TRAIL is a web platform for reproducible human–AI teaming experiments. It makes an AI teammate configurable across Big Five personality traits, communication style, participation timing, and memory, while supporting instrumented real-time collaboration, chained longitudinal studies, and exportable analytics. In a six-session classroom deployment involving about 51 students, the platform maintained the AI as a stable minority participant and enabled text-similarity analysis. A single blind persona change produced a design-consistent double dissociation: a cognitive-scaffolding agent received stronger contribution ratings and closer linguistic alignment, while a socially supportive agent produced warmer team climate and lower over-reliance.
No heat snapshots are available in the last 24 hours.