Thinking in Graphs with CoMAP: A Shared Visual Workspace for Designing Project-Based Learning

Researchers Ruijia Li and Bo Jiang have introduced CoMAP, a groundbreaking AI system that reimagines how educators design project-based learning (PBL) curricula. Unlike traditional linear tools or conversational AI assistants, CoMAP provides a shared visual workspace based on graph structures, allowing teachers to map out highly interdependent learning components in a non-linear fashion. The system embodies theories of distributed cognition, transforming the human-AI relationship from a simple prompt-response loop into what the researchers call a "transparent and equitable partnership." This graph-based approach specifically addresses the limitations of current tools that struggle to capture the creative, iterative nature of educational design.

In their study presented at CHI 2026 involving 30 educators, the researchers found CoMAP delivered significant improvements across multiple metrics. Teachers using the system showed enhanced design expression, stronger divergent thinking capabilities, and more effective iterative practice compared to those using dialogue-only AI interfaces. The visual, artifact-centric nature of CoMAP reduced cognitive load while fostering greater trust in the AI collaboration process. By providing persistent, shared context through graph visualizations, the system enables educators to maintain control over their creative process while leveraging AI support for complex curriculum planning tasks that involve numerous interconnected elements.