# Adversarial review — psychedelic-microdosing-symptoms (ROUND 2)

Authoring model: **Sol 6.1**. Adversarial reviewer: **Claude Opus 5.5 (Anthropic)**. Date: **2026-10-14**.
Canonical file reviewed: `hsef_v53_traffic_ranks_16_20_2026-10-13/psychedelic-microdosing-symptoms/psychedelic-microdosing-symptoms.review.md`.

## Summary

| Item | Result |
|---|---|
| Level before → after | Γ → Γ (no change) |
| Confidence before → after | Moderate → Moderate (no change) |
| Causal role | epiphenomenon (unchanged) |
| Findings | Critical 0 · Major 0 · Minor 5 (11 distinct string corrections, 17 occurrences) · Editorial 0 |
| REFERENCE PMIDs verified | 64/64 resolve; 0 failures; 0 retraction/EoC flags |
| DOIs cross-checked | 64/64 match PubMed |
| Load-bearing claims spot-checked | 16 |
| Collective-author builder bug | not present |
| Validator (package-bundled v2.3) | PASS — 0 fail, 0 warn, 30 pass (`validator.txt`) |

## Findings

| ID | Severity | Location | Original → corrected |
|---|---|---|---|
| 1 | Minor | Deciding experiment / PROPOSED DECISIVE EXPERIMENTS / Boundary experiment (3 identical copies) | "base-equivalentLSD twiceweekly" → "base-equivalent LSD twice weekly"; "with equalcontact/co-care" → "with equal contact/co-care"; "follow throughwithdrawal and relapse" → "follow through withdrawal and relapse" |
| 2 | Minor | AT A GLANCE, Best-supported rival | "home LSD36997080" → "home LSD 36997080"; "diagnosed-MDD41202873" → "diagnosed-MDD 41202873" |
| 3 | Minor | Competing models, Rank 1 | "from pre/postnull" → "from pre/post null" |
| 4 | Minor | Competing models, Rank 2 | "psychedelic andketamine" → "psychedelic and ketamine"; "dependency;active-controlled 23982301 andwithdrawal" → "dependency; active-controlled 23982301 and withdrawal"; "LSDmicrodose" → "LSD microdose" |
| 5 | Minor | Competing models, Rank 3 | "MDD41202873" → "MDD 41202873"; "restricted 48 hour highBDI38042914 positive" → "restricted 48-hour high-BDI 38042914 positive" |

All five are instances of the known builder rendering defect (lost spaces around numerals and between
words). Approximately 130 further sites of the same class remain in the technical body; they are
recorded for an upstream generator fix rather than hand-patched (see the review's ADVERSARIAL REVIEW
AND APPROVAL section).

## Rejected findings (checked, no change)

1. Reference #30 (PMID 33591231) prints 2022 against PubMed's 2020 ahead-of-print date — the print issue
   is 2022;36(1), so the review is correct.
2. Aggregation: pre-registered primary-subclaim with SC2 primary; headline Γ equals SC2. Reported
   alternatives (weakest-link Α, scope-matched Γ) are both correct — SC2 is the only scope-matched
   subclaim.
3. Momentum "Mechanistic only" carries the required artifact and is correct; the decisive trial is
   completed, not underway.
4. Dose-unit conversions (13/26 µg tartrate = 10/20 µg base; 29 µg tartrate = 20 µg base) verified
   verbatim against the stored original Methods of PMID 35106880 and PMID 40105807.
5. Framework-root validator v2.4 flags `v5.lay_summary`, mandatory only from v5.4; this is a v5.3
   review and the package validator passes.

## Approval

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