What question did this study set out to answer?

The aim is to confirm dose-response monotonicity of AC1 under new seeds and assess substrate family moderation in the analysis.

June 11, 2026Open Access

PHILIA-EcoSensory Swarm Phase SC-8 Pre-registration: Confirmatory Validation Protocol for AC1 Dose-Response Monotonicity and Substrate Family Moderation — 182-Substrate Integrated Registry × v32 Apparatus, Independent Seed Replication 67–71

Key Points

The aim is to confirm dose-response monotonicity of AC1 under new seeds and assess substrate family moderation in the analysis.
Pre-registration of confirmatory hypotheses under Bonferroni correction with independent replication tests.
Use of locked normalization apparatus and a comprehensive dataset from a 182-substrate integrated registry.
Power analysis conducted with Monte Carlo simulations to validate effect sizes.
Primary hypothesis shows 7/7 binary monotonicity with a replication probability of 0.70.
Secondary hypothesis reveals significant difference in substrate family moderation with |Z|≥2.241, achieving an effect size threshold of 41%.
Power at observed SC-7 effect is calculated at 0.9638, indicating strong support for findings.

Abstract

This document constitutes the pre-registration protocol for PHILIA Phase SC-8 confirmatory analysis. SC-8 Confirm will apply the locked v32 self-raw × 3-channel normalization apparatus to the 182-substrate integrated registry (SHA256: 3948b55114a6bad44296fa78cb6d53f8fc89d553137d3e6c0d683023344f7c9a) using five independent seeds 67–71, totaling 910 runs (182 substrates × 5 seeds × T=80, 000 steps). Two confirmatory hypotheses are pre-registered under Bonferroni correction (αfamily=0. 05, k=2, αₚerₜest=0. 025). H-SC8-Primary: Independent replication test of SC-7 Criterion B (Part 2 n=90, 7/7 perfect AC1 θ-sweep monotonicity) under new seeds 67–71. Primary verdict: 7/7 binary monotonicity (no escape clause). Sensitivity: Jonckheere-Terpstra (SA-1) + Spearman θ-rank (SA-2). Power: P (7/7|π=0. 5) =0. 0078 < α ✅; replication probability at π=0. 95: 0. 70. H-SC8-Secondary: Substrate family as moderator of AC1–MSE↔z direction. Fisher-z comparison of ρᵦ (RealWorld, AC1<0. 95, n=52) vs ρᵦ (Synthetic, AC1<0. 95, n=24). Verdict: |Z|≥2. 241 AND |Δρ|≥0. 40. Effect size threshold: 41% of SC-7 observed effect (Δρ=0. 968). Power at SC-7 observed effect: 0. 9638 (Monte Carlo, N=100, 000). Four pre-registered exploratory analyses (verdict-neutral, §4): Traffic+Environment sub-analysis, four-family Cochran's Q (z-level), outlier/boundary case sensitivity, AR (1) φ-sweep z-verification. Power analysis script SHA256: 40f031e97f12c19d42118f5d58ac8ec59924bcf6b086fc64566a50298fae944e Gate-0 v10 Cycle 3: UNCONDITIONAL ACCEPT (크리스 / Claude Opus / Anthropic). 5-AI adversarial review: Pandora / Lucas / Seogsaeng / Seonbi / Chris — all PASS. Part of the PHILIA-EcoSensory Swarm SC series. Measurement is life. Description, not proof. Pre-registration protects us. — PHILIA OS | 0∞1∞0. 5∞ SC-8 Explore Report DOI: 10. 5281/zenodo. 20603071SC-8 Annex v1. 3 DOI: 10. 5281/zenodo. 20576962SC-7 Results Report DOI: 10. 5281/zenodo. 20565918

Read Full Paperexternally

Mark Helpful

Bookmark

Relay

View Full Paper