# Adversarial review log — zone2-cardiorespiratory-fitness

Authoring model: **Sol 6.1**. Adversarial reviewer: **Claude Opus 5.5 (Anthropic)**. Date: **2026-10-14**.
Round 2 of the adversarial pass (brief: `deliverables/ADVERSARIAL_REVIEW_BRIEF_2026-10-06.md`, ROUND 2 section).

| Item | Value |
|---|---|
| Level before | Γ — Mechanistically plausible, partial system support |
| Level after | Γ — unchanged |
| Confidence | Moderate — unchanged |
| Role | reinforcing / stabilizing factor — unchanged |
| M / S / D | Mechanistic only / Stable / Contested — unchanged |
| PMIDs checked | 63 / 63 resolve to the stated title, first author, journal and year |
| PMIDs failed | 0 |
| Retractions / EoC | none found |
| Load-bearing claims spot-checked | 14 |
| Findings | 1 Major, 1 Minor, 1 Editorial (78 instances) |
| Validator | PASS — 0 fail, 0 warn, 30 pass (`validator.txt`) |

## Findings

| ID | Severity | Location | Change |
|---|---|---|---|
| A1 | Major | Cap audit "SC2 cap"; score card `cap_rules_applied[SC2]` | Universal cap 4 was cited as the binding cap for SC2 although SC2's outcome is measured VO2max (a functional phenotype) and the document's own SC2 Γ hard-rule audit records that rule as not triggered. Replaced with the correct basis: no cap binds SC2; its Γ ceiling comes from the boundary check (no field-valid positive perturbation at the registered scope). Score-card row set `triggered: false`, `cap_source: "none"`. Level unchanged. |
| A2 | Minor | Evidence and causal interpretation, PMID 2072834 | "improved VO2max by 7% relative to nontraining controls" → "improved VO2max by 7%, significantly more than the change in nontraining controls". 7% is the within-group gain, not the control-adjusted difference. |
| A3 | Editorial | Throughout (78 instances) | Rendering defects: missing spaces around numerals, "ErratumIn: PMIDnot supplied" (×2), empty "DOI: ." fields (refs 9 and 51 — confirmed to have no DOI in PubMed), comma-run PMID lists (×3), "UKBiobank" in prose. |

## Systematic artifacts reported by the lead — all four checked

1. **False "earlier adversarial approval" boilerplate** — **absent.** This is a fresh review, not a 9 October revision; it carries no spliced Consensus-consistency block and no "earlier approval" clause.
2. **Confidence set from access depth** — **present as wording, band confirmed correct.** The score card's `confidence_rationale` reads "63 primary reports with 2 selected partial-body examinations; density, heterogeneous controls and incomplete access set the confidence, not correctness", which names reviewer access as a confidence input although the framework defines confidence by evidence density and reproducibility. The band was independently re-derived against the High text ("extensive independent replication across labs, models and assay modalities; reproducibility well-established") and **Moderate is correct**: the registered exposure has one small positive trial and one modern null, with no independent replication at the registered scope. **Superseded 2026-10-14:** the lead has since directed that the rationale be rewritten per case; that edit is finding A4 and is described under “Lead policy applied” above. The band is unchanged.
3. **"Best-supported rival" renderer bug** — **absent.** `models_evaluated[0]` is "Training dose and realized exposure govern adaptation" with `rank_in_tournament: 1` and `is_nominal: false`; the nominal ("Below-first-threshold aerobic training increases fitness") is the third entry. The prose names the correct rank-1 rival.
4. **Reviewer-notes / Consensus-splice contradiction** — **absent.** No spliced block exists, and the REVIEWER NOTES section affirms Consensus use consistently with the reconciliation section.

Score card `review_metadata.reviewer` left as issued; corpus-wide house provenance.

## Lead policy applied — confidence bands (2026-10-14)

Two-step test as issued by the lead. **Step 1, strike access depth as a determinant.** The score card's `confidence_rationale` named reviewer access ("incomplete access set the confidence"). How much of a paper the reviewer could read is neither evidence density nor reproducibility and never justifies a band, so the sentence was rewritten to drop it and to say explicitly that access depth is recorded per ledger row and is not a determinant. **Step 2, re-derive from what remains.** What remains is a substantive evidence-quality objection, not a reading-access one: at the registered below-first-threshold scope there is one small positive controlled trial and one modern randomized null, with no independent same-scope replication, and the 63 reports use heterogeneous comparators, prescription methods and control conditions. That is nowhere near the High text's "extensive independent replication across labs, models, and assay modalities". The band therefore **stands at Moderate**; access depth was not the only thing holding it down.

**Momentum re-checked against the same clarification** (a completed publication or trial never satisfies "Decisive work underway"): already "Mechanistic only"; no change needed.

## Rejected findings (checked, no change)

1. **Headline Γ rather than Ε or Δ.** Belman and Gaesser (PMID 2072834) is a controlled trial whose low-intensity arm (~72% of the study's lactate threshold) gained VO2max significantly more than control; granting it a Δ crossing in older adults and pairing it with the null moderate-domain arm of Inglis et al. (PMID 38376995) in young adults would read as Ε. Rejected — the framework does not force it, because the registered exposure is "below the *first* measured threshold" and the truncated 1991 abstract does not state the threshold algorithm. Flagged as the review's leading residual uncertainty and as the Consensus question.
2. **Citation years** on references 10, 18, 22, 23, 34, 40, 54, 55, 59 differ by one year from the PubMed e-pub year — rejected per the standing convention.
3. **Divergence "Contested" without rung arithmetic** — rejected; Contested is defined as the case where no single implied level applies.
4. **Momentum "Mechanistic only"** — the required artifact is named (PMID 40198065, 42560532, 41221589). No change.
5. **Aggregation** — primary-subclaim correctly returns Γ from SC2; the declared alternatives (weakest-link Β, scope-matched Γ) are both correct. No change.
6. **Collective-author reference defect** — none present.

## Files changed in this case folder

- `zone2-cardiorespiratory-fitness.review.md` (banner + ADVERSARIAL REVIEW AND APPROVAL section + A1–A3 corrections)
- `zone2-cardiorespiratory-fitness.scorecard.json` (A1; and "approximately72%" → "approximately 72%" in narrative fields)
- `validator.txt` (new)
- `sol61-original/` holds the pre-review copies of the review and score card.

Package-level files (`OVERVIEW.md`, `MANIFEST.json`, `READER.html`, `BATCH_COMPLETION.json`) were **not** edited; the lead should propagate per the final report.

**Approval:** Following adversarial review and the corrections above, the revised document is approved. — Claude Opus 5.5 (Anthropic), 2026-10-14.