Designing for cognitive resilience: a distributed cognition approach to education in the age of generative AI

Generative AI is changing the cost structure of cognitive effort in education, making it cheaper and more socially acceptable to bypass the attention, retrieval, reasoning, and judgement that learning is meant to cultivate. This paper develops cognitive resilience, the trained capacity to sustain these functions under conditions of pervasive AI availability, as a design requirement for education. Building on a distributed cognition view, in which tools and mediation have always been constitutive of human thought, the paper argues that the relevant question is not whether AI mediates learning, but what kind of mediation is used, at what developmental moment, and whether learners remain active, accountable participants in it rather than passive recipients of its outputs. Drawing on cognitive science, learning science, and emerging evidence on AI dependence, the paper proposes a nine-principle design framework and argues that pedagogical design alone cannot secure cognitive resilience without governance conditions, since commercial EdTech interests can be structurally misaligned with education's developmental purposes. The argument is developed internationally, using the Australian policy landscape as an illustrative rather than comparative case. A developmental pipeline logic, equity analysis, and research agenda close the paper.