Using Co-Design to Examine How Children Conceptualize Intelligent Interfaces
summary
Co-design with seven children ages 7–12 and 11 adults runs across four scaffolded co-design sessions (279 minutes of video, 308 child utterances) using bags-of-stuff, big paper, stickies and role-play to probe error detection and correction. Children build IUIs as whole systems: multimodal input (speech, handwriting, drawing, wearables), context- and emotion-aware output, and meta‑intelligence that admits confusion and asks clarification. Concrete recommendations: keep conversation, surface uncertainty instead of silent autocorrect, and let children teach the system (e.g., images to correct errors).
AI-written summary · about
chapters (11)
- 0:00Introduction to research
- 2:15Defining intelligent interfaces
- 5:51Participatory design approach
- 9:21Key themes in children's models
- 13:18Connections between themes
- 15:08Design recommendations
- 15:33Q&A session begins
- 16:16Value of co-design with children
- 17:41Session scaffolding process
- 19:28Children's prior tech experience
- 21:38Closing remarks
related talks
06mDesign and Evaluation of DIO Construction Toolkit for Co-making Shared Constructions
18mExamining Adult-Child Interactions in Intergenerational Participatory Design
45mModeling as Mind Redesign: Why participatory AI is better
17mThe Evolution of Engagements and Social Bonds During Child-Parent Co-design
46mDesign Patterns For AI Interfaces
34mDesign in the times of AI
25mArtificial intelligence & design: imagining the future of UX
09mHow to Support Users in Understanding Intelligent Systems? Structuring the Discussion