A Mechanism for Aligning Epistemic Curiosity with Systematic Learning: From Personal Upstream Understanding to Formal Knowledge and Problem-Solving Interfaces

This bilingual preprint proposes a mechanism for aligning epistemic curiosity with systematic learning. It addresses how curiosity-driven exploration may produce personal upstream models with broad explanatory scope but insufficient boundary resolution, and why such models are often functionally reset when learners enter formal curricula. The article argues that systematic learning need not begin by re-entering all foundational content. Instead, it can first identify the learner's existing understanding and then transform it through model externalization, absolutized boundary testing, adversarial challenge, localization of specific knowledge gaps, targeted supplementation, analysis of problem-construction mechanisms, and independent reconstruction. An autobiographical physics-learning case illustrates that deep understanding does not automatically become reliable problem-solving performance, while an informal counterexample involving another learner is used to specify the model's conditional applicability. The proposed mechanism is compared with research on epistemic curiosity, interest-driven learning, prior knowledge, knowledge integration, self-explanation, productive failure, problem posing, and AI-supported scaffolding. Its claimed contribution is not the invention of each component, but their organization into a continuous pathway from personal upstream models to formal knowledge and real-world task interfaces. This is a theoretical preprint and reports no original empirical data.