MarcoFLY Framework — Cognitive Forcing: Preregistration of Experimental Protocol for Testing the MarcoFLY-1 Postulate on Human Reasoning Quality in Structured Human-AI Dialogue

Primary Hypothesis (MarcoFLY-1 Postulate): In human-AI dialogues structured with systematic epistemic labeling (CERTAINTY/INFERENCE/HYPOTHESIS) and cognitive forcing protocols (grounding, contradiction generation, external verification ≥50%), human reasoning quality — measured by task accuracy and bias resilience — increases significantly compared to unstructured AI interaction baseline. Statistical criteria: p