Metaheutagogy: A Framework For AI-Mediated Metacognitive Learning

Artificial Intelligence is rapidly reshaping how we learn by mediating inquiry, feedback, collaboration, and artefact creation. As these tools grow more capable, the key educational challenge is judgement, metacognition, and human agency. The recent preprint authored by Alexander Mikroyannidis and John Domingue (The Open University, UK) introduces metaheutagogy: a framework that goes beyond pedagogy, andragogy, and heutagogy by positioning AI as a reflective and dialogic partner in learning. Metaheutagogy uses the analogy of learners as F1 drivers: success comes from the ability to steer, govern, and critically oversee powerful AI systems.

The paper offers:

  • A comparative model from pedagogy → andragogy → heutagogy → metaheutagogy
  • A description of an AI‑mediated learning ecosystem
  • Practical pedagogical scenarios from Socratic AI tutors to collaborative inquiry and constructionist studios
  • A discussion of risks and how metaheutagogy mitigates them

Metaheutagogy is intended to provide a useful conceptual and practical foundation for educators, institutions, and policymakers designing learning for an AI‑shaped world.

The preprint is available here: https://doi.org/10.35542/osf.io/xt6c7_v1.