Operational Transition Test (OTT): Experimental Crucible v0.3 — Prospective Reduced-VCX Results (H1–H4 PASS)

Abstract

This record contains the prospective results of the Operational Transition Test (OTT) Experimental Crucible v0.3.0 — the reduced V–C–X criterion (scope-wide interventional validity ∧ robust control-cost compression ∧ attributable constrained operational expansion). The protocol, hypotheses H1–H4, thresholds, benchmark families, seed derivation, development/evaluation allocation, and scoring rules were publicly frozen before execution under DOI 10.5281/zenodo.22117014 (freeze timestamp 2026-08-26T19:09:17Z; run authorization 2026-08-26T19:13:03Z). Protocol root SHA-256 verified identical in both records: 44cc90e34397b83a4bce29ab9307511dcdf78e5a22b97bc39f03ecd4bfa29b06. Frozen verdict: H1 PASS · H2 PASS · H3 PASS · H4 PASS · overall_success = PASS. Full VCX on the frozen evaluation set (1520 cases): TP 260, TN 1259, FP 1, FN 0 — TPR 1.000000, FPR 0.00079365, balanced accuracy 0.99960317. H2 minimality: each of V, C, X is independently necessary on its dedicated isolation family (two-factor ablations reach FPR 1.0 where full VCX has FPR 0). H3 sealed challenge (400 zero-development cases): balanced accuracy 1.0. H4 topology stress: balanced accuracy ≥ 0.9984 on all four transition topologies. The single false positive occurred in context_fragile_validity (family FPR 0.0125, below the frozen 0.10 ceiling) — the family targeting the v0.2 failure mode (v0.2 H1 failed via context_fragile_macro, DOI 10.5281/zenodo.22116481). Interpretation boundary. This is a PASS of the frozen prospective internal synthetic benchmark. It is not independent external validation, not cross-domain validation, and not empirical confirmation of a universal law of emergence or of LOGOS-44. The narrower supported statement: a three-condition criterion distinguished the prospectively generated positive and negative transition mechanisms in the frozen v0.3 synthetic Crucible, including sealed zero-development families and multiple transition topologies. No thresholds were changed after observing results; no post-hoc retuning, reclassification, or rerun was performed. Contents. The ZIP is the canonical complete result artifact (run authorization, development freeze, evaluation lock, predictions, ground truth, scoring, summary/report, logs, integrity manifests). RUN_RESULT_REPORT_v0.3.md is included unchanged for human-readable preview. ZIP SHA-256: 7538425869499d1784d1b1c1fc3dc1f25f8bc6eac87fa71c678619d767c46c2c; uploaded files confirmed byte-identical via server checksums. Licensing. Record license CC BY 4.0 (text/data); code components remain under Apache License 2.0 per the research-line license map (LICENSES.md in the frozen protocol record). Copyright © 2026 Sławomir Grzegorz Gątkowski.