Answers are grounded in this paper and cite their sources.
Version v0.4 — 16 September 2026Adds Technical Amendment TA-02, which removes the preregistered pre-switch adversarial-collaboration waiting period and replaces it with independent execution followed by post-result notification of Lane, Seifert and Wiltshire.No scientific design element is changed: data, code locks, priors, endpoints, seeds, qualification gates, result classes, matched-N controls, sensitivities, secondary grid, Phase II and reporting rules remain unchanged. Preregistration of a controlled likelihood-architecture switch analysis for Pantheon+ supernova cosmology. The primary objective is to determine how much the Bayes factor between timescape and spatially flat ΛCDM changes when the hierarchical Lane latent-variable architecture is replaced by the Seifert Tripp-only fixed-x1/c architecture, while holding prespecified analysis components fixed. This version adds Technical Amendment TA-01 v1.3, deposited before any locked-data numerical evidence run. TA-01 replaces the non-portable dual source-era runtime reconstruction with one portable common Linux/Python runtime and a hashed source-logic adapter. It adds hard runtime-pin and Python-3.10 compatibility checks, source-faithful importance-nested-sampling evidence adjudication, fixed source-anchor qualification and confirmation rules, deterministic P+580 covariance derivation, resumable execution, recorded BLAS/LAPACK thread state, external-Linux execution instructions, and fail-fast scheduling. The original v0.2 preregistration remains unchanged as the pre-result scientific design. The source lock, input lock, model-comparison convention, source priors, primary controlled-switch endpoints, matched-N rule, Phase-II definitions, and downstream result-class rules remain unchanged except where TA-01 explicitly states otherwise. No Bayes factor, posterior, primary switch endpoint, matched-N result, or Phase-II result was inspected before TA-01. Original normative preregistration artifact:Preregistration_Lane-Seifert_Architecture-Switch_v0-2_DEPOSIT_MASTER.md Original v0.2 deposit-container SHA-256:34f0a0a6dc2a8d8e51fdbe02558efa6c516e9c9e31b953415f19874ab727de38 TA-01 v1.3 deposit-package SHA-256:bb9a1022aafb47e8838615a1227197bff7f4faa0a295af912957dc2b93e06804
No takes yet. Share an insight, caveat, or question.
Harald Zierhut (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: