Bowen Liu

Background: Generative artificial intelligence has expanded the possibilities for teaching agents to provide instructional explanations and feedback. However, limited research has examined how such support can be integrated into teacher‐guided classroom instruction and how students experience and engage with this form of teaching. Objectives: This study aimed to iteratively design and refine teaching agents within a teacher‐guided undergraduate classroom and to examine changes in students' sense of presence and learning engagement across three design iterations. Methods: A single‐group design‐based research study was conducted in a 10‐week Introduction to Educational Technology course. 36 first‐year undergraduate students majoring in educational technology participated in three design iterations: gesture‐supported presentation, teacher‐mediated real‐time Q&A, and individualized text‐based Q&A. Data were collected through questionnaires, classroom observations, student feedback, question records, and discussion texts. Repeated‐measures analyses were used to examine changes in students' sense of presence and learning engagement, and epistemic network analysis was used to examine changes in co‐occurrence patterns among presence‐related codes in students' post‐class reflections. Results and Conclusions: The findings show that students' sense of presence significantly increased across the three iterations, whereas learning engagement did not show a statistically significant change, although the descriptive means showed a small upward pattern. Epistemic network analysis identified changes in co‐occurrence patterns among presence‐related codes in students' post‐class reflections. The strongest connections involved direct instruction, affective expression, and resolution in the first iteration, and affective expression with direct instruction, exploration, and integration in the third iteration. Given the single‐group sequential design, these findings should be interpreted as observed changes during the iterative implementation process rather than causal evidence for the independent effects of specific design strategies. The study contributes process‐oriented evidence and design considerations for integrating teaching agents into teacher‐orchestrated human‐AI collaborative classrooms.

1. The emergence of generative artificial intelligence (GenAI) tools is reshaping the landscape of learning in education (Cooper, 2023; Feldman-Maggor et al., 2025). By interpreting natural languag...

Learning burnout is an important factor affecting the quality of learning. Current research indicates that learning burnout is closely related to metacognitive ability and learning engagement, yet the influencing mechanisms among the three remain unclear. About 431 college students participate in the study. According to self-regulated learning theory and social cognitive theory, this study explores the chain mediating role of learning engagement between metacognitive ability and learning burnout. The results show that: Metacognitive ability and learning engagement negatively correlate with learning burnout, while metacognitive ability positively correlates with learning engagement. Metacognitive ability negatively predicts learning burnout through the chain mediating role of learning engagement. The findings provide theoretical support and practical guidance for reducing learning burnout among college students.

Introduction: In online learning context, achievement emotions are of great significance and exert an influence on students’ learning performance. However, the research conclusions about the impact of achievement emotions on learning performance are not uniform.Methods: Following PRISMA guidelines, we searched Web of Science, Scopus, EBSCO and Google Scholar (2012–2024). 21 studies (240 independent effect sizes) met inclusion criteria and were analyzed with a random-effects model. Moreover, multiple moderating variables were investigated in the models, such as learner levels, disciplines, online learning styles, technological types, and types of performance measure.Results: Positive emotions showed a moderate positive correlation with online learning performance (r = 0.478), whereas negative emotions showed a moderate negative correlation (r = –0.303). Technology type and performance-measure type significantly moderated these relationships, while learner level, discipline, and online learning style did not.Discussion: Findings underscore the importance of fostering positive emotions in online learning and of adopting technologies that lessen the detrimental influence of negative emotions. Implications for instructional design and future research are discussed.

Teacher support is a key factor in influencing student learning engagement. However, the influence of teacher support on learning engagement in technological settings remains to be explored. The research aims to examine how teacher support correlates with online learning engagement, as well as to elucidate the mediation role of self-efficacy. It employed a Structural Equation Modeling (SEM) methodology, with a sample size of 671 college students. Results showed that teacher support exerted a significant impact on online learning engagement. Self-efficacy mediated between teacher support and online learning engagement. The study holds considerable significance for enhancing the effectiveness of online learning among college students.

Self‐direction has become an important skill in the 21st century. To cultivate learners with a high level of self‐direction, it is necessary to diagnose their self‐directed learning (SDL) ability. This study diagnosed and predicted learners' SDL ability based on their actual SDL behaviours. The study was performed in a self‐directed 3D design class lasting 90 minutes. A total of 193 middle school students participated in the study. The results of the Pearson correlation analysis ( p < 0.001) showed that the reported perception of SDL ability was significantly correlated with SDL behaviours. The results of the hierarchical multiple linear regression analysis showed that the SDL behaviours explained 84.9% of the variance in SDL ability (adjusted R ² = 0.849, p < 0.001). Therefore, SDL behaviours had significant predictive effects on the reported perception of SDL ability. Moreover, based on the random forest algorithm, the study built an SDL ability prediction model with high performance (accuracy = 0.83, precision = 0.82, recall = 0.84) using SDL behaviours as features. The study provides evidence for the design of effective strategies to enhance SDL ability and promote SDL behaviours. Practitioner notes: What is already known about this topic: To cultivate learners with a high level of self‐direction, it is necessary to diagnose their self‐directed learning (SDL) ability. SDL is a combination of internal personal attributes and external autonomous behaviours. Few studies have focused on diagnosing SDL ability based on learners' external SDL behaviours occurring during the learning process. What this paper adds: The reported perception of SDL ability was significantly correlated with SDL behaviours. SDL behaviours had significant predictive effects on the reported perception of SDL ability. Based on the random forest algorithm, the study built an SDL ability prediction model with high performance using SDL behaviours as features. Implications for practice and/or policy: The findings indicate that instructors could design effective strategies to promote SDL behaviours for the purpose of enhancing learners' SDL ability. The method and process of building an SDL ability prediction model might provide a reference for related research on ability prediction with behaviours.