Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Draw2Learn: A Human-AI Collaborative Tool for Drawing-Based Science Learning

Yuqi Hang

TL;DR

Draw2Learn investigates how a teammate-oriented AI can support drawing-based science learning by converting learning goals into structured drawing quests, offering optional scaffolds, monitoring progress, and delivering multidimensional feedback. Grounded in learning theories such as Bloom's Taxonomy, Zone of Proximal Development, and formative feedback principles, the system aims to preserve learner autonomy while providing scalable guidance. Formative evaluations with six participants suggest high usability and perceived usefulness, with themes highlighting the value of AI scaffolding and learner autonomy, though adaptive scaffolding and stronger teammate framing require refinement. The work contributes a design framework and interaction patterns for AI-assisted generative learning, and points to future research on learning outcomes, broader applicability, and classroom deployment.

Abstract

Drawing supports learning by externalizing mental models, but providing timely feedback at scale remains challenging. We present Draw2Learn, a system that explores how AI can act as a supportive teammate during drawing-based learning. The design translates learning principles into concrete interaction patterns: AI generates structured drawing quests, provides optional visual scaffolds, monitors progress, and delivers multidimensional feedback. We collected formative user feedback during system development and open-ended comments. Feedback showed positive ratings for usability, usefulness, and user experience, with themes highlighting AI scaffolding value and learner autonomy. This work contributes a design framework for teammate-oriented AI in generative learning and identifies key considerations for future research.

Draw2Learn: A Human-AI Collaborative Tool for Drawing-Based Science Learning

TL;DR

Draw2Learn investigates how a teammate-oriented AI can support drawing-based science learning by converting learning goals into structured drawing quests, offering optional scaffolds, monitoring progress, and delivering multidimensional feedback. Grounded in learning theories such as Bloom's Taxonomy, Zone of Proximal Development, and formative feedback principles, the system aims to preserve learner autonomy while providing scalable guidance. Formative evaluations with six participants suggest high usability and perceived usefulness, with themes highlighting the value of AI scaffolding and learner autonomy, though adaptive scaffolding and stronger teammate framing require refinement. The work contributes a design framework and interaction patterns for AI-assisted generative learning, and points to future research on learning outcomes, broader applicability, and classroom deployment.

Abstract

Drawing supports learning by externalizing mental models, but providing timely feedback at scale remains challenging. We present Draw2Learn, a system that explores how AI can act as a supportive teammate during drawing-based learning. The design translates learning principles into concrete interaction patterns: AI generates structured drawing quests, provides optional visual scaffolds, monitors progress, and delivers multidimensional feedback. We collected formative user feedback during system development and open-ended comments. Feedback showed positive ratings for usability, usefulness, and user experience, with themes highlighting AI scaffolding value and learner autonomy. This work contributes a design framework for teammate-oriented AI in generative learning and identifies key considerations for future research.
Paper Structure (12 sections, 4 figures)

This paper contains 12 sections, 4 figures.

Table of Contents

  1. Introduction
  2. Theoretical Grounding
  3. Research Questions
  4. System Design
  5. User Workflow
  6. Interface Architecture
  7. Formative Feedback from Users
  8. Setup
  9. Feedback Summary
  10. Limitations and Future Research
  11. Discussion
  12. Conclusion

Figures (4)

  • Figure 1: User workflow: (1) Learner inputs learning goal; (2) AI generates quest sequence; (3) Learner draws on canvas; (4) AI monitors progress and provides 4-dimensional feedback; (5) Optional helpers available on request; (6) Gem reward upon each task completion; (7) Style transformation after completing all tasks.
  • Figure 2: Three-panel interface: (Left) Quest panel with progress tracking and helper buttons; (Center) Drawing canvas with style selectors; (Right) AI teammate delivering multidimensional feedback.
  • Figure 3: User feedback ratings with individual data points: Positive ratings across dimensions including attractiveness (M=6.17), overall experience (M=6.15), positive affect (M=5.83), ease of use (M=5.67), and usefulness (M=5.67), with overall mean of 5.47/7.0.
  • Figure 4: Qualitative themes from thematic analysis: Five strengths emerged including AI scaffolding, deep learning, engagement, clear guidance, and learner autonomy. Five improvement areas include performance, input modalities, AI transparency, adaptive scaffolding, and interface refinement.