A short progress note. No conclusions; only what is done, what changed and what comes next. The detailed account is in the documentation we are preparing, which the first publication will link to.

Done

  • Part 1 — corpus map — closed in its second iteration. Just under three hundred candidate papers on functional MRI, EEG/MEG, PET and near-infrared imaging screened: which papers carry the field’s claims, which concepts they introduced, and how the concepts support one another.
  • One paper taken apart step by step. For one of the most-cited resting-state papers we reconstructed the whole chain of inference: what was measured, what follows from it, what is assumed without being checked, and where the forks are — points where a different assumption leads to a different research design. The same method was then applied to three pairs of papers, each an original claim and its later check.
  • How a claim travels. We followed what citing papers took from one article — the most-cited, a random sample and a split by genre — then the review articles, then the public channel (encyclopaedia, popular science). The question behind it: where does a claim settle as established in this field.
  • Part 2 — physics of measurement — opened. For each method we separate what the instrument physically measures from the step at which the model begins. The first rows are filled from the field’s own methodological papers.
  • Part 3 — statistics — opened from four pilot reconstructions: for each result we ask whether the test could have failed had the claim been false.

What changed in the method

  • Every number traces to a saved query with a date and a checksum.
  • We keep a list of our own errors: where we were wrong, it is recorded the same way as the field’s errors.
  • The criteria we apply to the field have been turned on our own numerical projects.

Next

Three short pieces — one per site — await the author’s reading and corrections; they appear in the coming weeks. After them, the full documentation with a public repository where the whole audit trail can be inspected.