Falsifiable Science

Experimental Predictions

A framework that makes no predictions is not physics. These are specific, numerical, falsifiable predictions derived from the GPP framework — not post-hoc fits.

Dark Matter

Dark Matter Core Radius

Pending — Euclid ~2027
rc = 9.98 – 9.99 kpc

The T-symmetric dark matter model predicts a universal soliton core radius in the range 9.98–9.99 kpc for Milky Way–scale halos. This is derived from the shadow kernel structure at the conformal boundary, not fitted to rotation curves. The Euclid satellite will constrain halo core radii to sub-kpc precision; this prediction will be falsifiable by approximately March 2027.

Test window: Euclid DR1 ~ March 2027 • Paper: GPP Dark Matter Preprint

Dark Matter Self-Interaction Cross-Section Scaling

Pending
σ/m ∝ v−4

The framework predicts a velocity-dependent self-interaction cross-section scaling as v⁻⁴ in the dwarf galaxy regime, a distinctive signature that distinguishes this model from WIMP and axion alternatives. Current Bullet Cluster and cluster-scale constraints are consistent with this scaling.

Test window: Future cluster surveys • Status: Consistent with existing data

Dark Energy

Dark Energy Equation of State

Pending — DESI 2026
w0 = −0.85    wa = −0.21

The shadow-paired vacuum energy calculation yields a dynamical dark energy equation of state with w₀ = −0.85 and wₐ = −0.21 in the Chevallier–Polarski–Linder parametrisation. This departs from ΛCDM (w = −1) and is consistent with the 2024 DESI BAO results suggesting w > −1 at low redshift. The DESI full survey will constrain these parameters to ~5% precision.

Test window: DESI Year 3 ~ 2026 • Status: Directionally consistent with DESI 2024

Gravitational Waves & Cosmology

NANOGrav Gravitational Wave Background Spectral Index

Pending
nt = 2/3

The T-symmetric cosmology predicts a stochastic gravitational wave background with spectral index nt = 2/3, arising from the conformal structure of the T-boundary. The NANOGrav 15-year dataset is consistent with this but not yet sufficient to distinguish it from other models. Future PTA data and LISA will constrain this to percent-level precision.

Test window: IPTA + LISA ~ 2030s • Status: Consistent with NANOGrav 15yr

CMB Hemispherical Asymmetry Amplitude

Pending
A = 0.07

The preferred axis from the T-boundary structure predicts a hemispherical power asymmetry A = 0.07 in the CMB temperature maps, consistent with the observed Axis of Evil anomaly. The angular separation between ℓ = 2 and ℓ = 3 multipoles is predicted to be Δθ₂₃ = 5°. Both values are derived from the shadow kernel geometry, not fitted.

Test window: LiteBIRD ~ 2030 • Status: Consistent with Planck data

Particle Physics

Neutrino Mass Sum & Lightest Mass

Pending — KATRIN / Euclid
mν1 = 0    Σmν < 0.12 eV

The T-boundary Majorana condition forces the lightest neutrino mass to zero: the lightest neutrino is exactly massless, protected by the topology of the T-boundary. This gives normal ordering with mν1 = 0, which combined with measured mass splittings gives Σmν ≈ 0.059 eV. The Majorana condition is also a prediction: neutrinoless double beta decay exists.

Test window: KATRIN full dataset 2027; Euclid neutrino mass sum 2027 • Status: Consistent with current upper limits

Lowest Glueball Mass

Pending — LHCb / PANDA
M0++ = 2N/(k+N) · ΛQCD

The Yang–Mills mass gap formula M = 2N/(k+N)·ΛQCD with N = 3 (SU(3) colour) and Kac–Moody level k = 4π/g² evaluated at the QCD scale predicts the scalar glueball mass. For k ≈ 1 this gives M0++ ≈ 1.5–1.7 GeV, consistent with the f₀(1500) candidate. The ratio M0++QCD is a parameter-free prediction.

Test window: PANDA at FAIR ~ 2026+ • Status: Consistent with lattice QCD and f₀(1500)

Majorana Neutrinos (Neutrinoless Double Beta Decay)

Pending — nEXO / LEGEND
0νββ exists

The T-boundary condition forces ν = ν̄ᶜ (Majorana condition) without any additional assumptions beyond the framework. Neutrinoless double beta decay must therefore occur. The effective Majorana mass mββ is predicted to be in the range 1–5 meV for normal ordering with mν1 = 0, at the sensitivity floor of next-generation experiments.

Test window: nEXO ~ 2030; LEGEND-1000 ~ 2028 • Status: Predicted positive signal

Note on Methodology

All predictions on this page are derived from the framework before comparison with data, not fitted to existing measurements. Where a prediction is directionally consistent with existing data, this is noted as confirmatory evidence but not treated as verification — that requires the dedicated future experiments listed above. The framework will be considered falsified if any confirmed experimental result contradicts a bold prediction listed here.

View Underlying Papers