Editorial: Recent Advances in Computer-Assisted Learning ^†

1. Introduction

Educational Technology (EdTech) is undergoing a profound transformation through three converging forces: the rapid maturation of artificial intelligence, the rise in immersive and game-based learning environments, and the large-scale phantomization of education through cloud computing, smart systems, and sustainable digital infrastructures. EdTech is no longer limited to digitizing textbooks or delivering content online; it increasingly seeks to personalize learning, anticipate learner needs, support educators with intelligent tools, and design engaging environments that address motivational and cognitive challenges. This shift reflects a broader move from treating technology as an add-on to pedagogy toward treating it as an integral component of how learning is designed, experienced, assessed, and governed.

Across contemporary research, three dominant themes stand out. First, AI and machine learning are being used to analyze learner behavior, automate feedback, generate learning content, and support decision-making. Second, game-based and immersive pedagogies aim to address declining engagement by embedding learning in interactive, experiential, and motivational environments. Third, education is increasingly organized around platforms and smart systems that integrate collaboration, mentoring, scalability, and sustainability. Together, these trends suggest that EdTech is evolving into an ecosystem of intelligent, immersive, and scalable learning infrastructures rather than isolated digital tools.

2. AI/ML in Learning

Artificial intelligence has become the most transformative force in contemporary EdTech, particularly in learning analytics, intelligent tutoring, automated assessment, and generative systems. One major strand of AI focuses on predicting learning outcomes and supporting early intervention. Contribution 1, by Canale et al., demonstrates that machine learning models trained on version control system data can predict student exam performance well before the end of a course, while also emphasizing the importance of explainable AI to enable teachers to understand and trust predictions. Similarly, Contribution 2, by Psathas et al., uses classic machine learning models to predict dropout in MOOCs, showing that behavioral traces combined with self-reported self-regulated learning data provide strong early indicators of risk. These studies position AI as an early warning and decision-support system for educators rather than as a replacement for pedagogical judgment.

Another major area is intelligent tutoring and conversational agents. Contribution 3, by Pérez-Marín et al., explores the use of pedagogic conversational agents designed to teach programming to children and shows that involving pre-service teachers in co-design significantly increases acceptance and classroom integration. This work highlights that technological sophistication alone is insufficient; social and professional acceptance by educators is critical for real impact. Therefore, conversational agents not only emerge as technical systems, but also as socio-technical artifacts that must align with teacher identity and practice.

Automated assessment is another rapidly growing domain. Contribution 4, by Nakamoto et al., addresses the problem of scoring mathematical self-explanations and proposes a semi-supervised approach that uses large language models to generate synthetic training data. Their results show that carefully integrated LLM-generated data can significantly improve scoring accuracy, although excessive synthetic data can degrade performance. This work illustrates how generative AI can act as a data amplifier in educational contexts where high-quality labeled data is scarce.

Generative AI has also become a subject of critical reflection in its own right. Contribution 5, by Montenegro-Rueda et al., conducted a systematic review of the existing research on the use of ChatGPT in education, showing its overall positive impacts on personalization, feedback, writing support, and collaboration, but also identifying major challenges such as a lack of teacher training, ethical risks, hallucinations, and misuse. Their analysis reveals three major research clusters: the changing role of teachers, the impact of chatbots on students, and the socio-technical context of adoption. This not only positions generative AI as a technical innovation, but also as a force that can reshape educational roles, assessment practices, and academic integrity.

AI also has relevance at the infrastructure level. Contribution 6, by Govea et al., shows that integrating AI with cloud computing significantly improves scalability, reduces administrative errors, and enables personalized content delivery in educational platforms. This demonstrates that AI is not only being increasingly embedded in learning activities, but also in the operational backbone of education.

Together, these studies portray the use of AI/ML in learning as moving from experimentation toward systemic integration. AI supports prediction, feedback, content generation, assessment, and platform management. However, across these works, a consistent message emerges: the success of AI in education depends less on algorithmic sophistication than on transparency, teacher training, ethical governance, and thoughtful pedagogical integration.

3. Game-Based and Immersive Pedagogy

Parallel to the rise of AI is a strong focus on engagement, motivation, and experiential learning through games and immersive technologies. Contribution 7, by Nadeem et al., shows that digital game-based learning significantly improves student engagement and motivation compared to traditional online activities, although features such as leaderboards can motivate some students while demotivating others. This highlights that gamification is not universally beneficial; its effects are mediated by learner differences and design choices.

Contribution 8, by Karavidas et al., directly compared the use of serious games for self-assessment in web technologies with the use of online quizzes, finding that while serious games increase engagement and the number of questions attempted, they do not necessarily improve learning outcomes compared to quizzes. This challenges the assumption that more engaging formats automatically yield better learning outcomes and underscores the need to carefully align game mechanics with learning objectives.

Immersive technologies extend this engagement agenda further. Contribution 9, by Esteves et al., provides a systematic review of immersive virtual reality applications for English learning, identifying several categories of design recommendations related to learning, interaction, and sensory experience. Their work shows that immersion and presence can support vocabulary retention, motivation, and contextual learning, but their effectiveness depend heavily on the quality of the designs and their alignment with pedagogical goals.

These game-based and immersive approaches also intersect with other EdTech domains. Contribution 5, by Montenegro-Rueda et al., notes that ChatGPT and similar tools are sometimes used to support creative, interactive, and collaborative activities that can be integrated into game-like learning designs. Likewise, the use of immersive VR for language learning often incorporates gamified elements such as exploration, tasks, and narrative progression, as shown in Contribution 9.

Across these studies, game-based and immersive pedagogy emerges less as a single method and more as a design philosophy focused on experience, motivation, and participation. However, evidence also shows that engagement alone is insufficient. Effective learning requires careful alignment between game mechanics, immersive features, and cognitive learning goals. The challenge for EdTech is therefore not simply to make learning fun, but to make engagement pedagogically meaningful.

4. Education Platforms and Smart Systems

The third major theme concerns the infrastructure that supports learning: platforms, smart environments, and sustainable systems. Contribution 10, by Nguyen et al., proposes a smart education system that integrates collaborative learning and e-mentoring and demonstrates, through empirical evaluation, that such systems can support diverse learners by combining peer interaction with professional guidance. Their model frames smart education around three elements—smart learners, smart pedagogy, and smart environments—with technology acting as an enabler of flexible learning pathways.

Govea et al.’s work on AI and cloud computing demonstrates how platforms can scale to large user bases while maintaining personalization and reducing administrative burden, as shown in Contribution 6. This reflects a broader shift toward platform-based education in which learning is delivered through integrated ecosystems rather than isolated tools.

Sustainability is also becoming a central concern. Contribution 11, by Mikroyannidis et al., describes the OpenLang Network, a sustainable online platform for language learning across Europe that combines open educational resources, community participation, and long-term governance strategies. Their evaluation shows that sustainability is not only technical, but also social, and depends on openness, community engagement, and institutional support.

In addition, AI intersects with platform design. Predictive analytics from Canale et al. and Psathas et al. in Contribution 1 and 2 can be embedded in platforms to monitor progress and flag risks. Generative tools such as ChatGPT are increasingly being integrated into platforms as learning companions or content generators, as discussed by Montenegro-Rueda et al. in Contribution 5. Thus, platforms are becoming intelligent environments that combine analytics, interaction, automation, and collaboration.

These studies show that the future of EdTech is not just about better tools, but also about better systems: scalable, intelligent, collaborative, and sustainable environments that support diverse learners and long-term educational goals.

5. Conclusions: Toward Intelligent, Immersive, and Scalable Education

Taken together, the eleven papers in this Special Issue depict EdTech as a field moving rapidly toward integrated learning ecosystems. AI and machine learning are reshaping assessment, feedback, prediction, and content generation. Game-based and immersive pedagogies address engagement and experience, but require careful pedagogical design. Platforms and smart systems provide the infrastructure that enables large-scale, personalized, and sustainable education.

Across all themes, a common thread emerges: technology alone does not improve education. Impact depends on design quality, teacher training, ethical governance, and alignment with pedagogy. Whether through AI models predicting dropout, VR environments teaching languages, serious games assessing knowledge, or smart platforms supporting collaboration, the future of EdTech lies in the thoughtful integration of intelligence, immersion, and infrastructure. This Special Issue not only maps the current practices, but also the conceptual foundations for the next generation of educational systems—systems that are adaptive, engaging, and socially responsible.