Decisive Tests: Status

KILL IS SHARED WITH ΛCDM (2026-09-27); WIN SIDE SHARED WITH CONSTANT-w CARDASSIAN (2026-10-08: wCDM fits any DR2-allowed point on the framework's curve to Δχ² ≤ 0.42 on DR3-like data; resolving the shape needs γ ≈ 0.30, excluded by DR2 at Δχ² = 137): the family contains ΛCDM at γ = ½, so its kill fires only when DR3 excludes Λ in the thawing or crossing direction, and that refutes ΛCDM equally. It cannot single out the framework. Its only non-Λ outcome (freezing away from Λ) is a win branch for Cardassian-class prior art. | The framework's cosmic sector (Session 100: G_eff = G/C in the Friedmann equation, C₀ = Ω_m forced, γ the only parameter) cannot produce the w = −1 crossing DESI DR2 prefers (w₀ > −1 with wₐ < 0, all four data combinations) — and as of 2026-08-11 that statement holds for the model CLASS, not just the model as specified. The originally drafted criterion targeted the substituted model's literal sign lock, sign(w₀ + 1) = sign(wₐ); the covariant completions were then derived and the literal lock DIED (the Brans-Dicke completion populates mixed-sign pairs, and the Appendix-D equation as written has no dark-energy sector at all) while the no-go HARDENED: every consistent completion misses the DESI quadrant — 0/192 γ values at four Brans-Dicke ω; forced wₐ comes out +0.23…+0.60, wrong sign every time (this is the Brans-Dicke covariant completion with w₀ forced to DESI's value; the non-covariant substituted family's best fit is a different object, (w₀, wₐ) = (−0.993, +0.023), i.e. ΛCDM; see /dark-energy). The class statement to register: dark energy slaved to matter density reaches the DESI quadrant iff ρ_DE(x) has an interior maximum, and no completion of C = tanh(γ·ln(1+x)) produces one. Kill fires if DR3 (full published covariance, Ω_m and r_d·H₀ marginalised, one SNe compilation fixed in advance) excludes Λ in the thawing or crossing direction (w₀ > −1 with wₐ < 0, whether or not w = −1 is crossed; worded "requires the crossing" until 2026-10-26, which a thawing-only result would not have fired) — with a pre-committed projection-robustness check, since the CPL (w₀, wₐ) crossing preference is itself debated as a parameterization artifact (Shlivko & Steinhardt 2024; Cortês & Liddle 2024 — a debate that cuts in the framework's favor and is cited both ways). NOT discriminating in the confirming direction — though the tie is now conditional: the γ = 1/2 = ΛCDM identity is a property of the substitution (completion B has no ΛCDM member), so "kill-or-tie" is substitution-conditional. Timeline: DESI DR3 ~2027–2028. Full derivation and honest bounds: /dark-energy and /honest-assessment#dark-energy.

This page advertised TEST-02 as the #1 active discriminating test — "requires zero-cost Gaia DR3 analysis" — for weeks after /tier-1-existing had closed it. It has no outcome that favours the framework. If Gaia confirms a MOND-scale ~18% anomaly (Chae 2023–2025), C(ρ) is refuted, because C(ρ) predicts a Newtonian null (γ_g ≡ 1 exactly in the wide-binary ratio, not a small signal below reach; current status on Tier 1, 2026-10-18). If Gaia confirms the null (Banik et al. 2024; Pittordis & Sutherland 2023), the framework merely reproduces Newton and is not selected over it. Both branches lose or tie. And at the knees where the ratio is null, the density law already fails lunar laser ranging (68/74 laws) and SPARC (ΔBIC +2843), so neither Gaia outcome reaches a live density-keyed model. The algebraic EFE-free corner on Tier 1's TEST-25 card is a separate reading that Gaia DR4 can kill; that cannot select the framework either. (Until 2026-10-23 this card said the density branch predicted 0.05–0.4%, ~80× below the Gaia DR3 systematics floor; that figure needs a knee no other page uses, and a researcher simulated-visitor persona found it contradicting Tier 1.) Separately, Desmond, Hees & Famaey (2024) — the same paper TEST-25 inherits — pre-empts this test. Demoted here 2026-08-10 after a simulated-visitor Pass 3 found this page still selling it; the closure was on the Tier-1 ledger since 2026-06-07 and never propagated to the page a researcher opens to find live tests.

Ranked #2 among decisive tests at $150K while this same page conceded that D and S — the two parameters in the consciousness equation the experiment would measure — are "not yet operationally defined." A test whose observables have no definition cannot be ranked among the most discriminating; there is no calibration procedure mapping any EEG quantity onto the C axis, so no outcome could be scored against a prediction. The gap is upstream of the experiment and no amount of funding closes it. It stays in the catalog as Tier 2 (TEST-11 — the ID retained; the Cassini squeeze that had been colliding with it was renumbered TEST-25 on 2026-08-10) and returns to a ranking only if D and S acquire measurement protocols. Demoted after a simulated-visitor Pass 3 flagged the contradiction between the ranking and the page's own caveat.

Session 107 predicted fσ₈(z=0.51) ≈ 0.418 (suppression), kill criterion fσ₈ > 0.46 for >3σ. DESI DR1 full-shape (arXiv:2411.12021): LRG1 fσ₈/(fσ₈)_fid = 1.09 (+0.12/−0.14), Table 9 → fσ₈ = 0.516 (+0.057/−0.066) — exceeds 0.46 by only ~0.9σ, short of the registered >3σ (corrected 2026-10-10: the 1.16 ± 0.13 quoted here since May was the QSO row). Combined σ₈ = 0.841 ± 0.034 gives a 2.4σ tension, but σ₈ is a different, GR-conditioned statistic — using it to falsify a modified-growth model risks circularity. DESI's own modified-gravity analysis (Ishak et al. arXiv:2411.12026) gives a weaker verdict (μ₀ within 1σ of zero). Honest verdict: post-hoc either way; the test as registered lacked the power to discriminate this framework from GR. See /tier-1-existing and /honest-assessment for the full correction.

Framework's own Session 107 forecasts BAO matches ΛCDM at 0.0% — the sound horizon is set at z~1100 when C ≈ 1 everywhere. The 10⁻⁴ modulation number had no session-level derivation.