Tai-Ping Hsu

National Taiwan Normal University, Taiwan // taipinshe@gmail.com

Mu-Sheng Chen

National Taiwan Normal University, Taiwan // mushengchen946@gmail.com

Ting-Chia Hsu

National Taiwan Normal University, Taiwan // ckhsu@ntnu.edu.tw

ABSTRACT: 

While generative artificial intelligence (GenAI) such as ChatGPT offers potential for education, its widespread use by teachers often leads to inefficient interactions, and it lacks pedagogical structure. This can diminish learners’ engagement and increase their cognitive load, hindering effective professional development. This study investigated whether a Pedagogical GenAI Agents Teaching Mode (Pedagogical GenAI Agents-TM), integrating Design Thinking as a pedagogical framework, could enhance in-service teachers’ professional development compared to a conventional ChatGPT Teaching Mode (GPT-TM). A quasi-experimental design involved 76 Taiwanese in-service teachers assigned to either Pedagogical GenAI Agents-TM or GPT-TM groups, with a focus on designing Sustainable Development Goal (SDG)-related teaching activities. Quantitative and qualitative analyses revealed that, compared to GPT-TM, Pedagogical GenAI Agents-TM significantly improved teachers’ SDG learning achievement and TPACK self-efficacy, while reducing their cognitive load. The Pedagogical GenAI Agents-TM group demonstrated more collaborative AI interaction and adopted effective learning strategies, perceiving AI as a learning partner, contrasting with the GPT-TM group’s tool-centric view and less effective information processing. The study concludes that pedagogically structured AI interaction, as embodied by Pedagogical GenAI Agents-TM, more effectively overcomes the limitations of general-purpose AI for teacher professional development. Integrating established pedagogical frameworks into AI systems fosters deeper learning, enhances teacher confidence, and alleviates cognitive strain. We therefore advocate for specialized, Pedagogical GenAI Agents designed as collaborative partners rather than mere informational tools.

Keywords: 

Pedagogical GenAI agents, Teacher professional development, Cognitive load reduction, Teacher agency, TPACK

Harun Dursun

Ministry of National Education, Turkey // Department of Education, Gazi University, Turkey // hdursun3890@gmail.com

Neşe Işık Tertemiz

Department of Education, Gazi University, Turkey // tertemiz@gazi.edu.tr

ABSTRACT: 

In the current study, the process of designing a mathematics learning environment at the primary school level with WEB 2.0 tools according to the ADDIE design model is aimed. It addresses the gap in the field by taking into account learning and teaching design models in developing content for mathematics education with new generation technological tools. In this study, it was aimed to design a learning environment for the subject of natural numbers at the fourth grade level of primary school using WEB 2.0 tools according to the ADDIE design model (analysis, design, development, implementation and evaluation). In the ADDIE stages, planning was made to prepare a suitable learning environment by revealing the students’ learning deficiencies related to the subject. In addition, 24 different WEB 2.0 tools were used in the learning environment activities designed and implemented on a group of 25 fourth-grade students who had previously studied the learning objectives of the subjects but had learning deficiencies and errors. The effectiveness of the prepared and implemented design was evaluated by looking at the learning levels of the participating students. In the light of the evaluation results, the design was improved and implemented again. This application was again implemented on a new group of 24 students. The findings revealed that the students’ learning level of the subject significantly increased as a result of the implementation of the activities within the learning environment developed with WEB 2.0 tools according to the ADDIE design model.

Keywords: 

ADDIE design model, Mathematics education, Natural numbers, WEB 2.0 tools

Xiao-Fen Shi

Institute of Teacher Education and Advanced Studies, Jiangxi Normal University, Jiangxi Province, P. R. China // 1297248249@qq.com

Zhi-Xian Zhong

Institute of Teacher Education and Advanced Studies, Jiangxi Normal University, Jiangxi Province, P. R. China // Jxzzx@126.com

Ran-Xi Yan

China-Korea Institute of New Media, Zhongnan University of Economic and Law, China // 2648004215@qq.com

ABSTRACT: 

Amid the global trend toward educational digitalization, the concept of human-machine collaboration is reshaping the ecosystem of basic education. However, the digital transformation of K-12 education faces profound challenges. Uneven resource allocation has led to a pronounced stratification pattern, resulting in a widening digital divide among K-12 teachers. This divide has emerged as a critical barrier to achieving the goal of inclusive education. To address this issue, the present study adopts grounded theory with a three-level coding approach to systematically analyze the manifestations of the digital divide among K-12 teachers. Through iterative refinement using the Delphi method, an assessment framework was developed, comprising three primary dimensions, eleven secondary indicators, and thirty-five tertiary indicators. Subsequently, the Analytic Hierarchy Process (AHP) was employed for subjective weighting, while the entropy weight method was used for objective weighting, leading to the determination of integrated weight coefficients. This study provides a theoretical basis and methodological tools for the future assessment of teachers’ digital development index and for bridging the digital divide within the teaching profession.

Keywords: 

Behaviour pattern extraction and analysis, K-12 education, Teacher training, Information literacy

Hsin-Yun Wang

Taoyuan Municipal Shou Shan Senior High School, Taiwan // wanghsinyun@gmail.com

Jerry Chih-Yuan Sun

Institute of Education, National Yang Ming Chiao Tung University, Taiwan // jerrysun@nycu.edu.tw

ABSTRACT: 

This study examined how objective and subjective engagement data predict learning outcomes in virtual reality (VR) co-creation environments. Conducted at a public high school in northern Taiwan, 83 students participated in an 18-week online-merge-offline (OMO) VR co-creation course. Random Forest (RF) models compared objective indicators (e.g., EEG, logs, performance scores) with self-reported surveys. Results showed that objective data provided greater predictive accuracy and explanatory power, while subjective measures were constrained by retrospective bias and unidimensional data structure. However, incremental validity analyses revealed comparable gains in explained variance across both data types with each added engagement dimension. This suggests that although objective data excel in predictive performance, subjective measures capture complementary insights into learners’ internal states, metacognitive awareness, and contextual experiences. Moreover, multimodal data outperformed unidimensional inputs, underscoring the value of data fusion in representing the complexity of student engagement. Sentiment analysis revealed trust as the dominant emotion, along with a mix of negative feelings, underscoring the emotional dynamics of immersive learning. To enhance instructional relevance, the study calls for the development of real-time multimodal dashboards co-designed with teachers and instructional designers – ensuring that feedback is pedagogically meaningful and context-sensitive. These findings support the integration of diverse data sources to advance adaptive support and deepen understanding of engagement in immersive learning.

Keywords: 

Multimodal learning analytics, Performance prediction, Sentiment analysis, Student engagement, Virtual reality co-creation

Siu Cheung Kong

Department of Mathematics and Information Technology, The Education University of Hong Kong, Hong Kong SAR // Artificial Intelligence and Digital Competency Education Centre, The Education University of Hong Kong, Hong Kong SAR // siucheungkong@gmail.com // sckong@eduhk.hk

Luwei Ye

Artificial Intelligence and Digital Competency Education Centre, The Education University of Hong Kong, Hong Kong SAR // yealoowei@gmail.com // lye@eduhk.hk

ABSTRACT: 

Research has highlighted the importance of self-regulated learning (SRL) for personal development and lifelong learning. Therefore, it is essential to foster SRL skills in learners from a young age to prepare them to adapt to and succeed in a rapidly changing world. Generative AI (GenAI) tools offer significant potential for helping learners develop these skills through immediate, individualised feedback. This study evaluated the use of GenAI learning tools to enhance students’ SRL skills. Eighty-four senior secondary students completed a 30-hour GenAI application course and were required to reflect on their SRL skills. A mixed-methods approach was employed to evaluate the course. Pre- and post-survey comparisons indicated that the students’ SRL skills improved significantly after attending the course. A thematic analysis of their self-reflective writings revealed that the students strongly believed the course had inspired them to apply GenAI tools for acquiring subject knowledge and managing their learning. The effectiveness of this course can guide and inspire future empirical research and pedagogical practices focused on integrating GenAI tools into the educational process to develop students’ SRL.

Keywords: 

Course evaluation, Generative artificial intelligence tools, Secondary students, Self-regulated learning, Subject learning