CALM v1.1: EEG Zoom Law Dataset and Minimal Evaluator

CALM (Cognitive Alphabet for Latent Microstates) v1.1 proposes that resting-state human EEG contains a robust “zoom law”: refining a specific latent state (S1) into finer substates yields a consistent gain in predictive / compression metrics across subjects, relative to comparable non-hierarchical refinements. This dataset provides the numerical backbone and minimal evaluator for that claim. It includes: zoom_S1_summary.json — summary statistics for the S1 zoom effect (median Δxent, confidence intervals, p-value). zoom_S1_subjects.csv — per-subject zoom metrics used to compute the summary. calm_eval_min.py — a compact evaluator script to apply the CALM v1.1 zoom test to new EEG datasets. Together, these files let researchers recompute the zoom law on the original cohort or test it on their own data. The broader CALM framework (three-state alphabet, full figures, and narrative report) is described in preparation; this v1.1 Zenodo record focuses specifically on the zoom law statistics and evaluator, making the core test easily falsifiable and reusable. Authored by Caleb “Telos” Holmes using an AI research assistant (“Aetheron”) as a tool for code and documentation. Responsibility for the scientific claims rests with the human author.