AHCI RESEARCH GROUP
Publications
Papers published in international journals,
proceedings of conferences, workshops and books.
OUR RESEARCH
Scientific Publications
How to
Here you can find the complete list of our publications.
You can use the tag cloud to select only the papers dealing with specific research topics.
You can expand the Abstract, Links and BibTex record for each paper.
You can use the tag cloud to select only the papers dealing with specific research topics.
You can expand the Abstract, Links and BibTex record for each paper.
2025
Wei, X.; Chen, Y.; Zhao, P.; Wang, L.; Lee, L. -K.; Liu, R.
In: Interactive Learning Environments, 2025, ISSN: 10494820 (ISSN).
Abstract | Links | BibTeX | Tags: 5E learning model, generative artificial intelligence, Immersive virtual reality, Pedagogical agents, primary students, Science education
@article{wei_effects_2025,
title = {Effects of immersive virtual reality on primary students’ science performance in classroom settings: a generative AI pedagogical agents-enhanced 5E approach},
author = {X. Wei and Y. Chen and P. Zhao and L. Wang and L. -K. Lee and R. Liu},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-105007642085&doi=10.1080%2f10494820.2025.2514101&partnerID=40&md5=94fee41fcdce74ebb9e91c6430ed9507},
doi = {10.1080/10494820.2025.2514101},
issn = {10494820 (ISSN)},
year = {2025},
date = {2025-01-01},
journal = {Interactive Learning Environments},
abstract = {Immersive virtual reality (IVR) holds the potential to transform science education by offering opportunities to enhance learners’ engagement, motivation, and conceptual understanding. However, the integration of generative AI pedagogical agents (GPAs) into IVR environments remains underexplored. Specifically, the application of GPAs as a scaffold within the framework of the 5E learning model in science education has not been fully examined. To address these gaps, this study explored the impact of a GPA-enhanced 5E (GPA-5E) learning approach in IVR on primary students’ academic achievement, self-efficacy, collective efficacy, and their perceptions of the proposed method. Adopting a mixed-methods design, eighty sixth-grade students from two complete classes were assigned to either an experimental group engaging IVR science learning with a GPA-5E approach or a control group following the traditional 5E method. The results indicated that the GPA-5E approach in IVR science learning significantly improved students’ academic achievement, self-efficacy, and collective efficacy compared to the traditional method. Students in the experimental group also reported positive perceptions of the GPA-5E method, emphasizing its benefits in IVR science learning. These findings underscore the potential of integrating GPA-enhanced scaffolds within IVR environments to enrich pedagogical strategies and improve student outcomes in science education. © 2025 Informa UK Limited, trading as Taylor & Francis Group.},
keywords = {5E learning model, generative artificial intelligence, Immersive virtual reality, Pedagogical agents, primary students, Science education},
pubstate = {published},
tppubtype = {article}
}
Immersive virtual reality (IVR) holds the potential to transform science education by offering opportunities to enhance learners’ engagement, motivation, and conceptual understanding. However, the integration of generative AI pedagogical agents (GPAs) into IVR environments remains underexplored. Specifically, the application of GPAs as a scaffold within the framework of the 5E learning model in science education has not been fully examined. To address these gaps, this study explored the impact of a GPA-enhanced 5E (GPA-5E) learning approach in IVR on primary students’ academic achievement, self-efficacy, collective efficacy, and their perceptions of the proposed method. Adopting a mixed-methods design, eighty sixth-grade students from two complete classes were assigned to either an experimental group engaging IVR science learning with a GPA-5E approach or a control group following the traditional 5E method. The results indicated that the GPA-5E approach in IVR science learning significantly improved students’ academic achievement, self-efficacy, and collective efficacy compared to the traditional method. Students in the experimental group also reported positive perceptions of the GPA-5E method, emphasizing its benefits in IVR science learning. These findings underscore the potential of integrating GPA-enhanced scaffolds within IVR environments to enrich pedagogical strategies and improve student outcomes in science education. © 2025 Informa UK Limited, trading as Taylor & Francis Group.
Wei, X.; Wang, L.; Lee, L. -K.; Liu, R.
Multiple Generative AI Pedagogical Agents in Augmented Reality Environments: A Study on Implementing the 5E Model in Science Education Journal Article
In: Journal of Educational Computing Research, vol. 63, no. 2, pp. 336–371, 2025, ISSN: 07356331 (ISSN).
Abstract | Links | BibTeX | Tags: 5E learning model, Augmented Reality, elementary science education, generative artificial intelligence, Pedagogical agents
@article{wei_multiple_2025,
title = {Multiple Generative AI Pedagogical Agents in Augmented Reality Environments: A Study on Implementing the 5E Model in Science Education},
author = {X. Wei and L. Wang and L. -K. Lee and R. Liu},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85211165915&doi=10.1177%2f07356331241305519&partnerID=40&md5=ab592abf16398732391a5dd3bd4ca7ed},
doi = {10.1177/07356331241305519},
issn = {07356331 (ISSN)},
year = {2025},
date = {2025-01-01},
journal = {Journal of Educational Computing Research},
volume = {63},
number = {2},
pages = {336–371},
abstract = {Notwithstanding the growing advantages of incorporating Augmented Reality (AR) in science education, the pedagogical use of AR combined with Pedagogical Agents (PAs) remains underexplored. Additionally, few studies have examined the integration of Generative Artificial Intelligence (GAI) into science education to create GAI-enhanced PAs (GPAs) that enrich the learning experiences. To address these gaps, this study designed and implemented a GPA-enhanced 5E model within AR environments to scaffold students’ science learning. A mixed-methods design was conducted to investigate the effectiveness of the proposed approach on students’ academic achievement, cognitive load, and their perceptions of GPAs as learning aids through using the 5E model. Sixty sixth-grade students from two complete classes were randomly assigned to either an experimental group engaged in AR science learning with a GPA-enhanced 5E approach or a control group that followed the traditional 5E method. The findings revealed that the GPA-enhanced 5E approach in AR environments significantly improved students’ academic achievement and decreased cognitive load. Furthermore, students in the experimental group reported positive perceptions of the GPA-enhanced 5E method during the AR science lessons. The findings offer valuable insights for instructional designers and educators who leverage advanced educational technologies to support science learning aligned with constructivist principles. © The Author(s) 2024.},
keywords = {5E learning model, Augmented Reality, elementary science education, generative artificial intelligence, Pedagogical agents},
pubstate = {published},
tppubtype = {article}
}
Notwithstanding the growing advantages of incorporating Augmented Reality (AR) in science education, the pedagogical use of AR combined with Pedagogical Agents (PAs) remains underexplored. Additionally, few studies have examined the integration of Generative Artificial Intelligence (GAI) into science education to create GAI-enhanced PAs (GPAs) that enrich the learning experiences. To address these gaps, this study designed and implemented a GPA-enhanced 5E model within AR environments to scaffold students’ science learning. A mixed-methods design was conducted to investigate the effectiveness of the proposed approach on students’ academic achievement, cognitive load, and their perceptions of GPAs as learning aids through using the 5E model. Sixty sixth-grade students from two complete classes were randomly assigned to either an experimental group engaged in AR science learning with a GPA-enhanced 5E approach or a control group that followed the traditional 5E method. The findings revealed that the GPA-enhanced 5E approach in AR environments significantly improved students’ academic achievement and decreased cognitive load. Furthermore, students in the experimental group reported positive perceptions of the GPA-enhanced 5E method during the AR science lessons. The findings offer valuable insights for instructional designers and educators who leverage advanced educational technologies to support science learning aligned with constructivist principles. © The Author(s) 2024.
2024
Liu, Z.; Zhu, Z.; Zhu, L.; Jiang, E.; Hu, X.; Peppler, K.; Ramani, K.
ClassMeta: Designing Interactive Virtual Classmate to Promote VR Classroom Participation Proceedings Article
In: Conf Hum Fact Comput Syst Proc, Association for Computing Machinery, 2024, ISBN: 979-840070330-0 (ISBN).
Abstract | Links | BibTeX | Tags: 3D Avatars, Behavioral Research, Classroom learning, Collaborative learning, Computational Linguistics, Condition, E-Learning, Human behaviors, Language Model, Large language model, Learning experiences, Learning systems, pedagogical agent, Pedagogical agents, Students, Three dimensional computer graphics, Virtual Reality, VR classroom
@inproceedings{liu_classmeta_2024,
title = {ClassMeta: Designing Interactive Virtual Classmate to Promote VR Classroom Participation},
author = {Z. Liu and Z. Zhu and L. Zhu and E. Jiang and X. Hu and K. Peppler and K. Ramani},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85194868458&doi=10.1145%2f3613904.3642947&partnerID=40&md5=0592b2f977a2ad2e6366c6fa05808a6a},
doi = {10.1145/3613904.3642947},
isbn = {979-840070330-0 (ISBN)},
year = {2024},
date = {2024-01-01},
booktitle = {Conf Hum Fact Comput Syst Proc},
publisher = {Association for Computing Machinery},
abstract = {Peer influence plays a crucial role in promoting classroom participation, where behaviors from active students can contribute to a collective classroom learning experience. However, the presence of these active students depends on several conditions and is not consistently available across all circumstances. Recently, Large Language Models (LLMs) such as GPT have demonstrated the ability to simulate diverse human behaviors convincingly due to their capacity to generate contextually coherent responses based on their role settings. Inspired by this advancement in technology, we designed ClassMeta, a GPT-4 powered agent to help promote classroom participation by playing the role of an active student. These agents, which are embodied as 3D avatars in virtual reality, interact with actual instructors and students with both spoken language and body gestures. We conducted a comparative study to investigate the potential of ClassMeta for improving the overall learning experience of the class. © 2024 Copyright held by the owner/author(s)},
keywords = {3D Avatars, Behavioral Research, Classroom learning, Collaborative learning, Computational Linguistics, Condition, E-Learning, Human behaviors, Language Model, Large language model, Learning experiences, Learning systems, pedagogical agent, Pedagogical agents, Students, Three dimensional computer graphics, Virtual Reality, VR classroom},
pubstate = {published},
tppubtype = {inproceedings}
}
Peer influence plays a crucial role in promoting classroom participation, where behaviors from active students can contribute to a collective classroom learning experience. However, the presence of these active students depends on several conditions and is not consistently available across all circumstances. Recently, Large Language Models (LLMs) such as GPT have demonstrated the ability to simulate diverse human behaviors convincingly due to their capacity to generate contextually coherent responses based on their role settings. Inspired by this advancement in technology, we designed ClassMeta, a GPT-4 powered agent to help promote classroom participation by playing the role of an active student. These agents, which are embodied as 3D avatars in virtual reality, interact with actual instructors and students with both spoken language and body gestures. We conducted a comparative study to investigate the potential of ClassMeta for improving the overall learning experience of the class. © 2024 Copyright held by the owner/author(s)
Sikström, P.; Valentini, C.; Sivunen, A.; Kärkkäinen, T.
Pedagogical agents communicating and scaffolding students' learning: High school teachers' and students' perspectives Journal Article
In: Computers and Education, vol. 222, 2024, ISSN: 03601315 (ISSN).
Abstract | Links | BibTeX | Tags: Adversarial machine learning, Agents communication, Augmented Reality, Contrastive Learning, Federated learning, Human communications, Human-Machine Communication, Human-to-human communication script, Human–machine communication, Human–machine communication (HMC), pedagogical agent, Pedagogical agents, Scaffolds, Scaffolds (biology), Secondary education, Student learning, Students, Teachers', Teaching, User-centered design, User-centred, Virtual environments
@article{sikstrom_pedagogical_2024,
title = {Pedagogical agents communicating and scaffolding students' learning: High school teachers' and students' perspectives},
author = {P. Sikström and C. Valentini and A. Sivunen and T. Kärkkäinen},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85202198552&doi=10.1016%2fj.compedu.2024.105140&partnerID=40&md5=dfb4a7b6c1f6352c5cc6faac213e938f},
doi = {10.1016/j.compedu.2024.105140},
issn = {03601315 (ISSN)},
year = {2024},
date = {2024-01-01},
journal = {Computers and Education},
volume = {222},
abstract = {Pedagogical agents (PAs) communicate verbally and non-verbally with students in digital and virtual reality/augmented reality learning environments. PAs have been shown to be beneficial for learning, and generative artificial intelligence, such as large language models, can improve PAs' communication abilities significantly. K-12 education is underrepresented in learning technology research and teachers' and students' insights have not been considered when developing PA communication. The current study addresses this research gap by conducting and analyzing semi-structured, in-depth interviews with eleven high school teachers and sixteen high school students about their expectations for PAs' communication capabilities. The interviewees identified relational and task-related communication capabilities that a PA should perform to communicate effectively with students and scaffold their learning. PA communication that is simultaneously affirmative and relational can induce immediacy, foster the relationship and engagement with a PA, and support students' learning management. Additionally, the teachers and students described the activities and technological aspects that should be considered when designing conversational PAs. The study showed that teachers and students applied human-to-human communication scripts when outlining their desired PA communication characteristics. The study offers novel insights and recommendations to researchers and developers on the communicational, pedagogical, and technological aspects that must be considered when designing communicative PAs that scaffold students’ learning, and discusses the contributions on human–machine communication in education. © 2024 The Authors},
keywords = {Adversarial machine learning, Agents communication, Augmented Reality, Contrastive Learning, Federated learning, Human communications, Human-Machine Communication, Human-to-human communication script, Human–machine communication, Human–machine communication (HMC), pedagogical agent, Pedagogical agents, Scaffolds, Scaffolds (biology), Secondary education, Student learning, Students, Teachers', Teaching, User-centered design, User-centred, Virtual environments},
pubstate = {published},
tppubtype = {article}
}
Pedagogical agents (PAs) communicate verbally and non-verbally with students in digital and virtual reality/augmented reality learning environments. PAs have been shown to be beneficial for learning, and generative artificial intelligence, such as large language models, can improve PAs' communication abilities significantly. K-12 education is underrepresented in learning technology research and teachers' and students' insights have not been considered when developing PA communication. The current study addresses this research gap by conducting and analyzing semi-structured, in-depth interviews with eleven high school teachers and sixteen high school students about their expectations for PAs' communication capabilities. The interviewees identified relational and task-related communication capabilities that a PA should perform to communicate effectively with students and scaffold their learning. PA communication that is simultaneously affirmative and relational can induce immediacy, foster the relationship and engagement with a PA, and support students' learning management. Additionally, the teachers and students described the activities and technological aspects that should be considered when designing conversational PAs. The study showed that teachers and students applied human-to-human communication scripts when outlining their desired PA communication characteristics. The study offers novel insights and recommendations to researchers and developers on the communicational, pedagogical, and technological aspects that must be considered when designing communicative PAs that scaffold students’ learning, and discusses the contributions on human–machine communication in education. © 2024 The Authors