Research Projects
I am a passionate researcher and innovator in educational technology and inclusive learning. My projects integrate interactive, robot-assisted learning systems designed to engage diverse learners—particularly those with specific learning difficulties (SpLD). Here, you can explore mobile and AR-enhanced tools that make learning Chinese characters more accessible and enjoyable, as well as pre-screening games that support the early identification of dyslexia, dysgraphia, and dyspraxia
This space offers a deeper look into how I combine human–computer and human–robot interaction to transform education. You will find my work on developing interactive robots—such as Kebbi and Minibo—that dance, sing, gesture, and respond, capturing attention and motivating students with varying needs. These projects stem from a strong commitment to inclusive education, grounded in research on SpLDs, accessibility, and diverse learners.
Let this section be your gateway to understanding my vision: creating effective, engaging learning experiences for every student—powered by human-centred design and emerging technologies.
Published Papers
2025
DysVis: A User-Centred Data Visualization System for Dyslexia Pre-screening
Fung K. Y., Lee, L. H., Yuan L., Fung K. C., Sin K. F., Lui, TL., Qu, H., & Song, S.
The 2025 CHI Conference on Human Factors in Computing Systems (pp. 1-18).
DysVis provides teachers with multi-level visualizations, such as performance overviews, task analyses, handwriting observations, and behavioural insights, enabling them to identify the root causes of learning difficulties.
Motivating Students with Different Needs to Learn Chinese in a Mixed-Background Classroom by Robot-Assisted Learning
Fung K. Y., Fung K.C., Lui, TL., Sin, K. F., Lee, L. H., Qu, H., & Song, S.
IEEE Transactions on Learning Technologies
Interactive design has been proven effective in enhancing students’ learning performance and engagement. However, developing a learning tool for students with diverse backgrounds is challenging. This study proposes a robot-assisted Chinese learning system (RACLS) for those with diverse backgrounds and investigates its impact on learning motivation by a comparison study.
Exploring the Impact of Robot Interaction on Learning Engagement: A Comparative Study of Two Multi-Modal Robots
Fung K. Y., Fung K. C., Lui, TL., Sin, K. F., Lee, L. H., Qu, H., & Song, S.
Smart Learning Environments, 12(1), 12.
To propose an inclusive learning system developed on two multi-modal robots, Kebbi and Minibo, with interactive (i.e., movable hands) and straightforward features. The system integrates various interactive elements, such as animations, songs, dance, gestures, and touch, to enhance students’ learning engagement, interaction, and motivation and cater to their diverse needs.
A Study on Using ChatGPT to Help Students With Dyslexia Learn Chinese and English Writing
Fung K. Y., Fung K. C, Lee, L. H., Lui, TL., Qu, H., Song, S. & Sin, K. F.
ECNU Review of Education
This study examines the role of large language models—specifically ChatGPT—in supporting inclusive education. It investigates how ChatGPT can assist students with dyslexia, focusing on its potential to improve learning outcomes and inclusion in the classroom.
Research News / Press Release
Effects of Different Demographics When Evaluating Interactive Robot-Empowered Language Learning in Human-Robot Interaction
Fung K. Y., Fung K. C., Lui, TL., Mow W. H., Sin, K. F.
2025 20th ACM/IEEE International Conference on Human-Robot Interaction (HRI) (pp. 1010-1014). IEEE.
The previous work developed a language learning system with human-like robots utilizing self-determination theory (SDT) to enhance student’ learning engagement and proficiency. Initial evaluations indicated that typical students encountered stimulated interest and efficacy with the robot. To test the consistency of these findings, we conducted a replication study using the same system, robot, and tasks but with students with dyslexia. Through questionnaires, observations, and proficiency tests, the results confirmed that the interactive robot also improved learning motivation and performance in those with dyslexia, with the same system, robot, and tasks.
2024
ADPS—A Prescreening Tool for Students With Dyslexia in Learning Traditional Chinese
Fung K. Y., Tang, K. Y., Lui, T. L. R., Sin, K. F., Lee, L. H., Qu, H., & Song, S.
IEEE Transactions on Learning Technologies, 17, 1428-1444.
Early identification of specific learning disabilities, such as dyslexia, is crucial for timely intervention, yet many children in Hong Kong miss the optimal screening age due to resource limitations. This study presents an Automatic Dyslexia Pre-screening System (ADPS) for Chinese learners, addressing the unique challenge of the language’s lack of sound–script correspondence. Using a two-stage, gamified approach grounded in linguistic features, ADPS was evaluated with 62 students and successfully differentiated those with and without dyslexia, while its interactive design encouraged independent participation.
Humanoid Robot-Empowered Language Learning Based on Self-Determination Theory
Fung K. Y., Lee, L. H., Sin, K. F., Song, S., & Qu, H.
Education and Information Technologies, 1-30.
With the ability to provide feedback and assistance, humanoid educational robots have been proven effective in assisting students to overcome learning challenges and enhancing individual learning outcomes. However, the strength of humanoid robots in promoting social and emotional skills has not been well investigated. Socially supportive behaviour can contribute more to students’ learning engagement than knowledge transfer. This study focuses on the design of humanoid robots to engage students from functional and affective perspectives.
A Robot-Assisted Scenario Training for Students With ASD
Fung K. Y., Fung K. C., Lui, TL, Pang F., Sin, K. F., Qu, H., & Song, S.
International Conference on Computers in Education 2024.
Best Technical Design Paper Nominee
We developed a robot-assisted scenario training (RAST) system to facilitate inclusive learning and arouse students’ learning interests. With the RAST system, we seek to identify effective interactions that can improve students’ engagement.
Create-to-Learn Paradigm: A Proxy Visual Storytelling Tool (PVST) for Stimulating Children's Story Sense and Structure
Fung K. Y., Lee, L. H., Qu, H., Li, Y., Song, S., & Yip, D.
In Proceedings of the 17th International Symposium on Visual Information Communication and Interaction (pp. 1-8).
To introduce a proxy visual storytelling tool (PVST) that employs a character-based approach (i.e., proxy character assembling) to enhance children’s creativity and storytelling skills. Through a comparative study using Kurt Vonnegut’s “The Shape of Stories” theory, the PVST was evaluated.
Utilizing Humanoid Robots to Improve Learning Proficiency and Support for Students With Dyslexia: An Empirical Investigation
Fung K. Y., FUNG, K. C., SIN, K. F. K., LUI, T. L. R., QU, H., & SONG, S.
This study investigates the use of humanoid robots to enhance learning outcomes and provide targeted support for students with dyslexia. Through interactive, robot-assisted activities, the research explores how such technologies can improve engagement, learning proficiency, and accessibility in Chinese language education, offering insights into the potential of robotics for inclusive education.
Robot-Assisted vs. Conventional Learning of Kindergarten Students Using Scenario-Based Learning
Fung K. Y., Fung, K. C., Sin, K. F., Leung, R. L. T., Yow, W. Q., Perrault, S., ... & Song, S.
In INTERNATIONAL JOURNAL OF PSYCHOLOGY (Vol. 59, pp. 232-232). THE ATRIUM, SOUTHERN GATE, CHICHESTER PO19 8SQ, W SUSSEX, ENGLAND: JOHN WILEY & SONS LTD.
This study, published in the International Journal of Psychology, compares the effectiveness of robot-assisted and conventional teaching in scenario-based learning for kindergarten students. It examines how robot interaction can enhance engagement, motivation, and learning performance, providing empirical evidence for integrating robotics into early childhood education.
2023
Can Underprivileged Children Learn Effectively at Home? A Six-Month Study of Game-Based Traditional Chinese Learning During the Pandemic Lockdown
Fung K. Y., Lee, L. H., Hui, P., & Song, S.
IEEE Transactions on Learning Technologies
The COVID-19 pandemic disproportionately impacted underprivileged students, limiting their access to quality learning. This study investigates the effectiveness of unsupervised, game-based learning for traditional Chinese, using data from 3,245 quizzes completed by 26 students over six months. Results show significant gains in literacy and phonology, along with notable reductions in several common writing errors, highlighting the potential of game-based approaches to support autonomous learning in disadvantaged contexts.
Can Students With Dyslexia Learn Independently? A Seven-Week Study of Chinese Character Learning in an Informal Learning Environment
Fung K. Y., Perrault, S. T., Lee, L. H., Fung, K. C., & Song, S.
IEEE Transactions on Learning Technologies, 16(5), 619-632.
This study explores how large-scale digital learning can support primary school students with dyslexia in mastering Chinese character writing—an area of particular difficulty in Hong Kong. A mobile application with three progressive writing conditions was developed to guide learning, memory reinforcement, and recall. Testing showed that students with dyslexia significantly improved their writing skills and speed, narrowing the gap with peers, demonstrating the potential of structured, self-paced digital tools for inclusive education.
2022
Designing a Game for Pre-Screening Students With Specific Learning Disabilities in Chinese
Fung K. Y., Sin, K. F., Wen, Z. A., Lee, L. H., Song, S., & Qu, H.
In Proceedings of the 24th International ACM SIGACCESS Conference on Computers and Accessibility (pp. 1-5).
To design a Chinese morphological awareness puzzle that challenges students to recognize different words made up with the first character that is identical and the second character that is different, such as 樹枝 (literally means tree branch), 樹幹 (literally means tree truck), 樹葉 (literally means tree leaves), and 樹根 (literally means tree root).
Designing a Data Visualization Dashboard for Pre-Screening Hong Kong Students With Specific Learning Disabilities
Fung K. Y., Wen, Z. K., Li, H.T., Wang, X. B., Song, S. H. & Qu, H. M.
ACM SIGACCESS Conference on Computers and Accessibility (ASSETS ’22)
To design a new visualization dashboard for Hong Kong special education coordinators (SENCOs) and special education (special-ed) teachers to assist them in pre-screening students with special learning disabilities (SpLD).
The Impact of COVID-19 Pandemic on Learning Chinese at Home of Dyslexic Students
Fung K. Y., Fung, K. C.
Transforming education & strengthening society: Redesigning Pedagogy International Conference 2022: Conference proceedings (pp.209-227). Book.
The COVID-19 pandemic disrupted in-person classes in Hong Kong, limiting language learning opportunities for students, particularly those with dyslexia. This study developed a digital game-based learning tool for primary students to practice Chinese reading, writing, and pronunciation at home. Over four months, results showed that dyslexic students preferred radical–character matching games and improved in handwriting speed and accuracy, narrowing performance gaps with peers in some tasks. However, challenges such as reduced engagement, unsuitable home learning environments, and technical issues in audio games were observed, offering insights for future inclusive digital learning design.
2021
A Digital Tool to Provide Pre-Screening to Dyslexia in Hong Kong
Fung K. Y., Fung, K. C., Aidan, C., & Ching, Y. Y.
In 2021 IEEE International Conference on Engineering, Technology & Education (TALE) (pp. 755-761). IEEE.
To develop a digital tool to pre-screen dyslexia via handwriting in Hong Kong.
Research and Development of a Novel Mobile Chinese Application for Dyslexia Pre-Screening
Fung K. Y.
MPhil dissertation, Hong Kong University of Science and Technology
This study presents a novel mobile application for early pre-screening of dyslexia in Hong Kong, tailored to the Chinese language context. The app integrates six gamified tests covering handwriting, sentence reading, and language skills, with AI-based scoring criteria to identify key dyslexia indicators. In trials with 20 primary students, including eight with dyslexia, results showed slower writing speeds and distinctive handwriting patterns—such as disproportionate character components—among dyslexic students. The system offers an efficient tool for teachers to detect at-risk learners and enable timely intervention.
2019
Augmented Reality and 3D Model for Children Chinese Character Recognition—Hong Kong Primary School Education
Fung K. Y., Fung, K. C., & Wan, S. Y.
In Proceedings of 27th International Conference on Computers in Education, ICCE 2019 (Vol. 1, p. 673).
To shows an interactive and interesting way to promote Chinese character learning via the AR environment and 3D models.
Inquiry-Based Learning With Kanban Game
R. W. C. Lui, P. T. Y. Lee and K. Y. Fung
2019 IEEE 11th International Conference on Engineering Education (ICEED), Kanazawa, Japan, 2019, pp. 108-112
To present the design and the effectiveness of a computer game that supports inquiry-based learning of Agile development with Kanban.
2020
HCI Technology With Mastery Learning Approach for Children Learning Chinese Characters Writing in Hong Kong
Fung K. Y., Fung, K. C.
In 2020 IEEE 20th International Conference on Advanced Learning Technologies (ICALT) (pp. 226-227). IEEE.
To demonstrate an interesting way to enhance self-learning Chinese character writing via human-computer interaction (HCI) technology.