VoidStep-H0 presents a pre-registered, binary observational test designed to directly probe environment-dependent modulation of the local expansion rate using low-redshift Type Ia supernovae. The Hubble tension — the persistent discrepancy between early- and late-universe measurements of the Hubble constant — has reached high statistical significance and remains unresolved. Most proposed solutions rely on extensions to the standard cosmological model, often introducing additional degrees of freedom and degeneracies that limit direct falsifiability. This work adopts a fundamentally different approach. Instead of expanding model complexity, it defines a localized observable: an environmental step in standardized distance modulus, Δμₑnv, measured between supernovae in underdense and overdense regions of the large-scale structure. The experiment is explicitly pre-registered, with a fixed decision threshold and time horizon. A detection of Δμₑnv ≳ 0. 04 mag would indicate that local density contributes significantly to the observed discrepancy, while a null result with |Δμₑnv| < 0. 015 mag excludes this class of explanations at the relevant amplitude scale. The methodology is fully reproducible and relies exclusively on publicly available datasets, including Pantheon+, ZTF DR2, 2M++, and DESI DR2. A dual-pipeline architecture (Bayesian and frequentist) combined with strict blinding and pre-defined validation criteria ensures robustness against methodological bias. This repository contains the full pre-registration protocol, implementation details, and reproducible analysis pipeline, enabling independent verification and execution of the experiment. More broadly, VoidStep-H0 introduces a shift in methodology for addressing cosmological tensions: transforming model-dependent inference problems into direct, falsifiable observational tests with explicit termination conditions. This work is intended as both a scientific contribution and a reproducible experimental framework for future investigations in precision cosmology.
Eduardo Parra (Sat,) studied this question.