Evidence & Research
Methodology & Sources
How a record is built, what it is allowed to say, and what happens when the source does not answer the question.
Record construction
Six steps, in order
The same sequence is applied to every record, whichever publication it originates from.
- 01
Identify the tested intervention
A record is anchored to what was actually administered — a purified compound, a defined extract, a named strain, a formulation, a dose and a route — not to a common mushroom name.
- 02
Assign the evidence owner
The result belongs first to that intervention. Transport to any other preparation is a separate question, recorded separately.
- 03
Grade maturity, direction and replication
How far the proposition travelled into direct human testing, which way it pointed, and how often it survived meaningful independent retesting — kept apart so a rigorous negative is never read as weak evidence.
- 04
Measure the claim gap
The distance between the public sentence and the experiment underneath it is graded on its own scale.
- 05
Record what remains uncertain
Each record names its dominant uncertainty and, where the source states one, the experiment that would resolve it.
- 06
Leave unknown fields empty
Where the source material does not establish a value, the field is left blank and flagged for review. Nothing is filled in to complete a layout.
Sources
Where the current records come from
The evidence records published on this site are derived from the Foundation Evidence Map blocks of The Fungal Evidence Atlas (Richard L. Pope, The Spore Patch Ltd, 2026), FA-089 Publication Master.
Grades, propositions, ownership statements and uncertainties are reproduced from that source rather than reinterpreted for the web.
Primary references are reproduced within each approved full record in the Evidence Library. References are transcribed from the verified Atlas source material and remain attached to the proposition and intervention they support.
Where a later source, taxonomic revision or correction changes a record, the change is documented in the correction record rather than silently overwritten.
Protocol
Audit protocol and editorial provenance
What kind of review this is, how a record travels from source text to published record, and where the boundaries of the method lie. Sections that depend on facts not yet confirmed for publication are not shown rather than approximated.
Type of review, and its limits
This is an editorial evidence audit. It asks what a body of published work actually demonstrates about a defined proposition, and where a claim has travelled beyond the experiment that produced it. It is not a systematic review, a meta-analysis or a clinical guideline, and it does not produce treatment recommendations.
Inclusion and exclusion principles
A study is included where it bears on a defined proposition about a defined intervention. Work that concerns a different preparation, organism, population, outcome or exposure is not folded into that proposition; it is recorded as a separate question, or as a transfer whose confidence must be stated.
Grouping papers into research programmes
Replication is assessed proposition-by-proposition, not paper-by-paper. Repeated publications from the same programme, or studies of a materially different preparation or endpoint, do not raise a replication grade on their own.
Extraction and verification
Every record is extracted from the source chapter, then held as an unverified extraction until it is reconciled field by field against that chapter. A record moves to checked, and only then to approved for publication. Nothing is bulk-imported, and unapproved extractions are never served publicly.
Use of AI, and where it stops
AI tools are used to assist first-pass extraction of fields from the source chapters and to help draft and build this website. They are not permitted to set or alter a grade, direction, replication value, safety or conservation position, reference or conclusion. Every field is reconciled by a human against the source chapter before a record can be marked checked, and publication requires a further human approval. Where the two disagree, the source chapter governs.
Provenance of each value
Individual values carry their own provenance: transcribed directly from the source, derived from it under a stated rule, or human-verified against the underlying material. Provenance travels with the value rather than with the record as a whole.
Inaccessible or incomplete sources
Where the source does not establish a value, the field is left unfilled and flagged, and a verified absence is distinguished from a field that has simply not been reviewed. A field that does not apply to a record is marked as such rather than left ambiguous.
Correction and regrading
A grade may be revised when new evidence, a taxonomic revision or an identified error requires it. The change is recorded in the correction history, with the previous position preserved rather than silently overwritten.
Responsibility
Published grades, propositions, references, ownership statements and corrections are the responsibility of Richard L. Pope and The Spore Patch Ltd, and are checked against the relevant source material before publication.
Methodological limitation
The Atlas framework is an editorial evidence-audit system developed to make evidential boundaries and transfers visible. It should not be understood as an externally validated clinical instrument, formal risk-of-bias tool or replacement for specialist systematic-review methods.
Standard of truth
Accuracy before completeness
An empty field is information. It records that the question was asked and the source did not answer it. A plausible sentence written to fill that space would destroy the only thing this catalogue is for.