Human Capability Test (HCT): Evidence, Reasoning & Judgment — A Proposed Methodological Protocol for Longitudinal Assessment

This methodological companion to Human Development through AI-Mediated Environments (HD-AIME) presents a proposed research protocol for investigating changes in independently demonstrable human capability during sustained AI-mediated activity. The initial Human Capability Test (HCT) pilot focuses on Evidence, Reasoning, and Judgment (ERJ). It outlines three experimental conditions: independent work without generative AI, performance-oriented AI assistance, and developmentally structured AI assistance. The proposed design includes independent baseline and longitudinal reassessments, a provisional behavioral scoring rubric, the Project Atlas assessment scenario, and supplementary Cognitive Provenance documentation. The protocol identifies methodological requirements concerning task equivalence, scoring reliability, experimental fidelity, participant protection, statistical analysis, and future preregistration. Research status: This is a proposed, unvalidated methodological protocol. It does not report participant data, completed experiments, or validated assessment instruments. The procedures and measures require independent methodological review and feasibility testing before confirmatory research. Joseph Korson, Independent Researcher. Version 0.1 — November 2026.