AI for Youth Learning

会議の名前
CHI 2026
"Let’s talk about data": Co-Designing Critical Data Literacy Tools for K-12 Education through Dialogic Learning
要旨

This paper investigates dialogic learning as a pedagogical lens for designing tools that support critical data literacy (CDL) in K-12 education. We present a research-through-design (RtD) project conducted with three teachers to operationalize dialogic learning in design. The resulting prototype, Datafy, enables students to produce and analyze personally meaningful music data and co-create playlists, fostering critical reflection and dialogue around data practices. Through a classroom observation with 40 sixth-grade students, we show how the tool shaped learners’ collaborative exploration and construction of knowledge about data. We contribute a theoretical framework and an empirical case study of embedding dialogic learning in the design of CDL technologies that promote dialogue and critical engagement with data. We recommend: (1) anchoring co-design with teachers in learning activities and learning goals, (2) designing diverse opportunities for dialogue about critical data literacy, and (3) treating dialogue linked to learning goals as design evidence for tool evaluation.

著者
Yu-Yu Liu
Aarhus University, Aarhus, Denmark
Maja Dybboe
Aarhus University, Aarhus N, Denmark
Johannes Ellemose
Aarhus University, Aarhus N, Denmark
Johanne Birkkjær. Bjerrum
Aarhus University, Aarhus, Denmark
Line Have. Musaeus
Aarhus University, Aarhus, Denmark
Christian Dindler
Aarhus University, Aarhus, Denmark
Ole Sejer. Iversen
Aarhus University, Aarhus, Denmark
Balancing Attention Support for Early-learners (BASE) in Digital Technology Interactions: A Framework for Design and Use
要旨

Digital technologies increasingly shape young children’s attention, yet current design paradigms often prioritise engagement over developmental appropriateness. This paper responds to growing concerns about the attention economy’s influence on children’s lives. While many theories address attention and digital engagement, there is no unified design framework to support balanced attentional development in early childhood (ages 3 to 5). We introduce the Balanced Attention Support for Early-learners (BASE) framework, which comprises three principles: (1) ally indoor and outdoor learning, (2) balance attentional processes in technology interactions, and (3) co-design with children’s attention in mind. Developed through a meta-narrative review of psychology, early childhood education, and human–computer interaction literature, the framework identifies five attentional modes: focused attention, restoration, switching, mindfulness, and relational attending. BASE reframes familiar design practices through the lens of attentional development, offering new directions for resisting attention-capture paradigms and supporting socio-cognitive growth, ethical engagement, and the whole child.

著者
Kellie Vella
Queensland University of Technology (QUT), Brisbane, Australia
Nelli Holopainen
Queensland University of Technology, Brisbane, Queensland, Australia
Marina Torjinksi
Deakin University, Melbourne, Australia
Madeleine Rose. Dobson
Curtin University, Bentley, Western Australia, Australia
Karen Murcia
Curtin University, Perth, Australia
Margot Brereton
QUT, Brisbane, Brisbane, Australia
Generative AI in Children's Creative Collaboration: Impact, Perception, and Design Guidelines
要旨

The advent of Generative AI (GenAI) has raised discussions about its effects on individuals. However, little is known about its impact on children’s creative collaboration, despite its importance for social and cognitive development. We examined GenAI’s role in children’s creative collaboration through five co-design sessions with 28 children (ages 5-11) using diverse GenAI tools (text, image, video, voice); 17 parents participated in focus group interviews. Our findings show that GenAI can foster positive social dynamics by enabling “Human vs. AI” teaming and children’s co-creation with shared ownership. However, GenAI disrupted collaborations when roles between children were unclear, AI ignored group dialogue, and AI dominated children’s agency. Children and parents envisioned socially attuned AI that could play an “older sibling” role--scaffolding while allowing playful disagreement--while raising concerns about children’s overreliance on GenAI. This work advances understanding of GenAI in collaboration and proposes design implications for designing AI systems that support child-centered collaboration.

著者
Daeun Yoo
University of Washington, Seattle, Washington, United States
Michele Newman
University of Washington, Seattle, Washington, United States
Caroline Pitt
University of Washington, Seattle, Washington, United States
Kevin Huu. Vo
University of Washington, Seattle, Washington, United States
Michelle Zhang
University of Washington , Seattle, Washington, United States
Michelle Kim
University of Washington Information School, Seattle, Washington, United States
Katie Davis
University of Washington, Seattle, Washington, United States
Jason Yip
University of Washington, Seattle, Washington, United States
Generative AI and Creative Mediums for Youth’s Emotion Regulation: An Interview Study with Clinicians
要旨

Emotion regulation (ER) is essential to youth well-being, and cognitive-behavioral therapy (CBT) is an established approach for building ER skills. Clinicians often use creative mediums such as visuals and narratives to support ER through CBT, yet access and personalization remain limited. Generative AI (GenAI) shows promise for addressing these limitations, but its benefits and risks in youth ER remain underexplored, underscoring the need for expert perspectives. We interviewed 20 ER specialists--psychotherapists, art therapists, and psychiatrists--using a GenAI technological probe that generated CBT-based visuals and narratives. Clinicians highlighted GenAI’s potential as a “bridge” to help youth concretely identify and express emotions, practice personalized coping skills, and mediate ER conversations between home and clinics. They also cautioned that the vividness and unpredictability of GenAI outputs may trigger trauma or reinforce maladaptive thinking. We propose psychologically grounded design implications for GenAI to foster safe, engaging youth ER as a foundation for lifelong well-being.

著者
Daeun Yoo
University of Washington, Seattle, Washington, United States
Daniela E. Munoz Lopez
University of Washington, Seattle, Washington, United States
Xiaotian Daisy Hu
University of Washington, Albany, Oregon, United States
Jason Yip
University of Washington, Seattle, Washington, United States
Katie Davis
University of Washington, Seattle, Washington, United States
From Squishing to Meaning: Exploring Data Physicalization Through Children’s Embodied Experiences
要旨

Data physicalization is a promising approach for empowering children to understand and enjoy their own data. However, it relies on embodied metaphors to convey information effectively. This paper explores how to elicit children's embodied experiences using a set of shape-changing objects that can inform the design of dynamic physicalizations. We propose a set of auxetic metamaterials, which can bend, twist, scale and shear. Following principles of tangible interaction, we conducted a study with 59 children who participated in four movement-based games before being introduced to the collection of shape-changing tangibles. Children expressed metaphors based on these activities related to concepts such as containers, rhythm, resistance, and semantic analogies, which we categorised into embodied schemas. Our findings reveal that characteristics of the shape-changing tangibles aid children in connecting their bodily experiences to dynamic transformations. Translating these insights into idea sketches, we outline how to tailor these affordable shape-changing mechanisms into usable prototypes.

著者
Andres Alberto. Ramirez Duque
University of Glasgow, Glasgow, Glasgow, United Kingdom
Dushani Perera
University of Edinburgh, Edinburgh, United Kingdom
Dorsey B.. Kaufmann
University of Edinburgh, Edinburgh, United Kingdom
Ayça Atabey
University of Edinburgh, Edinburgh, United Kingdom
Uta Hinrichs
University of Edinburgh, Edinburgh, United Kingdom
Andrew Manches
University of Edinburgh, Edinburgh, United Kingdom
Stephen Anthony. Brewster
University of Glasgow, Glasgow, United Kingdom
Design and Evaluation of a Photorealistic AI Virtual Peer in Elementary Collaborative Classroom
要旨

In elementary education, students struggle to articulate uncertainties, limiting diverse perspectives in classroom discussions, particularly in small schools where limited participants constrain collaborative learning. This study designed and evaluated ``Saya,'' a photorealistic AI virtual peer functioning as an additional student. We implemented five teacher-controlled speech acts (expand, probe, summarize, lighten, and incorrect answer) through dynamic classroom dialogue generation using GPT-4o-mini. Field studies in Japanese elementary schools (large class: 27 students, small class: 2 students) demonstrated that Saya integration increased the proportion of student speaking time by 1.28 times and 2.07 times respectively, with 95.6% and 100% of students expressing desire for future Saya-integrated lessons. Teachers reported enhanced student concentration and listening behaviors, noting that interactions with Saya prompted students to reconstruct their own understanding of the learning material. This research provides new insights into design principles for collaborative learning agents in elementary education settings, effective implementation scenarios based on class size, and the future potential of AI-enhanced collaborative learning.

著者
Satomi Tokida
The University of Tokyo, Tokyo, Japan
Koki Usui
AISIN CORPORATION, Aichi, Japan
Godai Tanaka
AISIN CORPORATION, Aichi, Japan
Shin Osuga
AISIN CORPORATION, Aichi, Japan
Masanori Kayano
University of Yamanashi, Yamanashi, Japan
Yuta Itoh
Institite of Science Tokyo, Yokohama, Japan
Yoshio Ishiguro
The University of Tokyo, Tokyo, Japan
DOLLama: Fostering Family Anti-Bullying Learning through AI-Augmented, Toy-Mediated Educational Drama
要旨

Educational drama is a proven method for anti-bullying education, but its traditional reliance on teachers and peers limits its accessibility to children and families outside of school. HCI has rarely explored how to augment this practice with AI-infused, interactive role-playing or how to involve parents in the process. We introduce DOLLama, an AI-powered projection-augmented interactive system that transforms children's toys and family-created stories into gamified anti-bullying vignettes. A study with 20 families demonstrated how DOLLama facilitated children’s and parents’ learning. Children used their toys to enact the roles of the one being bullied and bystanders, developing empathy and practicing coping strategies in co-performance with AI-controlled toy characters. By observing this play, parents gained new insights into their child’s strengths and challenges and identified their own knowledge gaps. Based on these findings, we derive HCI design implications for AI-enhanced, toy-mediated educational drama that supports anti-bullying education for children and their families.

著者
Di Liu
Southern University of Science and Technology, Shenzhen, China
Zhenhao Zhang
Southern University of Science and Technology, Shenzhen, Guandong, China
Zhuoyi Zhang
University of Washington, Seattle, Washington, United States
Yufei Hu
Southern University of Science and Technology, Shenzhen, China
Keming Jiao
Lund University , Lund , Sweden
Xueliang Li
Southern University of Science and Technology, Shenzhen, Guangdong, China
Pengcheng An
Southern University of Science and Technology, Shenzhen, China