Situated Learning Accessibility: Difference, Indeterminacy, and the Logic of Dynamic States

Adaptive learning systems compress complex histories into tractable representations of a learner. Prior work already supplies several safeguards against naive compression: relational accounts of accessibility and affordance, state-space and partially observable models of student knowledge, non-monotonic and four-valued student models, Bayesian state-space models that preserve epistemic uncertainty, and dynamic epistemic logics of learning and conflicting evidence. This paper addresses a narrower problem that persists across those traditions: the silent change of status by which an estimate is read as the thing estimated, an observation becomes a trait, missing evidence becomes a negative conclusion, and a revision of the model is narrated as a change in the learner. We introduce the Dynamic Map of Learning Accessibility (MDAA), which conceptualizes learning accessibility not as a stable property of the learner but as a situated and time-dependent relation among subject, learning object, and conditions. Building on the established distinction between a latent process and estimates of it, we add a proposition-level epistemic layer in which the status λ(t; P) of a scoped claim is derived from provenance-preserving support and counter-support, using a provisional Belnap–Dunn scaffold that distinguishes support only, counter-support only, both, and neither. The candidate contribution is the joint architecture: configuration-scoped accessibility S×O×C×t, an independently specified process model x(t), a proposition-level evidence status with explicit provenance, and a falsifiable cross-layer hypothesis asking whether independently identified process transitions are associated with changes in the distribution of epistemic states. The proposal is developed in dialogue with Imre Hermann’s individual logic, Luiz Sérgio Coelho de Sampaio’s logic of difference, and contemporary learner-modeling and evidence-logical traditions. Preprint. Conceptual manuscript, not peer reviewed. No empirical results are claimed.