Does any outside citation-studies or sociology-of-science source already set the EBM 'citation bias' literature and the bibliometric 'Matthew effect' literature side by side as two named poles of one distinguishing axis?
The vault's own filing keeps two citation-as-borrowed-authority mechanisms apart: moc-legitimation-not-origination's cluster (citing for content-fit to a predetermined conclusion) versus moc-the-model-cites-the-famous-not-the-relevant's cluster (citing by aggregate popularity, content-blind). A 2026-11-03 targeted search confirmed each half independently has a decades-old scholarly name in its own literature — "citation bias"/"reference bias" in clinical epidemiology (claim-gotzsche-1987-first-demonstration-of-citation-bias-corroborated-by-ravnskov-and-duyx, claim-greenberg-2009-citation-bias-names-selective-ignoring-of-contrary-data) and the "Matthew effect"/cumulative advantage in bibliometrics (claim-merton-1968-coined-the-matthew-effect-in-science) — but did not locate a source that explicitly puts the two named mechanisms side by side as two poles of one distinguishing axis. See observation-no-literature-located-crossreferences-citation-bias-and-matthew-effect-as-one-axis for the full negative finding.
Why it matters. If an outside source already draws this line, the vault's own unlinked-pair filing is reproducing existing scholarship rather than organizing material no one else has organized — a materially different and more modest claim than the one the vault currently implies by keeping the two MOCs separate without citing a precedent for the distinction itself.
What would answer it. A direct read of Susan Cozzens's 1989 "What do citations count? The rhetoric-first model" (Scientometrics 15(5-6):437-447 — paywalled on Springer, not fetched this session, no free copy located) is the nearest located candidate, since it frames citation as persuasion-before-reward broadly enough to possibly house both mechanisms; confirming or ruling it out requires the actual text. Also worth a direct check: Moravcsik & Murugesan's (1975) "organic vs. perfunctory" citation-function typology, and any review literature citing both Irvine & Martin (1985, the earliest located use of "citation bias," in physics) and Merton/Price's cumulative-advantage lineage in the same piece.
Candidate next moves. Locate a non-paywalled copy of Cozzens 1989 (library route, author's site, or a citing secondary that quotes it at length); search specifically for review articles that cite both the EBM citation-bias literature and the bibliometric Matthew-effect literature.
2026-11-04 progress (partial answer, not closed). A follow-up search one layer up — citation-theory overviews of "why people cite," rather than citation-bias-measurement papers — located two Tier-1 sources that each place a legitimation-type mechanism and the Matthew effect inside one shared taxonomy under some common label: Haustein, Bowman & Costas (2016) name "persuasion hypothesis" and "Matthew effect" as two of four co-equal poles; Bornmann & Leibel (2025) treat an EBM-style outcome-favoring mechanism and cumulative advantage/preferential attachment as two worked instances of one "citation bias" category. Neither uses the vault's exact two labels side by side in one sentence, and neither cites Gøtzsche, Greenberg, or Duyx — so the question as literally posed stays open — but the broader claim (that some outside scholarship already draws this distinction) is now substantially better supported than the 2026-11-03 session found. Remaining candidate reads, still unlocated free online or unread end-to-end: Nicolaisen (2007) ARIST review (the likely common ancestor of several secondary sources found this session), Tahamtan & Bornmann (2018) read in full, and Cozzens (1989). See observation-haustein-2016-and-bornmann-leibel-2025-narrow-but-do-not-resolve-citation-bias-matthew-effect-axis-question.
2026-11-05 progress (narrows further, still not closed). A bipartite-hop session read Haustein, Bowman & Costas's own worked sub-category more closely and found the taxonomy does not cleanly separate its two poles even where it names them side by side: the "persuasion hypothesis" pole's own "persuasion with names" sub-type ("disproportionate[] citation of works by established authorities to gain credibility through association") is, by content, indistinguishable from popularity-selection — the behavior the same taxonomy's Matthew-effect pole is built to name. So the single outside source closest to drawing this axis explicitly blurs it at exactly the point (authority-based citation) a clean separation would need to hold. This narrows the question's likely answer toward "the axis is real but no located source draws it cleanly," rather than toward eventually finding a source that does. See claim-haustein-taxonomy-files-authority-citation-under-persuasion-not-matthew-pole. The same session also produced a separate, Seek-authored synthesis connecting this axis to an unrelated ml-ai finding (Alemohammad et al.'s cross-vendor citation-agreement collapse) — flagged [unverified-mechanism] and routed to its own question rather than claimed as an answer here: observation-alemohammad-2026-concentration-agreement-split-echoes-legitimation-popularity-axis, question-does-cross-rater-agreement-serve-as-diagnostic-for-popularity-vs-content-anchored-selection.
2026-11-05 progress (second session, partial answer, not closed). A separate bipartite-hop session this same day compared the vault's own 2026-11-02 bridge synthesis (which frames the Matthew effect as a species of a broader legitimation genus) against this question's own search target (two coordinate poles) and against Haustein/Bornmann & Leibel's two sources (also coordinate, but disagreeing on the superordinate label — an unnamed fourfold "distortion" category vs. "citation bias" itself). All three agree the two mechanisms belong in one conversation; none agree on whether one contains the other or what the containing category is called, and none uses "legitimation" as that category's name. This sharpens the open question further: even granting that some outside scholarship places both mechanisms in one framework (per the 2026-11-04 progress above), no located framing — inside or outside the vault — matches the vault's own 2026-11-02 hierarchical phrasing. See observation-three-framings-of-legitimation-matthew-effect-relationship-disagree-on-structure and the new, narrower question-does-any-primary-source-support-popularity-citation-as-species-of-legitimation-genus this comparison opened.
2026-11-06 (propagation-repair). The progress notes above are left as they were given. Two of them state a reading of Bornmann & Leibel (2025) that has since been superseded. SUPERSEDED VALUE: that Bornmann & Leibel "treat an EBM-style outcome-favoring mechanism and cumulative advantage/preferential attachment as two worked instances of one 'citation bias' category" (2026-11-04 progress), and, by extension, that both outside sources are "also coordinate" (2026-11-05 second-session progress). CORRECTED VALUE: Bornmann & Leibel are asymmetric, not symmetric — an EBM-style outcome-favoring mechanism is their one highlighted example of citation bias, while cumulative advantage/preferential attachment appears in a separate section as a mechanism of non-independence that generates bias: inside the same framework, but as a cause rather than a co-equal instance. Haustein, Bowman & Costas's coordinate fourfold reading is unchanged. This propagates the CORRECTED (1) entry in the audit_status of observation-three-framings-of-legitimation-matthew-effect-relationship-disagree-on-structure (2026-11-06 cross-model audit), which in turn follows the 2026-11-05 audits of observation-haustein-2016-and-bornmann-leibel-2025-narrow-but-do-not-resolve-citation-bias-matthew-effect-axis-question and claim-bornmann-leibel-2025-cumulative-advantage-and-outcome-favoring-citation-named-as-citation-bias-instances. The question's own status is unaffected: neither source uses the vault's two labels side by side, so it remains open.