Journal of Computer Assisted Learning最新文献

{"title":"Generative AI-Supported Creative Writing Among Primary School Students: A Study of Creative Writing Performance, Writing Self-Efficacy, and Flow State","authors":"Xin Liu,&nbsp;Yi Wen,&nbsp;Xiaozhe Yang,&nbsp;Yihe Gao","doi":"10.1002/jcal.70347","DOIUrl":"https://doi.org/10.1002/jcal.70347","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>The integration of Generative Artificial Intelligence (GenAI) into creative writing is a growing trend. However, few studies have examined primary school students' interactions with GenAI alongside their writing experiences and resulting writing products.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Objectives</h3>\u0000 \u0000 <p>This study compared creative writing performance, writing self-efficacy, and flow state across GenAI-supported, peer-collaborative and individual writing conditions. It also descriptively examined patterns of student–AI interaction during the writing activity.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>An experiment involved 78 primary school students in China assigned to three conditions: GenAI-supported writing (<i>n</i> = 20), peer collaboration (<i>n</i> = 38; 19 pairs), and individual writing (<i>n</i> = 20).</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results and Conclusions</h3>\u0000 \u0000 <p>Students in the GenAI-supported writing group obtained higher creative writing performance scores than those in the peer-collaborative and individual writing groups. Interaction analysis showed that students mainly used GenAI to generate and elaborate ideas. Among episodes with observable uptake, adapted or selective incorporation was more common than direct incorporation, and no complete GenAI response was reproduced verbatim. In addition, students in the GenAI-supported writing group also reported higher writing self-efficacy than those in the individual writing group, although the difference from the peer-collaborative writing group was not significant. However, they reported the lowest flow state. Writing self-efficacy was positively associated with creative writing performance, whereas no significant association between flow state and creative writing performance was observed in this sample. These findings suggest that GenAI may serve as a resource for developing ideas in primary school creative writing.</p>\u0000 </section>\u0000 </div>","PeriodicalId":48071,"journal":{"name":"Journal of Computer Assisted Learning","volume":"42 6","pages":""},"PeriodicalIF":5.7,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145598","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Which Evaluation Strategies Help Students Spot Visual Misinformation on Social Media?","authors":"Shai Goldfarb Cohen,&nbsp;Sarit Barzilai,&nbsp;Linor L. Hadar,&nbsp;Thuraia Copti-Mshael,&nbsp;Liron Primor","doi":"10.1002/jcal.70344","DOIUrl":"https://doi.org/10.1002/jcal.70344","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>The rapid spread of visual misinformation online, intensified by advances in generative artificial intelligence, presents new challenges for adolescents' digital information literacy. Although students' difficulties in evaluating online information are well documented, little is known about how they assess different forms of visual misinformation on social media.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Objectives</h3>\u0000 \u0000 <p>This study examined the extent to which lower secondary school students can successfully identify different types of visual misinformation on social media, the evaluation strategies they use, and whether these strategies predict misinformation detection across types of visual misinformation.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>The participants were 272 students in Grades 7–9 who evaluated social media posts containing visual misinformation, including manipulated videos, manipulated images, and AI-generated images. Students rated each item's credibility and provided written justifications. Evaluation strategies were coded using a theoretically grounded scheme.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>Students were moderately successful in identifying manipulated images and videos, but struggled to detect AI-generated images, which were often perceived as credible. Students relied primarily on evaluation of visual appearance and content, while source evaluation and corroboration strategies were infrequently used. Across all visual misinformation types, examination of visual features, knowledge-based evaluation, sourcing, and corroboration strategies predicted misinformation detection. However, corroboration was the only strategy that reliably predicted identification of AI-generated images.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusions</h3>\u0000 \u0000 <p>Students mostly rely on first-hand evaluation of visual appearance and content, but these strategies are inadequate for detecting realistic AI-generated visual misinformation. Instruction should emphasize active verification strategies, particularly corroboration, and support students in coordinating multiple evaluation strategies when engaging with online visual information.</p>\u0000 </section>\u0000 </div>","PeriodicalId":48071,"journal":{"name":"Journal of Computer Assisted Learning","volume":"42 6","pages":""},"PeriodicalIF":5.7,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/jcal.70344","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145766","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multimedia Learning Outcomes and Cognitive Load: The Roles of Spatial Ability, Enjoyment, Domain, and Resource Type","authors":"Natalie Toomey,&nbsp;Misook Heo","doi":"10.1002/jcal.70345","DOIUrl":"https://doi.org/10.1002/jcal.70345","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>Multimedia learning research has primarily emphasized cognitive predictors of learning, with less attention to affective engagement despite its potential relevance to effort allocation during learning tasks.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Objectives</h3>\u0000 \u0000 <p>This study examined the joint predictive associations of spatial ability, enjoyment, gender, and multimedia type (static versus animated) with learning outcomes and perceived cognitive load in parallel across two tasks (biology and mechanics).</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>Voluntary online participation was completed by 111 university students who were randomly assigned to static or animated resources. Participants completed a spatial ability measure (Paper Folding Test), two multimedia learning tasks, post-task enjoyment ratings, and a cognitive load questionnaire assessing intrinsic, extraneous, and germane load. Multivariate multiple linear regression was used to model learning outcomes and cognitive load within each task.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results and Conclusions</h3>\u0000 \u0000 <p>Across both domains, higher spatial ability predicted higher learning performance and lower perceived extraneous cognitive load. Enjoyment was positively associated with germane cognitive load in both tasks, with additional task-specific associations with intrinsic and extraneous load. Multimedia type and gender did not significantly predict learning outcomes or cognitive load in either task. This study contributes to the literature by examining cognitive (spatial ability) and affective (enjoyment) learner characteristics jointly along with multimedia type, while modelling learning outcomes and cognitive load as complementary parallel outcomes. Through this approach, this study clarified the distinct roles of cognitive and affective learner characteristics when static and animated materials are instructionally equivalent. Specifically, it identified spatial ability as a reliable individual-difference factor associated with more efficient multimedia processing and found that enjoyment was associated with higher reported germane cognitive load. These findings highlight the importance of considering both cognitive capacity and task-level engagement in multimedia learning design.</p>\u0000 </section>\u0000 </div>","PeriodicalId":48071,"journal":{"name":"Journal of Computer Assisted Learning","volume":"42 6","pages":""},"PeriodicalIF":5.7,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145765","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From Digital Competence to Collaborative Knowledge Construction: The Role of Social Presence in Online Learning Environments","authors":"Fahimeh Keshavarzi,&nbsp;Elham Heidari,&nbsp;Zahra Rafatjoo,&nbsp;Ghasem Salimi,&nbsp;Mohammadreza Farrokhnia","doi":"10.1002/jcal.70348","DOIUrl":"https://doi.org/10.1002/jcal.70348","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>Collaborative knowledge construction (CKC) is an important process in online higher education, requiring students to communicate, share information, and collectively develop knowledge through digital environments. Although digital competence (DC) may be relevant to these processes, CKC also depends on the social and interpersonal conditions reflected in social presence (SP). However, limited research has examined how DC and SP jointly relate to CKC in online learning.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Objective</h3>\u0000 \u0000 <p>This study examined the relationships among university students' perceived DC, SP, and CKC and investigated whether SP accounted for part of the relationship between DC and CKC in online learning environments.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>A total of 341 students at a public university in Iran reported on their overall synchronous online-learning experiences during the 2024–2025 academic year. DC encompassed information and data literacy, communication and collaboration, and digital content creation, while SP included affective expression, open communication, and group cohesion. Structural equation modelling was used to examine the proposed relationships, and bootstrapping was employed to assess the indirect association between DC and CKC through SP.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>Perceived DC was positively associated with CKC (<i>β</i> = 0.27, <i>p</i> &lt; 0.001) and SP (<i>β</i> = 0.47, <i>p</i> &lt; 0.001), while SP was positively associated with CKC (<i>β</i> = 0.61, <i>p</i> &lt; 0.001). The model explained 21.6% of the variance in SP and 60.2% of the variance in CKC. DC also showed a significant indirect association with CKC through SP (<i>β</i> = 0.29, 95% BC CI [0.21, 0.38]), while the direct association remained significant, a pattern consistent with partial statistical mediation.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusions</h3>\u0000 \u0000 <p>The findings suggest that the relationship between perceived DC and CKC is partly socially embedded. Perceived DC was related to CKC both directly and indirectly through SP, indicating that digital capabilities and interpersonal conditions may jointly matter for online CKC. Online learning may therefore benefit from combining support for students' DC with collaborative activities that foster meaningful interaction, interpersonal connection, and collective knowledge development.</p>\u0000 </section>\u0000 </div>","PeriodicalId":48071,"journal":{"name":"Journal of Computer Assisted Learning","volume":"42 6","pages":""},"PeriodicalIF":5.7,"publicationDate":"2026-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/jcal.70348","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149146863","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Impact of Language Proficiency on MOOC Performance: Insights From Video Interactions and Forum Participation","authors":"Tanguy Dubois,&nbsp;Damien De Meyere,&nbsp;Patrick Watrin,&nbsp;Magali Paquot","doi":"10.1002/jcal.70338","DOIUrl":"https://doi.org/10.1002/jcal.70338","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>Massive Open Online Courses or MOOCs have the potential to democratise education by making high-quality learning opportunities accessible to students from diverse socioeconomic backgrounds and geographical locations. However, many courses on major platforms like EdX are offered exclusively in English, limiting accessibility for non-native speakers. Beyond the simple native/non-native divide, it remains unknown how different levels of English proficiency influence learners' performance in MOOCs.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Objectives</h3>\u0000 \u0000 <p>Our study investigates how language proficiency influences performance in English-taught MOOCs and modulates the effect of participants' interactions with videos and discussion forums.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>We retrieved the self-assessed language proficiency from the pre- and post-course surveys of 1268 participants for two MOOCs on EdX, viz., International Law and International Humanitarian Law. To capture these participants' interactions with videos and discussion forums, we generated measures of their use of pauses, closed captions, video playback rate, forum posts, etc. based on log data gathered automatically by EdX. Using mixed-effects beta-binomial regression, we relate these variables to their performance, calculated as the score they obtained on multiple-choice questions throughout the course.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results and Conclusions</h3>\u0000 \u0000 <p>In line with our expectations, interactions with discussion forums and especially videos have a positive effect on the performance of low-proficiency participants, which could help them overcome the language barrier they face compared to more proficient and native participants. However, the majority of low-proficiency participants do not make sufficient use of these video and discussion-forum functionalities to fully benefit from their potential.</p>\u0000 </section>\u0000 </div>","PeriodicalId":48071,"journal":{"name":"Journal of Computer Assisted Learning","volume":"42 6","pages":""},"PeriodicalIF":5.7,"publicationDate":"2026-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149146818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Investigating Synchronous Interaction Behaviours Among Open University Students: A Latent Profile Analysis and Epistemic Network Analysis","authors":"Zhu Jin,&nbsp;Liu Jinglu,&nbsp;Wang Ran","doi":"10.1002/jcal.70339","DOIUrl":"https://doi.org/10.1002/jcal.70339","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>Online education plays a pivotal role in contemporary higher education systems, particularly for distance learners at open universities, who often face challenges such as low learner engagement and perceived social isolation. However, the heterogeneity of interaction behaviours among open university students—a population whose learning ecology is shaped by work-study conflicts and constrained synchronous availability—remains critically underexplored.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Objectives</h3>\u0000 \u0000 <p>This study aimed to investigate synchronous and asynchronous interaction behaviours among open university students, identify distinct interaction profiles, and provide evidence-based, profile-specific insights for differentiated instructional design.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>A mixed-methods approach was employed, combining latent profile analysis (LPA) and epistemic network analysis (ENA). Unlike prior studies, this study uses LPA to identify interaction profiles grounded in theoretically derived dimensions of teacher-student and student–student interactions across synchronous and asynchronous modalities, and ENA to uncover the cognitive discourse structures underlying those profiles, thereby bridging self-reported behavioural frequencies with actual epistemic engagement. Data were collected from 372 students at a Chinese open university over 12 weeks.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>Three distinct interaction profiles emerged: (1) Teacher-student interaction enthusiasts (TIE): preferred real-time interactions with instructors; (2) selective asynchronous collaborators (SAC): Balanced synchronous and asynchronous peer collaboration; and (3) high interaction achievers (HIA): Actively engaged in all interaction types. ENA revealed varying cognitive and social engagement patterns across profiles. One-way ANOVA confirmed significant performance differences among the three groups (<i>F</i>(2, 369) = 107.14, <i>p</i> &lt; 0.001, <i>η</i>² = 0.367), with HIA &gt; SAC &gt; TIE.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusions</h3>\u0000 \u0000 <p>The findings underscore the importance of learner-centric instructional design and provide actionable profile-specific strategies: scaffolded real-time responses for TIE, flexible blended designs for SAC, and leadership roles for HIA. These evidence-based profiles advocate for inclusive digital education models that accommodate cognitive diversity in open university settings.</p>\u0000 </section>\u0000 </div>","PeriodicalId":48071,"journal":{"name":"Journal of Computer Assisted Learning","volume":"42 6","pages":""},"PeriodicalIF":5.7,"publicationDate":"2026-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149029147","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Students' Perceptions of University Preparation and Perceived Industry Relevance of AI Skills: A Study in UAE Higher Education","authors":"Zeina Hojeij,&nbsp;Fahed Bizzari","doi":"10.1002/jcal.70342","DOIUrl":"https://doi.org/10.1002/jcal.70342","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>Artificial intelligence is changing how universities teach and how graduates prepare for work. While universities are expected to support students in developing AI-related knowledge and confidence, less is known about how students themselves perceive the relevance of this preparation to the industry needs they anticipate.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Objectives</h3>\u0000 \u0000 <p>This study explores students' perceptions of how well their institutions prepare them for AI-related workplace challenges, how relevant they perceive their AI education to be to industry expectations, and which learning experiences they see as most useful for developing valued AI competencies.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>The study used a survey design with open-ended elaboration, involving 114 students (86% at a single UAE university, with a small number from other UAE and non-UAE institutions). The survey included Likert-scale, multiple-choice, and open-ended items. Data were analyzed descriptively, with gender and year-of-study comparisons tested using Mann–Whitney <i>U</i> and Kruskal-Wallis tests; open-ended responses were analyzed thematically.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results and Conclusions</h3>\u0000 \u0000 <p>Students strongly believed universities should prepare all learners for AI-shaped workplaces but reported only moderate confidence in current preparation, with no significant differences by gender or year of study. Many described a gap between theoretical learning and practical application and wanted current tools, career guidance, and ethical use. Students valued hands-on projects, industry partnerships, internships, curriculum integration, and responsible AI education. These findings reflect student perceptions rather than verified industry requirements and suggest universities need more visible, applied, industry-connected AI education alongside continued attention to ethics and critical thinking.</p>\u0000 </section>\u0000 </div>","PeriodicalId":48071,"journal":{"name":"Journal of Computer Assisted Learning","volume":"42 6","pages":""},"PeriodicalIF":5.7,"publicationDate":"2026-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149027811","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Lifelong, Place-Based and Gameful Design Perceptions on Virtual Reality Immersive Learning Experiences","authors":"Robson M. Araujo-Junior,&nbsp;Alec M. Bodzin","doi":"10.1002/jcal.70335","DOIUrl":"https://doi.org/10.1002/jcal.70335","url":null,"abstract":"&lt;div&gt;\u0000 \u0000 \u0000 &lt;section&gt;\u0000 \u0000 &lt;h3&gt; Background&lt;/h3&gt;\u0000 \u0000 &lt;p&gt;There is a need to better understand how immersive technologies, specifically virtual reality (VR), can support learning across the lifespan and in different learning settings. Prior work emphasizes the importance of empirical investigations that include diverse learner characteristics, examine how these diverse elements relate to learning outcomes, and identify mechanisms that may enable or constrain effective learning. Scholars also advocate for interdisciplinary approaches that integrate immersive technologies and thoughtful design to support meaningful learning experiences.&lt;/p&gt;\u0000 &lt;/section&gt;\u0000 \u0000 &lt;section&gt;\u0000 \u0000 &lt;h3&gt; Objectives&lt;/h3&gt;\u0000 \u0000 &lt;p&gt;This study examines how learners' perceptions of the immersive learning experience with &lt;i&gt;Watershed Explorers: Environmental Issues&lt;/i&gt; in desktop VR (dVR) and headset VR (hVR) are associated with their reported engagement and motivation across formal, non-formal, and informal learning settings and a wide age range.&lt;/p&gt;\u0000 &lt;/section&gt;\u0000 \u0000 &lt;section&gt;\u0000 \u0000 &lt;h3&gt; Methods&lt;/h3&gt;\u0000 \u0000 &lt;p&gt;Data were collected over 6 months using convenience sampling. The dataset included 151 participants aged 8–90 across formal, non-formal, and informal learning settings. The Cognitive Affective Model of Immersive Learning (CAMIL) informed construct selection, and the Technology-enhanced Lifelong Learning Ecology (TeLLE) conceptual framework guided the regression approach. Analyses examined how learners' perceived VR affordances (immersion and presence), perceived gameful design features (narrative, tutorial, and agency), perceived local context (place-based content), as well as learning settings (formal, non-formal, informal), device (dVR or hVR), and age were associated with their reported affective outcomes (engagement and motivation). Survey-mode differences (i.e., embedded survey for dVR; paper survey for hVR) align with device conditions and represent a limitation for interpreting device comparisons.&lt;/p&gt;\u0000 &lt;/section&gt;\u0000 \u0000 &lt;section&gt;\u0000 \u0000 &lt;h3&gt; Results and Conclusions&lt;/h3&gt;\u0000 \u0000 &lt;p&gt;Participants across learning settings and the sampled age range reported high perceived immersion and presence. Perceived gameful design showed the strongest associations with self-reported affective outcomes, followed by perceived VR affordances and perceived local context. Age showed a small but statistically significant positive association with affective outcomes, suggesting that lifespan factors contribute meaningfully alongside design perceptions. Perceived affective outcomes mean scores were similar for dVR and hVR, which showed negligible associations with outcomes; however, caution is advised when interpreting device differences due to survey","PeriodicalId":48071,"journal":{"name":"Journal of Computer Assisted Learning","volume":"42 6","pages":""},"PeriodicalIF":5.7,"publicationDate":"2026-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149029149","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"One Prompt Does Not Fit All: Comparing Uniform and Adaptive Code-Specific Prompting for Deductive Qualitative Coding With Large Language Models","authors":"María Jesús Rodríguez-Triana,&nbsp;Luis P. Prieto,&nbsp;Mohamed Saban,&nbsp;Juan I. Asensio-Pérez,&nbsp;Cristina Villa-Torrano,&nbsp;Inmaculada Haba-Ortuño,&nbsp;Denis Gillet","doi":"10.1002/jcal.70334","DOIUrl":"https://doi.org/10.1002/jcal.70334","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>Large language models (LLMs) have demonstrated potential for (semi-)automating the qualitative analysis of unstructured data, particularly in deductive qualitative coding using codebooks. While prior research has shown the feasibility of this technology, model performance seems to vary depending on the nature of the constructs being coded. However, existing approaches typically apply a single LLM prompting strategy across entire datasets, often of discourse transcripts or questionnaire data, using a single coding scheme or theoretical frame.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Objectives</h3>\u0000 \u0000 <p>We aim to explore whether code-specific adaptive prompting, where LLM prompts are customised based on expert coders' or data-driven rules for specific codes, outperforms uniform prompting in terms of agreement with a negotiated human coding and, secondarily, reduced computational cost when agreement is maintained.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>We apply this approach to analyse the description of 35 multimedia learning designs (comprising 758 items/activities) created by teachers in an inquiry-based learning digital platform. Two human coders and two open-weights LLMs (Llama 3.3 and GPT-OSS) coded the dataset, attending to three different pedagogical frameworks.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results and Conclusions</h3>\u0000 \u0000 <p>Our results indicate that, with both LLMs, the data-driven adaptive prompting reached similar or higher agreement with the reference point (i.e., the coding agreed by humans) than uniform prompting strategies (namely, zero-shot and few-shot with and without context), while reducing costs. These results may be especially relevant for the coding of large-scale datasets, highlighting opportunities for human–AI collaboration in deductive qualitative coding of educational data.</p>\u0000 </section>\u0000 </div>","PeriodicalId":48071,"journal":{"name":"Journal of Computer Assisted Learning","volume":"42 6","pages":""},"PeriodicalIF":5.7,"publicationDate":"2026-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149029150","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Construction and Application of a Time-Series Sentiment Analysis Model for Online Learning: A Case of Online Teacher Training","authors":"Ning Ma,&nbsp;Benchao Fan,&nbsp;Chunping Liu","doi":"10.1002/jcal.70313","DOIUrl":"https://doi.org/10.1002/jcal.70313","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>Online teacher training has emerged as a crucial pathway for professional development. However, the spatiotemporal separation inherent in online learning often gives rise to emotional challenges, including loneliness, fatigue, and reduced engagement, which can compromise learning outcomes. While prior studies recognize the value of emotion analysis and feedback, they typically rely on static or snapshot measures, limiting insights into the temporal evolution of learners' affective states and constraining timely interventions.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Objectives</h3>\u0000 \u0000 <p>This study aims to develop a time-series sentiment analysis model that enables real-time, automated monitoring and feedback of teachers' emotional states in online learning environments, thereby enhancing engagement, self-regulation, and overall learning effectiveness.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>A BERT-based deep learning model was employed to classify textual sentiment, and an orthogonal polynomial regression approach captured temporal dynamics of affective states. The integrated model was embedded within an online Project-Based Learning platform under a blended education framework. A sequential within-subject quasi-experimental design was adopted, combining questionnaire surveys and semi-structured interviews, with Friedman tests and Wilcoxon signed-rank post hoc comparisons used to assess the impact of time-series emotional feedback on learners' online learning outcomes.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results and Conclusions</h3>\u0000 \u0000 <p>The proposed model significantly improves online teacher training by sustaining engagement, reducing attrition risk, and supporting adaptive cognitive-emotional regulation. Real-time visualization and automated feedback facilitate timely interventions, enabling teachers to adjust strategies and maintain reflective learning practices. These findings demonstrate the potential of time-series sentiment analysis to inform data-driven, emotion-aware design of online professional development, advancing both methodological innovation and practical applications in teacher training.</p>\u0000 </section>\u0000 </div>","PeriodicalId":48071,"journal":{"name":"Journal of Computer Assisted Learning","volume":"42 5","pages":""},"PeriodicalIF":5.7,"publicationDate":"2026-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148943718","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}