Computers & Education最新文献

{"title":"Effects of a human-robot collaborative teaching approach on preschoolers’ social-emotional competence and learning behaviors","authors":"Jian-Wen Fang ,&nbsp;Ting-Ting Ji ,&nbsp;Yun-Fang Tu ,&nbsp;Gwo-Jen Hwang ,&nbsp;Di Zou ,&nbsp;Jing Chen","doi":"10.1016/j.compedu.2025.105469","DOIUrl":"10.1016/j.compedu.2025.105469","url":null,"abstract":"<div><div>Preschoolers' social-emotional competence is key to their long-term academic success. Developing this competence requires meaningful social interactions and emotional support, which the conventional teaching (CT) approach may not fully provide. This study proposed a human-robot collaborative teaching (HRCT) approach to facilitate preschoolers' social-emotional learning (SEL). A human-robot collaborative teaching system was developed to foster preschoolers' social-emotional competence through collaboration between the humanoid intelligent robot and the teacher. To evaluate the effectiveness of this approach, 96 preschoolers with an average age of 4.6 years were recruited. Of these, 44 were randomly assigned to the experimental group, which utilized the HRCT approach, while 52 were assigned to the control group, which followed the conventional teaching approach. The results indicated that the experimental group outperformed the control group in terms of social interaction skills, emotion comprehension skills, and interpersonal problem-solving skills. Significant differences were also observed in the learning behaviors of preschoolers during activities related to social interaction, emotion comprehension, and interpersonal problem solving. Further analysis of the teacher's perceptions and reflections revealed that, despite some limitations of the robot, its anthropomorphic and embodied features effectively sparked preschoolers' interest and engagement. These features contributed to enhancing their social-emotional competence by simulating and practicing SEL processes.</div></div>","PeriodicalId":10568,"journal":{"name":"Computers & Education","volume":"240 ","pages":"Article 105469"},"PeriodicalIF":10.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145217981","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Learner autonomy and mode selection in dual-mode chatbot: Students' interaction patterns and learning outcomes in an online course","authors":"Songhee Han (한송희)","doi":"10.1016/j.compedu.2025.105464","DOIUrl":"10.1016/j.compedu.2025.105464","url":null,"abstract":"<div><div>While educational chatbots are increasingly used in online courses, most research has focused on single-mode chatbots, with little attention to how students engage with multi-mode chatbots. Limited empirical studies have examined how students exercise autonomy when choosing between a general large language model (LLM)-based mode and a reference-based mode powered by retrieval-augmented generation (RAG). This study examines how students in a large-scale online course engaged with a dual-mode chatbot that allowed them to choose between general (LLM-based) and reference-based modes (LLM-RAG). Grounded in a conceptual framework linking learner autonomy, user factors, and learning outcomes, the study employed a mixed-methods explanatory sequential design using chatbot interaction logs, learning management system logs, server data, and survey responses. Findings showed that 41.3 % of the participants were unaware of or indifferent to using the mode selection feature despite each interaction beginning with a system-generated message explaining each mode's functionality. Among those aware, general mode users prioritized convenience and open-ended use, while reference-based mode users emphasized course-aligned assistance and demonstrated greater hyperlink engagement. Hyperlink use was positively correlated with social presence, teaching and cognitive presence, self-regulation, and perceived ease of use. Course completion showed a modest positive correlation with perceived ease of use but not with other variables. Interaction behaviors differed between the two chatbot modes, with exploratory and socio-emotional talk more prevalent in the general mode, and course-related sentence questions dominating the reference-based mode. These findings highlight the need for intentional design that fosters learner autonomy and aligns chatbot functionality with instructional goals.</div></div>","PeriodicalId":10568,"journal":{"name":"Computers & Education","volume":"240 ","pages":"Article 105464"},"PeriodicalIF":10.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145263659","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Tailoring educational support with graph neural networks and explainable AI: Insights into online learners' metacognitive abilities","authors":"Hongjiang Wang ,&nbsp;Peiyu Chen ,&nbsp;Jinwen Luo ,&nbsp;Yunying Yang","doi":"10.1016/j.compedu.2025.105452","DOIUrl":"10.1016/j.compedu.2025.105452","url":null,"abstract":"<div><div>Metacognition—the awareness and regulation of one's thinking processes—plays a crucial role in self-regulated learning (SRL), yet traditional educational research methods struggle to capture how metacognitive abilities manifest in actual learning behaviors. While computer-assisted learning (CAL) environments offer rich opportunities to observe these processes in action, educational researchers have typically analyzed this data using approaches that fail to connect metacognitive abilities with the complex, sequential nature of SRL behaviors. Our study bridges this gap by examining how 49 university students' metacognitive abilities shaped their learning patterns over one semester. We introduced a novel methodological approach that transforms diverse digital traces into unified graph structures, allowing us to map connections between metacognitive abilities and the planning, monitoring, and evaluation phases of SRL. Using attributed graphs, we integrated both static indicators and sequential behavioral patterns to predict metacognitive abilities with significantly higher accuracy than traditional single-data approaches, including Long Short-Term Memory (LSTM), Recurrent Neural Networks (RNN), Artificial Neural Networks (ANN), and Random Forest (RF). Through Explainable AI techniques, we revealed that high-metacognitive learners exhibited comprehension-centered, goal-oriented strategies across learning phases, while low-metacognitive learners focused primarily on task completion with limited strategic planning. These insights enabled us to develop personalized metacognitive profiles that can guide targeted educational interventions. Our approach demonstrates how advanced analytical methods can transform educational data into meaningful insights about cognitive processes, offering educators new ways to understand and support students' metacognitive development.</div></div>","PeriodicalId":10568,"journal":{"name":"Computers & Education","volume":"240 ","pages":"Article 105452"},"PeriodicalIF":10.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145020938","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Children's engagement with digital technology in educational spaces: A scoping review","authors":"Jacquelyn Harverson ,&nbsp;Chris Zomer ,&nbsp;Celine Chu ,&nbsp;Sharon Horwood ,&nbsp;Marcus Horwood ,&nbsp;Maria Nicholas ,&nbsp;Louise Paatsch","doi":"10.1016/j.compedu.2025.105454","DOIUrl":"10.1016/j.compedu.2025.105454","url":null,"abstract":"<div><div>Children's engagement in play and learning is becoming increasingly complex as educational spaces integrate digital experiences. While engagement is often seen as a multifaceted construct, there is often no clear and generally accepted definition or understanding of what is meant when the term ‘engagement’ is used. This scoping review synthesised the ways in which engagement has been conceptualised within young children's digital experiences in educational spaces. A systematic search of seven databases captured studies published between 2011 and September 2024. Studies reporting children's engagement with digital technology (e.g., in play or learning via any digital experience) that were conducted in a formal or semi-formal educational setting were included. A total of 2661 records were screened, resulting in the inclusion of 88 studies. The concept of <em>engagement</em> was explicitly defined in 18 studies (19.3 %), theorised in 8 studies (9.1 %), and measured in 31 studies (35.2 %). Theoretical foundations of engagement drew from digital and educational disciplines, primarily rooted in motivational theories. Thematic analysis demonstrated the multidimensional nature of engagement, synthesised through ten key themes drawn from the definitions and measures provided. The most prominent themes identified in definitions were <em>participation</em> (<em>n</em> = 8, 44.4 %), <em>effort</em> (<em>n</em> = 7, 38.9 %) and <em>interpersonal skills and behaviour</em> (<em>n</em> = 6, 33.3 %), while those found in measures of engagement were predominantly associated with <em>enjoyment</em> (<em>n</em> = 14, 45.2 %), <em>participation</em> (<em>n</em> = 11, 35.5 %), and <em>interest</em> (<em>n</em> = 8, 25.8 %). This review highlights the need for more consistent approaches in defining and measuring children's digital engagement.</div></div>","PeriodicalId":10568,"journal":{"name":"Computers & Education","volume":"240 ","pages":"Article 105454"},"PeriodicalIF":10.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145094090","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A systematic literature review on designing self-regulated learning using generative artificial intelligence and its future research directions","authors":"Qi Xia ,&nbsp;Qian Liu ,&nbsp;Ahmed Tlili ,&nbsp;Thomas K.F. Chiu","doi":"10.1016/j.compedu.2025.105465","DOIUrl":"10.1016/j.compedu.2025.105465","url":null,"abstract":"<div><div>A growing body of research suggests that generative AI (GenAI) can significantly enhance self-regulated learning (SRL) through its immediacy and interactivity. Nevertheless, challenges remain, including the lack of clarity regarding the mechanisms by which generative AI influences SRL, as well as the difficulties teachers encounter when trying to incorporate it into their classrooms. This systematic review study investigates how to design SRL activities using GenAI. We examined 73 articles published over the past five years, drawn from four databases: Web of Science, ProQuest, ERIC, and Scopus. This study has three major empirical findings. First, six pedagogical affordances from GenAI across the three phases of SRL—forethought, Performance, and self-reflection—are (i) creating personalized learning objectives; (ii) searching for, analyzing, and integrating resources; (iii) monitoring and evaluating progress; (iv) recommending learning strategies; (v) recording progress and providing feedback; and (vi) generating new ideas and examples. Second, popular student learning activities using GenAI in each phase are information searching in the forethought; strategies for problem-solving in the performance, and obtaining feedback and conducting self-assessments in the self-reflection. Third, two major dimensions influencing student engagement in SRL using GenAI are individual and environmental. Finally, we visualize how the three empirical findings relate to each other. Our findings help us understand how AI as a human-machine collaborative tool affects the SRL process. This study provides practical recommendations for facilitating SRL in GenAI-enhanced environments and offers guidance for the design and development of personalized GenAI learning tools.</div></div>","PeriodicalId":10568,"journal":{"name":"Computers & Education","volume":"240 ","pages":"Article 105465"},"PeriodicalIF":10.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145155577","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring characteristics of primary school students’ self-regulated learning (SRL) behaviors in human-GenAI collaborative programming learning environments: Insights from a proposed framework","authors":"Xuechun Ma ,&nbsp;Cuixin Li ,&nbsp;Jie Xu ,&nbsp;Shiyun Zhu ,&nbsp;Yan Li","doi":"10.1016/j.compedu.2025.105453","DOIUrl":"10.1016/j.compedu.2025.105453","url":null,"abstract":"<div><div>Programming education in primary school is vital for nurturing future-ready talents, yet primary school students often struggle with self-regulated learning (SRL), particularly in resource utilization and strategy regulation. Although human-generative AI (GenAI) collaborative programming learning might have the potential to enhance personalized programming education, GenAI's interplay with SRL processes remains underexplored. To address this gap, this study first proposed a SRL Behavior Analysis Framework for human-GenAI collaborative programming learning environments and then examined SRL behaviors of a group of sixth-grade students (n = 36) in such an environment using this framework, along with various learning analytics methods including cluster analysis, descriptive statistics and lag sequential analysis. The analysis yielded the following results: (1) Based on their learning performance, primary school students were identified as three distinct clusters: programming specialized unit (PSU), high performance unit (HPU), and low performance unit (LPU). (2) Regarding SRL behaviors, students prioritized self-control (65.8 %), followed by self-observation (19 %), task analysis (12.1 %), and behavior stagnation (3.2 %). (3) Students in PSU and HPU consistently adopted goal-oriented SRL strategies, whereas students in LPU exhibited passive dependence and fragmented strategy use. GenAI's facilitative effect in supporting learning correlated with users' SRL capabilities. (4) Students in PSU and HPU exhibited frequent transitions between SRL behaviors, whereas students in LPU had insufficient ability to switch strategies when facing programming difficulties. Based on these findings, this study proposed four forward-looking design recommendations: effectively integrating GenAI with the programming environment, utilizing multimodal data and AI for learning assessment and feedback, building a cluster-driven early warning mechanism, and conducting dynamic SRL analysis and guidance based on fine-grained time-series data.</div></div>","PeriodicalId":10568,"journal":{"name":"Computers & Education","volume":"240 ","pages":"Article 105453"},"PeriodicalIF":10.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145047684","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating educational technology: Consolidating across multiple impact indicators and rating systems","authors":"Natalia Kucirkova ,&nbsp;Per Henning Uppstad ,&nbsp;Richard Holeton ,&nbsp;Dan Lin ,&nbsp;Alison Clark-Wilson ,&nbsp;Antonie Chigeda","doi":"10.1016/j.compedu.2025.105467","DOIUrl":"10.1016/j.compedu.2025.105467","url":null,"abstract":"<div><div>Procurement of educational technologies (EdTech) requires a research-based rubric with multiple quality indicators that reflect the technology’s impact across various settings. We outline the development of the Multiple EdTech Impact Index (MEII) and an impact benchmarking formula for EdTech tools designed for K12 education. MEII includes fifteen indicators of impact, categorized into three evidence weights (high, medium, low) across five domains: efficacy, effectiveness, ethics, equity, and environment. It calculates a total EdTech impact score, highlighting the strength and relevance of products. MEII integrates evaluations based on diverse national EdTech certification standards and proposes a unified approach for assessment. We demonstrate its value through two models: national EdTech certification systems and a comparative database ranking EdTech tools by impact weight.</div></div>","PeriodicalId":10568,"journal":{"name":"Computers & Education","volume":"240 ","pages":"Article 105467"},"PeriodicalIF":10.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145228757","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Towards reliable generative AI-driven scaffolding: Reducing hallucinations and enhancing quality in self-regulated learning support","authors":"Keyang Qian ,&nbsp;Shiqi Liu ,&nbsp;Tongguang Li ,&nbsp;Mladen Raković ,&nbsp;Xinyu Li ,&nbsp;Rui Guan ,&nbsp;Inge Molenaar ,&nbsp;Sadia Nawaz ,&nbsp;Zachari Swiecki ,&nbsp;Lixiang Yan ,&nbsp;Dragan Gašević","doi":"10.1016/j.compedu.2025.105448","DOIUrl":"10.1016/j.compedu.2025.105448","url":null,"abstract":"<div><div>Generative Artificial Intelligence (GenAI) holds a potential to advance existing educational technologies with capabilities to automatically generate personalised scaffolds that support students’ self-regulated learning (SRL). While advancements in large language models (LLMs) promise improvements in the adaptability and quality of educational technologies for SRL, there remain concerns about the hallucinations in content generated by LLMs, which can compromise both the learning experience and ethical standards. To address these challenges, we proposed GenAI-enabled approaches for evaluating personalised SRL scaffolds before they are presented to students, aiming for reducing hallucinations and improving overall quality of LLM-generated personalised scaffolds. Specifically, two approaches are investigated. The first approach involved developing a multi-agent system approach for reliability evaluation to assess the extent to which LLM-generated scaffolds accurately target relevant SRL processes. The second approach utilised the “LLM-as-a-Judge” technique for quality evaluation that evaluates LLM-generated scaffolds for their helpfulness in supporting students. We constructed evaluation datasets, and compared our results with single-agent LLM systems and machine learning approach baselines. Our findings indicate that the reliability evaluation approach is highly effective and outperforms the baselines, showing almost perfect alignment with human experts’ evaluations. Moreover, both proposed evaluation approaches can be harnessed to effectively reduce hallucinations. Additionally, we identified and discussed bias limitations of the “LLM-as-a-Judge” technique in evaluating LLM-generated scaffolds. We suggest incorporating these approaches into GenAI-powered personalised SRL scaffolding systems to mitigate hallucination issues and improve the overall scaffolding quality.</div></div>","PeriodicalId":10568,"journal":{"name":"Computers & Education","volume":"240 ","pages":"Article 105448"},"PeriodicalIF":10.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145021041","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Can virtual reality improve social-emotional learning among adolescents? An experimental study","authors":"Qinna Feng ,&nbsp;Gege Li ,&nbsp;Yanbei Chen ,&nbsp;Feng Zhang ,&nbsp;Wenhao Li ,&nbsp;Heng Luo","doi":"10.1016/j.compedu.2025.105462","DOIUrl":"10.1016/j.compedu.2025.105462","url":null,"abstract":"<div><div>The potential of virtual reality (VR) to support social-emotional learning (SEL) has been widely acknowledged, but design precedents for VR-based SEL interventions are still lacking in the literature. The effectiveness of such interventions also remains uncertain due to insufficient empirical evidence. To bridge this research gap, the present study adopted a between-subjects design to explore the impact of a VR-based intervention on enhancing SEL and learning experience. A total of 297 seventh-grade students in China were randomly assigned to one of three conditions: VR (n = 101), face-to-face (n = 107), and control (n = 89). Participants in both the VR and face-to-face conditions engaged in collaborative tasks designed to promote SEL, while the control condition received no such intervention. The findings revealed significant improvement in overall social-emotional competencies in the VR condition, especially in the subscales of task performance, collaboration, and engagement with others. The VR intervention also promoted a stronger sense of group cohesion and enriched social experiences. These findings offer practical implications for the design and implementation of VR-based SEL interventions in formal education settings.</div></div>","PeriodicalId":10568,"journal":{"name":"Computers & Education","volume":"240 ","pages":"Article 105462"},"PeriodicalIF":10.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145094088","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using eye tracking to explore differences and influencing factors of map reading ability in different problem situation among high school students","authors":"Xinxin Li ,&nbsp;Ping Li ,&nbsp;Huiming You ,&nbsp;Qiuwen Zhou ,&nbsp;Ling Zhao ,&nbsp;Yan Huang ,&nbsp;Ya Luo","doi":"10.1016/j.compedu.2025.105466","DOIUrl":"10.1016/j.compedu.2025.105466","url":null,"abstract":"<div><div>This study uses eye tracking to explore the differences in map-reading ability among high school students in the context of subject knowledge and real-life application. Unlike existing literature, which primarily emphasizes outcome-based assessments, this study integrates process- and outcome-based indicators to quantitatively evaluate high school students' map-reading abilities across different problem situations. Additionally, it comprehensively analyzes the internal and external factors influencing students' map-reading abilities and constructs a conceptual framework to illustrate the pathway through which these factors contribute to differences in performance among the students. The results reveal that in real-life scenarios, students exhibited longer reaction times, engaged in more extensive visual searches, processed information in a complex manner, and achieved low accuracy rates. However, despite these limitations, their cognitive load did not increase significantly. Furthermore, several limitations were highlighted in traditional geography education, particularly in fostering students' spatial reasoning and practical application skills. These findings contribute to the existing research on map-reading abilities, providing a new theoretical foundation and empirical support for advancements in geography education. Moreover, they can help educators better enhance students’ map-reading ability to meet diverse and complex practical needs.</div></div>","PeriodicalId":10568,"journal":{"name":"Computers & Education","volume":"240 ","pages":"Article 105466"},"PeriodicalIF":10.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145261908","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}