Bridging Pedagogy and Play: Introducing a Language Mapping Interface for Human-AI Co-Creation in Educational Game Design

A team of researchers from Northeastern University has introduced a novel web tool designed to revolutionize educational game design through human-AI collaboration. The tool, presented in a paper accepted for CHI EA '26, addresses a critical pain point: while educational games can boost motivation and critical thinking, instructors often struggle to design games that reliably hit specific learning targets. Existing authoring tools reduce programming needs but fail to solve the core design challenge, often leaving non-experts reliant on the 'black box' suggestions of AI assistants. This new interface positions structured language as the primary medium for collaboration with a Large Language Model (LLM).

The core innovation is a controlled natural language framework where users and the AI assistant collaboratively develop a shared language that explicitly links pedagogy to gameplay. This mapping occurs through four interconnected components within the interface, making the pedagogical intent behind every game mechanic transparent and editable. By externalizing this design logic, the tool aims to lower barriers for educators without game design expertise, preserve human agency in critical pedagogical decisions, and enable continuous alignment and reflection between learning goals and play. The approach represents a significant shift from AI as a mere suggestion engine to a true co-creator within a structured, human-guided process.