1. Definition
The Learning Cycle is a guided, iterative framework that structures inquiry into 6 specific phases (Notice, Ask, Generate, Test, Reflect, Share) to transform AI interaction from passive absorption into active, measurable discovery.
2. Use Case
Activated when the learner encounters a new domain or a conceptual anomaly, requiring a rigorous pathway to convert initial curiosity into testable hypotheses.
3. Human Role
Formulates initial questions, develops original hypotheses by cross-referencing intuition with emerging data, critically evaluates failures during the test phase, and synthesizes the discovery to argue it publicly.
4. AI Role
Responds to explorations by introducing methodological constraints and providing simulation scenarios. It returns limited or partial feedback to force the student’s autonomous exploration, refraining from delivering the final solution.
5. Friction
Interrupts the reflex to externalize heuristic thinking, forcing human ideas to collide against the structural limits of the system during the verification phase (“Test”) and blocking access to the easy answer.
6. Risk
Without this closed loop, the student consolidates a silent epistemic dependency: they consume the output as pre-digested truth without building and consolidating their own logical neural networks.
7. Observable Markers
During the public sharing phase, the student can publicly reconstruct their methodological errors, defending the exact reasons that forced them to adjust their starting hypotheses.