---
id: "20261023-0205-what-genuinely-connects-lieder"
title: "What genuinely connects Lieder & Griffiths (2020) citing Chater & Oaksford (1999) for the name 'rational analysis' and resource-rational analysis's §5.4 yardstick-swap"
type: "capture"
status: "promoted"
promoted_to: ["[[claim-chater-oaksford-1999-retrospective-itself-performs-yardstick-swap-on-wason-task]]","[[observation-chater-oaksford-1999-name-citation-section-5-4-swap-confirmed-bridge]]"]
not_promoted: ["Capture Claim 1 (the 1999 retrospective itself performs the yardstick-swap on the Wason task) → PROMOTED as [[claim-chater-oaksford-1999-retrospective-itself-performs-yardstick-swap-on-wason-task]]. Tier-1, capture-verified, seedling; carries the honest footnote-seam flag on source_quote_3 and is queued for the verifier bee. Distinct from the Marr note on the same PDF (different fact), from the 2009-précis-sourced Wason-recast note (same move, different source), and from the §5.4 note (same move, different paper).","Capture Claim 2 (the connection is specific to this 1999 paper, not a general fact about Oaksford & Chater's oeuvre) → FOLDED, not separately minted. It is the scope/significance of Claim 1 (now the claim-note's 'date' paragraph: the move predates the 2009 précis by eight years) plus the bridge argument (now the observation note). Splitting it off would violate 'atomic is not many' — it cannot be true or false independently of Claim 1.","Capture Claim 3 (the bridge is real but one notch below full novelty; the MOC already catalogues both halves; vault_novelty max_cosine 0.82) → FOLDED into [[observation-chater-oaksford-1999-name-citation-section-5-4-swap-confirmed-bridge]] as its novelty hedge (verdict: confirmed-but-tightening). The vault-search/embedding figures (0.88 seed cosine, 0.82 duplicate-risk max_cosine) were kept OUT of assertive note prose per the 'metadata/vault-search claims go on the question, not 30-notes' rule (5a) and recorded only as non-load-bearing [unverified-quant].","Further leads (footnote-60 syllogistic-reasoning third domain of the same logic→probability swap; the 1999 'Outstanding questions' box's limits-of-rational-analysis question; the disposition question about folding Claim 1 into the Marr note) → NOT promoted: leads/dispositions for future sessions, not claims. The fold-into-Marr disposition was decided AGAINST — Claim 1 is a distinct fact (the paper's Wason/ODS argument) from the Marr outstanding-question fact, so it earns its own note.","Entity candidates → John R. Anderson: no Anderson-specific new fact (the Marr note already records the acknowledgements crediting him), [[entity-john-anderson]] NOT updated. Nick Chater / Mike Oaksford: a genuine new fact (their 1999 retrospective itself performs the move) — [[entity-nick-chater]] and [[entity-mike-oaksford]] each got one dated 2026-10-24 line. Lieder / Griffiths / Wason: confirmation only, no update owed.","No new 50-questions note routed: the only verification gap is the footnote-seam grounding of source_quote_3, which the verifier bee's mechanical sweep already covers by design (no load-bearing doubt the kept claim rests on). No open question in 50-questions matches this capture's topic, so no closure to rule on."]
origin: "batch"
writer_model: "claude-sonnet-5"
date_created: "2026-10-23T00:00:00.000Z"
provenance: "Batch run 2026-10-23, bridge-seed check against topic pairing [[claim-lieder-griffiths-2020-name-chater-oaksford-1999-as-rational-analysis-source]] x [[claim-resource-rational-analysis-rationality-revisited-swaps-normative-yardstick-same-move-as-oaksford-chater]] (cosine 0.88, unlinked). 10-inbox/raw/ was grepped for the exact slug pair before research began; no sibling capture mentions both exact slugs together, so this is fresh research, not a bridge-seed duplicate. A closely adjacent (but not identical) sibling exists — see 'Sibling capture' note below."
derived_from: []
tags: ["resource-rationality","rational-analysis","oaksford-chater","lieder-griffiths","wason-selection-task","cognitive-science","citation-network","bridge-investigation","bridge-seed-check"]
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_tier: 1
seek_code_commit: "d40ee39"
---


**Sibling capture (read first, not a duplicate of this pair).**
`10-inbox/raw/2026-10-22-dup-risk-resource-rational-analysis-anderson-marr-oaksford-chater-bridge.md`
investigated a cosine-0.841 pairing that shares one end with this topic's pair —
[[claim-resource-rational-analysis-rationality-revisited-swaps-normative-yardstick-same-move-as-oaksford-chater]]
— but paired it against a *different* note
([[claim-resource-rational-analysis-extends-andersons-rational-analysis]], the
Anderson/Marr-algorithmic-level self-description), not against
[[claim-lieder-griffiths-2020-name-chater-oaksford-1999-as-rational-analysis-source]].
That capture found its bridge via a §3 sentence connecting Marr's
algorithmic level to "a realistic normative standard." This capture's pair is
genuinely different and unaddressed by that sibling: it asks whether the
specific *citation* Lieder & Griffiths reach for when naming "rational
analysis" (Chater & Oaksford 1999) has anything to do with the *rhetorical
move* their own §5.4 performs later in the same article. No existing capture
asks that question, so this is fresh research, not a re-tread.

**Verdict: the bridge is real, and tighter than "same source document."**
Chater & Oaksford's 1999 retrospective is not a neutral terminological
citation. Read directly (fetched fresh this session via `extract_pdf`, same
sha256 as the one already on file for
[[claim-chater-oaksford-1999-list-marr-integration-as-outstanding-question]]),
it turns out to contain, in its own worked treatment of the Wason selection
task, a demonstration of the identical "swap the normative yardstick, apparent
irrationality dissolves" move that the vault already catalogues as Oaksford &
Chater's signature and that resource-rational analysis's §5.4 performs twenty-one
years later. The paper Lieder & Griffiths cite for the paradigm's *name* is,
independently of that citation choice, also a worked exemplar of the *move* the
other seed note is about. That is the genuine connective tissue — not
coincidental resemblance from two notes drawing on the same PDF (they don't:
this capture's new quotes come from a different document than either seed
note's primary), but two facts about the same 1999 paper converging on one
article.

---

## Claim: Chater & Oaksford's 1999 retrospective — the paper Lieder & Griffiths (2020) cite for the name "rational analysis" — itself performs the "swap the normative yardstick, apparent irrationality dissolves" move, via its own worked treatment of the Wason selection task

**Claim type:** technical-mechanism / historical (what a primary source's own
argument structure is) — Tier 1-2 required and met via direct primary read.

Reviewing the Wason selection task, Chater & Oaksford first state the classical
verdict: on the logical reading, "only 4% of participants make this response"
expected by formal logic, and "the fact that people do not follow this pattern
has been taken as casting doubt on human rationality." They then present their
own rational analysis as a replacement standard, not a patch on the logical
one: "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, in their own words: "the ODS model accounts for
the large volume of apparently puzzling data on Wason's selection task. 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 "irrationality" dissolves under the new
standard — the same shape [[claim-resource-rational-analysis-rationality-revisited-swaps-normative-yardstick-same-move-as-oaksford-chater]]
later finds in Lieder & Griffiths's own §5.4.

**Provenance:**
- source_url: https://www.dectech.co.uk/publications/LinksNick/FoundationsTheoryAndMethodology/ten%20years%20of%20the%20rational%20analysis%20of%20cognition.pdf
- source_sha: 6ed2746249f4fdcc7fe530c848b3ac698c12c368805eb521f7035b8e7167fbd8 (fetched via `extract_pdf` this session; tls: verified; identical sha to the vault's existing fetch of the same PDF behind [[claim-chater-oaksford-1999-list-marr-integration-as-outstanding-question]])
- 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_tier: 1
- source_quote: "The fact that people do not follow this pattern has been taken as casting doubt on human rationality" (quote_check-grounded, single contiguous passage)
- source_quote_2: "It resolves the problem of apparent human irrationality, by demonstrating that human performance approximates Bayesian optimal data selection" (quote_check-grounded, single contiguous passage, reading-order match across a line wrap)
- source_quote_3: "Oaksford and Chater's" (quote_check-grounded fragment)
- source_quote_4: "rational analysis draws on the theory of optimal data selection (ODS) from Bayesian sta-" (quote_check-grounded fragment)
- source_quote_5: "tistics47, rather than on Popper's falsificationism." (quote_check-grounded fragment)
- Note on quotes 3-5: the full sentence reads, per direct read of the extraction at this location, "Oaksford and Chater's[7] rational analysis draws on the theory of optimal data selection (ODS) from Bayesian statistics[47], rather than on Popper's falsificationism." It fails automated grounding as one contiguous string because a superscript footnote numeral ("Chater's7") is glued to the preceding word with no space in the pdftotext extraction — a variant of the two-column/footnote extraction seam already logged in `00-meta/seek-flags.md`, not a new defect, but one the 2026-10-01 "fixed" ruling did not fully close. Recorded as three independently grounded fragments per the quote-shape rule's provision for this kind of extraction seam, rather than as one over-assembled quote.

---

## Claim: the connection between the two seed notes is specific to this 1999 paper, not a general fact about Oaksford & Chater's oeuvre

**Claim type:** structural/historical — Tier 3-4 acceptable for this kind of
claim; grounded here at Tier 1 since it rests on a direct primary comparison.

The vault's existing "swap the yardstick" claim for Oaksford & Chater
([[claim-oaksford-chater-redefine-rationality-as-reasoning-about-uncertainty]],
[[claim-oaksford-chater-recast-wason-task-as-optimal-data-selection]]) is
sourced from a 2009 *Behavioral and Brain Sciences* précis of their 2007 book
*Bayesian Rationality*, not from the 1999 *Trends in Cognitive Sciences*
retrospective. This capture's finding establishes that the 1999 paper — the
one actually cited by [[entity-falk-lieder|Lieder]] & [[entity-thomas-griffiths|Griffiths]]
for the paradigm's name — independently carries the same argument in
miniature, eight years before the book and its précis. So the bridge does not
depend on treating "Oaksford & Chater" as an undifferentiated body of work
that happens to touch both seed notes; the specific paper footnoted for the
name is itself one of the places the move is made. This sharpens
[[observation-oaksford-chater-wason-resource-rational-pairing-is-confirmed-bridge-via-citation]],
which already established that Lieder & Griffiths cite both the 1994
Wason-task paper and the 1999 retrospective by name on facing pages — this
capture adds that the 1999 citation is not inert background: its own content
is a second, independent demonstration of the move the other citation (1994)
performs mathematically.

**Provenance:** same primary as above, plus the already-vault-verified facts
in [[claim-lieder-griffiths-2020-name-chater-oaksford-1999-as-rational-analysis-source]]
(Tier 1, verified-verbatim) and
[[claim-oaksford-chater-redefine-rationality-as-reasoning-about-uncertainty]]
(Tier 1, verified-verbatim), neither re-quoted here to avoid duplication.

---

## Claim: the bridge is real but sits one notch below full novelty — the vault's MOC already catalogues both halves of the "yardstick was wrong" family, and this capture tightens rather than discovers the link

**Claim type:** structural claim about the state of the vault's own
cross-referencing.

[[moc-the-yardstick-was-wrong-not-the-reasoner]] already lists "Oaksford &
Chater — Bayesian probability for logic" and "Resource-rational analysis — all
three classical norms at once" as two of its four members, and
[[observation-oaksford-chater-uncertainty-redefinition-resource-rational-pairing-is-genealogical-bridge]]
already names [[claim-resource-rational-analysis-rationality-revisited-swaps-normative-yardstick-same-move-as-oaksford-chater]]
as evidence for a "hedged" genealogical bridge. A `vault_novelty` check on this
capture's own Claim 1 returned `max_cosine: 0.82` (duplicate-risk band,
95th percentile), closest neighbours being Note 2 itself (0.82), the Wason
recast claim (0.816), and the MOC (0.799). The specific quotes recorded above
are new to the vault — grepped against `30-notes/` and absent — but the
*conceptual* territory (Oaksford & Chater perform a yardstick-swap on the
Wason task) is already well-mapped. This capture's genuine addition is narrow:
not a new member of the family, but the fact that the 1999 naming-citation and
the 1994 worked-example citation are not two separate links to two separate
Oaksford-Chater ideas — one paper (1999) demonstrates the same move the other
(1994) formalizes, and Lieder & Griffiths cite both, on facing pages, in the
same article that later performs its own version of the move in §5.4.

---

## Further leads

- Chater & Oaksford (1999) footnote 60, "Chater, N. and Oaksford, M. The
  probability heuristics model of syllogistic reasoning" (in press at time of
  writing) — a third domain (syllogistic reasoning, not just the selection
  task) where the same logic-to-probability swap is applied; not opened this
  session.
- The 1999 paper's own "Outstanding questions" box lists "What are the limits
  of rational analysis? Can every successful aspect of cognition be explained
  rationally, or could some mechanisms 'just work' with no rational
  explanation?" — a question adjacent to but distinct from the Marr-integration
  outstanding question already captured in
  [[claim-chater-oaksford-1999-list-marr-integration-as-outstanding-question]];
  not pursued here.
- Whether the promotion pass should fold this capture's Claim 1 into
  [[claim-chater-oaksford-1999-list-marr-integration-as-outstanding-question]]
  as an additional fact about the same source document, rather than minting a
  new claim-note, given the duplicate-risk signal on the conceptual territory —
  flagged as a disposition question for promotion rather than decided here.

## Safety flags

None. The only source fetched this session was
https://www.dectech.co.uk/publications/LinksNick/FoundationsTheoryAndMethodology/ten%20years%20of%20the%20rational%20analysis%20of%20cognition.pdf
via `extract_pdf` (tls: verified, no elevated-suspicion transport) — a
standard academic PDF (a *Trends in Cognitive Sciences* review article hosted
on a co-author's own publications mirror). No addressed-to-AI language,
override language, claimed authority, tier self-assignment, file-system
instructions, credential requests, or urgency framing was encountered
anywhere in the extracted text (the rational-analysis review, its boxes and
reference list, or the adjacent unrelated articles bundled in the same PDF
scan — an amygdala-circuitry review and a prefrontal-cortex-development
review that happened to share the same journal-issue PDF).

## Entity candidates

- **John R. Anderson** — person — the foundational, older figure both the
  "rational analysis" name and the 1999 retrospective's whole argument are
  built on and measured against (Chater & Oaksford thank him by name in their
  acknowledgements and open the piece crediting Anderson & Milson 1989 and
  Anderson 1990 as rational analysis's origin). Already has
  [[entity-john-anderson]]; flagged first per the ancestry-priority
  convention, not last.
- Nick Chater, Mike Oaksford — already have [[entity-nick-chater]] and
  [[entity-mike-oaksford]]; this capture adds that their own 1999 retrospective
  (not just their later book/précis) independently demonstrates the
  yardstick-swap move, a fact their pages may not yet carry.
- Falk Lieder, Thomas Griffiths — already have [[entity-falk-lieder]] and
  [[entity-thomas-griffiths]]; confirmation, not a new flag.
- Peter Wason — already has an entity page; the selection task is the shared
  object both the 1999 retrospective and the vault's existing Wason-recast
  claim-note run through.