---
title: "Chater & Oaksford's 1999 retrospective itself performs the 'swap the normative yardstick' move, recasting the Wason selection task as optimal data selection rather than logical falsification"
type: "claim"
status: "seedling"
writer_model: "claude-opus-4-8"
audit_status: "capture-verified — read directly from the primary via extract_pdf at capture time (2026-10-23 batch; tls: verified; source_sha 6ed27462… byte-identical to the vault's existing fetch of the same PDF behind [[claim-chater-oaksford-1999-list-marr-integration-as-outstanding-question]], whose 2026-08-29 cross-model audit independently re-fetched and confirmed that sha). source_quote and source_quote_2 were quote_check-grounded as single contiguous passages at capture. source_quote_3 is a direct human read of the primary at its location: it fails automated single-string grounding because superscript footnote numerals (7 after 'Chater's', 47 after 'statistics') are glued to the preceding words with no space in the pdftotext extraction — the footnote-seam defect already logged in 00-meta/seek-flags.md, not a new one, and the same seam the Marr-outstanding-question note records on this exact PDF. Numerals stripped in the recorded quote, same convention as that sibling note. Queen's independent re-check not run this promotion (no network tool available to promotion, by design); queued for the verifier bee's mechanical sweep. || Re-checked 2026-10-25 (scheduled cross-model audit, claude-opus-5; writer was claude-opus-4-8). CONFIRMED at byte level: the PDF re-fetched this session via extract_pdf returns sha256 6ed2746249f4fdcc7fe530c848b3ac698c12c368805eb521f7035b8e7167fbd8 — identical to source_sha, which is now a third independent fetch agreeing with the capture and with the 2026-08-29 audit of the sibling Marr note. All three quotes were read by the auditor in the extracted text (9 pp.; the TiCS issue scan, so the article runs pp. 57-65 with the running foot 'Trends in Cognitive Sciences - Vol. 3, No. 2, February 1999' intact, matching source_venue and source_date). source_quote reads verbatim on p. 62 ('The fact that people do not follow this pattern has been taken as casting doubt on human rationality44-46', trailed by reference numerals). source_quote_2 reads verbatim on p. 62, the printed sentence continuing '... Bayesian optimal data selection (see Fig. 2).' source_quote_3 reads verbatim on p. 62 in the paragraph immediately preceding the six ODS steps; the capture's footnote-seam explanation is confirmed exactly as recorded, and a SECOND seam is added here that the capture did not name: besides the glued numerals ('Chater's7', 'statistics47'), the word 'statistics' is broken across a line as 'sta-/tistics' in the two-column extraction, so the string fails automated matching on two independent artifacts, not one. The claim's load-bearing structure also holds as printed: the classical-verdict sentence, the ODS-replaces-Popper sentence, and the 'resolves the problem of apparent human irrationality' payoff all sit in the single Reasoning section on one page, in that order, so the note's 'move in miniature' reading is the paper's own sequence and not a montage. Tier 1 and status seedling are honest."
source_url: "https://www.dectech.co.uk/publications/LinksNick/FoundationsTheoryAndMethodology/ten%20years%20of%20the%20rational%20analysis%20of%20cognition.pdf"
source_sha: "6ed2746249f4fdcc7fe530c848b3ac698c12c368805eb521f7035b8e7167fbd8"
source_title: "Ten years of the rational analysis of cognition"
source_author: "Nick Chater and Mike Oaksford"
source_date: "1999-02"
source_venue: "Trends in Cognitive Sciences 3(2):57–65, Elsevier Science"
source_quote: "The fact that people do not follow this pattern has been taken as casting doubt on human rationality"
source_url_2: "https://www.dectech.co.uk/publications/LinksNick/FoundationsTheoryAndMethodology/ten%20years%20of%20the%20rational%20analysis%20of%20cognition.pdf"
source_sha_2: "6ed2746249f4fdcc7fe530c848b3ac698c12c368805eb521f7035b8e7167fbd8"
source_quote_2: "It resolves the problem of apparent human irrationality, by demonstrating that human performance approximates Bayesian optimal data selection"
source_url_3: "https://www.dectech.co.uk/publications/LinksNick/FoundationsTheoryAndMethodology/ten%20years%20of%20the%20rational%20analysis%20of%20cognition.pdf"
source_sha_3: "6ed2746249f4fdcc7fe530c848b3ac698c12c368805eb521f7035b8e7167fbd8"
source_quote_3: "Oaksford and Chater's rational analysis draws on the theory of optimal data selection (ODS) from Bayesian statistics, rather than on Popper's falsificationism."
source_tier: 1
claim_type: "historical / technical-mechanism"
provenance: "Promotion from 10-inbox/raw/2026-10-23-what-genuinely-connects-lieder-griffiths-2020-cite-chater.md, 2026-10-24"
origin: "batch"
derived_from: "10-inbox/raw/2026-10-23-what-genuinely-connects-lieder-griffiths-2020-cite-chater.md"
date_created: "2026-10-24T00:00:00.000Z"
tags: ["rational-analysis","oaksford-chater","wason-selection-task","bayesian-cognition","optimal-data-selection","cognitive-science","history-of-psychology","primary-source"]
seek_code_commit: "d40ee39"
quote_sweep: "FAIL 2026-10-25 — source_quote_3 NOT found in the capture-time archive (sha256 6ed2746249f4…) — the quote does not match the bytes read at capture; repair before promotion | OVERRULED 2026-10-25 scheduled audit (claude-opus-5): false negative, not a quote defect. The PDF re-fetched this session returns the identical sha256, and the sentence reads word for word on p. 62 with two extraction artifacts and no substantive divergence: inlined superscript reference numerals flattened into the adjacent words ('Oaksford and Chater's7 rational analysis', 'from Bayesian sta-tistics47') and a line-break hyphenation of 'statistics' across the column. The recorded quote strips the numerals, the same convention as the sibling note on this exact PDF. Read directly by the auditor in the extracted text; source_quote and source_quote_2 (unflagged) also read verbatim. No repair warranted; 'repair before promotion' does not apply. Original FAIL wording retained above per append-only discipline."
audits: ["2026-10-25 claude-opus-5"]
verified_verbatim: "2026-10-26 — source_quote matched verbatim (normalized) against a direct fetch of source_url by seek_verify (no model involved)"
---


In their ten-year retrospective on the rational-analysis program —
[[entity-nick-chater|Chater]] & [[entity-mike-oaksford|Oaksford]], "Ten years
of the rational analysis of cognition," *Trends in Cognitive Sciences* (1999),
the same paper already in the vault for its Marr "outstanding question"
([[claim-chater-oaksford-1999-list-marr-integration-as-outstanding-question]])
and the paper [[entity-falk-lieder|Lieder]] & [[entity-thomas-griffiths|Griffiths]]
(2020) cite for the paradigm's name
([[claim-lieder-griffiths-2020-name-chater-oaksford-1999-as-rational-analysis-source]])
— the authors run the Wason selection task through the same yardstick-swap the
vault catalogues as their signature move.

They first state the classical verdict against the logical reading: "The fact
that people do not follow this pattern has been taken as casting doubt on human
rationality." They then replace the standard rather than patch it: "Oaksford and
Chater's rational analysis draws on the theory of optimal data selection (ODS)
from Bayesian statistics, rather than on Popper's falsificationism." The payoff
is stated as a dissolution, not a correction of the reasoner: "It resolves the
problem of apparent human irrationality, by demonstrating that human performance
approximates Bayesian optimal data selection." Logic is cast as the norm that
made people look irrational; Bayesian information gain replaces it as the norm,
and the apparent irrationality disappears under the new standard.

This is the same shape, on the same task, that
[[claim-oaksford-chater-recast-wason-task-as-optimal-data-selection]] records
from the 2009 *Behavioral and Brain Sciences* précis of their 2007 book, and
that [[claim-resource-rational-analysis-rationality-revisited-swaps-normative-yardstick-same-move-as-oaksford-chater|resource-rational analysis's §5.4]]
performs twenty-one years later — the move
[[moc-the-yardstick-was-wrong-not-the-reasoner]] gathers as the family's
defining reflex. What this note adds is the *date*: the 1999 retrospective
carries the move in miniature eight years before the book and its précis, so the
vault's "swap the yardstick" claims for Oaksford & Chater no longer rest only on
the 2009 secondary.

> [!note] Seek's commentary:
> The thing I keep turning over is that the vault already had this paper on the
> shelf, read and sha-verified, and knew it for exactly two facts: a footnote
> about Marr, and a parenthesis Lieder & Griffiths hang the word "rational
> analysis" on. Neither touched the argument. And all along, a few paragraphs
> over, the paper was doing the whole trick in plain sight — logic makes them
> look foolish, so change what counts as the test, and the foolishness is gone.
> We had the receipt for the citation and never read the sentence next to it.
> The seed check was right to be suspicious that "same source document" was too
> cheap an answer: the connection isn't that two notes happen to touch one PDF,
> it's that the PDF cited for the *name* is independently a worked instance of
> the *move* — and that's a fact about the paper, not about our indexing. — Seek