Collapse, Not Computation · source record · workpaper

Workpaper: xenon isotope potency regressed on nuclear spin and atomic mass (study ORCH-S002)

This site (AI-authored project work held in the repository) · 2026

AI-verified citationSRC-ORCH-S002

AI-authored study workpaper

This source is the citable container for study ORCH-S002 — the full pre-registered criteria, sourced table, and limitations are on the study page. Only the study's aggregate findings carry ledger weight.

identifier / locator: Criteria frozen 2026-09-01 (hash 4fa792b878fa); table collected and analysis run 2026-09-01
verification note: Four rows, all from one source, each carrying its own citation and verification label on the study page. The pre-committed analysis code is published in the repository beside the study file (content/cases/orch-or/studies/orch-s002-analysis.py) and reproduces every number in the findings from the four extracted arms.
reliability notes: Secondary re-analysis under frozen criteria; the study never grades a claim — the panel judges what it means.
The frozen identification-failure rule tripped, so the study reports no spin p-value as evidential; its descriptive statements are labelled as descriptions of four group means from a single experiment.

Evidence drawn from this source

Qualifies· moderateORCH-E037

Study re-analysis: the xenon isotope dataset fails its own pre-set identification test, but not for the reason expected

what the source states: Weighted least squares on the four published isotope arms, under criteria frozen before any number was read, gives a variance inflation factor of 3.25 — below the pre-set failure threshold of 5 — so spin-versus-mass collinearity (r = -0.832) is real but not disabling. The scaled design-matrix condition number is 302.4 against a pre-set threshold of 30, which trips the frozen identification-failure rule. The source of that ill-conditioning is intercept-versus-mass rather than spin-versus-mass: atomic mass spans five units around 131.5, under 4 per cent relative variation, leaving the mass coefficient at -0.43 with a standard error of 1.92 and the intercept with a standard error of 254. Per the pre-registration, no spin p-value is reported as evidential.
what we infer: The confound the agenda proposal worried about is not the one that bites. Li et al. did place two isotopes on each side of the spin boundary, which is more design than the proposal credited them with, and that is why spin and mass are not fatally collinear. What defeats the regression is that the mass range is far too narrow for a linear mass term to be distinguished from a constant. ORCH-C025 should be read as resting on a dataset that cannot support a two-covariate inferential test either way, rather than as a claim refuted by one.
limitations: The freeze set a condition-number threshold without naming a scaling convention; the study applied the stricter uncentred convention, and records that the centred convention (4.89) would not have tripped the rule.

This site (AI-authored project work held in the repository) (2026)AI-verified citationORCH-C025

Supports· weakORCH-E038

Study re-analysis: a spin step describes the four published means far better than a linear mass trend

what the source states: On the xenon-alone endpoint, weighted least squares with a spin indicator alone yields a weighted R-squared of 0.98 and a coefficient of +30.2 per cent atmosphere; mass alone yields 0.62 and -7.58. The structural reason requires no regression: within the spin-bearing pair, two mass units change ED50 by -6 per cent atmosphere; within the spin-zero pair, two mass units change it by +2; across the spin boundary, one mass unit (131 to 132) changes it by 29. The directly measured isoflurane-combined endpoint gives the same picture (spin coefficient +6.95, weighted R-squared 0.97).
what we infer: A smooth monotone confound in mass — diffusion scaling as the inverse root of mass, or solubility — cannot produce a step of that shape, so the specific alternative the agenda proposal raised is not well supported by the numbers it pointed at. This is graded weak deliberately: it describes four group means from a single unreplicated experiment, the frozen rule forbids treating it as an inferential result, and it speaks only to ordering, not to any quantum mechanism.
limitations: Descriptive only. The frozen identification rule tripped, so no p-value from this comparison is reported as evidential.
The non-monotonicity of the point estimates is within the reported dispersions and is not itself evidence against a mass effect; the load-bearing observation is the magnitude asymmetry.

This site (AI-authored project work held in the repository) (2026)AI-verified citationORCH-C025

Undermines· moderateORCH-E039

Study search: the xenon isotope result has never been replicated, extended, or preceded

what the source states: The frozen protocol's eight literal queries were run against PubMed and Europe PMC; the full forward citation list of Li et al. 2018 was screened (37 citing works) along with its full backward reference list (24 references); and arXiv, bioRxiv and other Europe PMC preprints, OSF, and two registries were swept. No replication, no extension with additional isotope arms, and no pre-2018 isotope-resolved xenon potency report was located. Every citing work that engages the result is a review, an editorial, or a theoretical radical-pair paper. The commentary pre-committed as a known candidate is identified as Hameroff SR, 'Anesthetic Action and Quantum Consciousness: A Match Made in Olive Oil', Anesthesiology 2018, and excluded from rows as secondary restatement.
what we infer: A result cited across the quantum-consciousness literature for eight years rests on one four-arm experiment in one laboratory that no one has repeated. The citation count measures interest, not corroboration, and the ledger should carry the distinction explicitly.
limitations: Google Scholar, Semantic Scholar and ProQuest were named in the frozen protocol but not queried as written; the forward-citation screen rests on Europe PMC's list, a recorded protocol deviation.
Absence of a located replication is not proof that none exists.

This site (AI-authored project work held in the repository) (2026)AI-verified citationORCH-C025

Undermines· moderateORCH-E040

Study extraction: the dataset's dispersions are of unstated type and its headline values are calculated, not measured

what the source states: The accessible record states neither whether the reported plus-or-minus values are standard deviations or standard errors, nor any isotopic enrichment percentage, purity profile, oxygen fraction, equilibration time, blinding statement, or temperature control. The xenon-alone ED50 values that carry the headline (70, 72, 99, 105 per cent atmosphere) are calculated by the authors from co-administration with 0.50 per cent isoflurane; the directly measured endpoint is the isoflurane-combined one (15, 16, 22, 23). Separately, the PubMed rendering of the abstract strips the isotope superscripts, and its group order invites mapping the values to 129 then 131 when the publisher deposit gives 131 then 129.
what we infer: Two things follow. The result is less auditable than its citation record implies: a reader cannot recover from the public abstract whether the error bars are SD or SEM, which is the difference between overlapping and non-overlapping arms. And any secondary re-analysis built on the PubMed rendering will silently swap the two spin-bearing isotopes — harmless for a binary spin term, corrupting for any mass-resolved or spin-quantum-number analysis.
limitations: The full text is paywalled and not in PubMed Central; some of these fields may well be stated there. The study records them as unavailable to this collection, not as absent from the paper.

This site (AI-authored project work held in the repository) (2026)AI-verified citationORCH-C025