers interested in adopting naming practices into their curriculum; (2) we provide insights regarding the student perspective on naming practices, derived from the activities, revealing potential issues and opportunities in teaching the topic.