Invisible Pedagogy: An Algorithmic and Neuroplasticity Model for Ambient Scaffolding and Implicit Skill Acquisition

This working paper introduces “Invisible Pedagogy,” a biophysically inspired computational framework for AI tutors that mitigate cognitive offloading-induced skill decay. It formalizes human–AI interaction as a POMDP with EKF personalization and a Synaptic Tagging & Capture-inspired consolidation layer, evaluates adaptive spacing and scaffolding policies against standard baselines in simulation within the model, and outlines ethical design principles and a 2×2×2 empirical study to validate the framework. It is a theory/model foundation and does not yet report large-scale human trials.