Asymmetric-Amplitude Error Correction¶

A correction issued through the same channel as the original assertion but at systematically lower amplitude, so it reaches far fewer receivers and population belief stays stuck on the now-incorrect original even though a correction is formally on the record.

Core Idea¶

Asymmetric-amplitude error correction is the structural failure mode in publisher-mediated communication in which a correction is issued through the same channel as the original assertion but at a systematically lower amplitude — reaching fewer receivers, with less prominence, with less push — so that the population-level effect of the correction is negligible even when it has been formally issued.

The mechanism runs as follows. A publisher controls an amplitude-variable channel whose parameters (headline placement, broadcast slot, push-notification dispatch, search-ranking, recommendation-graph routing, social distribution) determine how many receivers a message reaches and with what salience. The original assertion is emitted at high amplitude: front page, lead segment, full following-push, prominent search position. The correction is emitted through the same channel at low amplitude: a corrections column buried inside the print edition, an editor's note appended to an archived web article without re-push, a walk-back in a follow-up segment with a fraction of the original's audience, a quiet re-ranking that never surfaces the correction to the original-snippet's query population. The asymmetry is not produced by passive forgetting over time; it is built in at issuance: even at the moment the correction is published, far fewer receivers are reached than were reached by the original. The publisher's records show "correction issued"; the receiver population's beliefs have not moved, because most receivers were never reached.

The population-level consequence is a belief ratchet: each original-correction cycle with amplitude asymmetry adds to a stock of audiences holding the now-incorrect original belief, and that stock is not unwound by a formally complete but low-amplitude correction. The diagnostic is not "was a correction issued" but "what fraction of original-receivers were reached by the correction at comparable amplitude." The structural fix is issuance-amplitude symmetry: corrections to front-page stories run on the front page; corrections to push-notified content trigger a push to the same subscriber list; corrections to socially distributed content are re-distributed through the original sharing graph; corrections to search-snippet content are surfaced to the same query population. The pattern is distinct from the cognitive continued-influence effect (in which a receiver persists in a belief even after seeing a correction) — that is downstream; asymmetric-amplitude error correction names the upstream channel condition that determines how many receivers see the correction in the first place.

Structural Signature¶

Sig role-phrases:

the single publisher — one source with exclusive control over how prominently each message goes out
the amplitude-variable channel — a distribution surface whose parameters (placement, push, broadcast slot, ranking, sharing-graph routing) set how many receivers a message reaches and how saliently
the high-amplitude original — the initial assertion emitted at full prominence and reach
the low-amplitude correction — the walk-back emitted through the same channel but with systematically fewer parameters firing
the issuance-time asymmetry — the reach shortfall built in at the moment of publication, not accrued by later forgetting
the under-reached receiver population — far fewer audience members exposed to the correction than were exposed to the original
the belief ratchet — population belief stuck on the now-incorrect original, accumulating a fresh uncorrected cohort with each cycle
the record–belief divergence — the publisher's log reads "corrected" while audience belief reads "uncorrected," the fingerprint of the shortfall

What It Is Not¶

Not the absence of a correction. The whole point is that a correction was issued — formally, honestly, often promptly — and still failed at the population level. The defect is not editorial neglect or a refusal to set the record straight; it is that the correction went out at a fraction of the original's amplitude, so most of the original's audience was never reached. A desk that "ran a correction" can be squarely inside this failure mode.
Not passive decay over time. The shortfall is built in at issuance: even at the moment the correction is published, fewer receivers are reached and reached more faintly. It is not that a correction broadcast at full amplitude slowly fades from memory — the gap is present at t=0 and would show in an audit run the instant the correction goes out. Temporal forgetting is a separate process layered on top.
Not the continued-influence effect. That cognitive effect is downstream: it describes receivers who persist in a belief even after they see the correction. This pattern is upstream — its receivers were never reached at all. The two are distinguished by a single test: estimate the fraction of original-receivers the correction reached at comparable prominence. A fraction near zero is this pattern; a fraction near one, with belief still stuck, hands the post-mortem to continued influence.
Not information asymmetry. Information asymmetry is differential knowledge between parties; this is differential reach of two related messages from one source. No party here holds private information — the correction is public and on the record. What differs is how many receivers each message's issuance-amplitude actually exposed.
Not necessarily bad faith or concealment. The asymmetry is a structural property of issuance discipline — corrections column versus front page, no-push versus push — and arises from default channel routing even with no intent to bury. Deliberate suppression is one way to produce the gap, but the ordinary case is a desk following standard correction practice that simply runs fainter than the original by default.

Scope of Application¶

Asymmetric-amplitude error correction lives wherever a single publisher controls an amplitude-variable channel and an audience updates on what reaches it rather than what is formally on the record — that is, across the publisher–channel–receiver subfields of communication and media; its reach is bounded to that substrate (the looser "they buried the retraction" analogues are carried by the parent primes signal_amplification and ratchet, not by this concept).

Journalism / newsroom corrections practice — the home turf: a front-page or lead-segment original answered by a buried corrections-column line, an un-pushed editor's note, or a walk-back segment drawing a fraction of the original audience.
Scientific publishing — an indexed, cited, pushed original article versus a retraction notice published with no comparable push, often poorly linked back and not retro-cited, so the flawed result keeps circulating at full amplitude.
Public-health communication — prominent initial guidance broadcast widely, then a quieter walk-back issued when the science shifts, leaving the superseded advice the higher-amplitude message.
Legal-record publication — a publicized conviction or trial judgment versus an appellate reversal or expungement with a materially smaller publication footprint.
Audit and accountability reporting — a press-released headline finding answered by a later substantive amendment buried in an appendix with no comparable distribution push.
Software changelogs and release notes — marketing-grade launch announcements versus regression acknowledgments and bug-fix notes tucked into low-traffic changelog entries.
Algorithmic / recommender distribution — a viral item reaching audience N, with a moderation label or correction the recommender never re-routes through the original distribution graph, so it reaches only a fraction of N.
Search-result amendment — an original snippet preserved in cache and index, corrected by a quiet re-ranking rather than by surfacing the correction to the original snippet's query population.

Clarity¶

Naming this pattern relocates the standard a correction must meet. The everyday editorial question — "did we run a correction?" — is a record-keeping question, answerable yes by appending an editor's note or printing a corrections-column line. Asymmetric-amplitude error correction makes legible that the record-keeping standard and the population-belief standard come apart: a correction can be fully, honestly, promptly issued and still leave the audience's beliefs untouched, not because readers refused to update but because most of them were never reached. That dissolves a confusion that otherwise invites the wrong post-mortem — blaming reader stubbornness, "the damage is done," or the continued-influence effect — when the failure was upstream, in how faintly the correction went out. The sharper question a desk can now ask is not whether a correction exists but what fraction of the original story's audience the correction reached at comparable prominence.

The concept also sharpens a distinction editorial practice routinely blurs: issuance discipline versus retraction policy. A retraction policy governs whether and how the record gets set straight; issuance discipline governs the placement, push, broadcast slot, and distribution-graph routing through which the corrected record actually travels. Holding these apart tells an editor that strengthening the wording or the verification behind a correction cannot move public belief if the correction still runs at a fraction of the original's amplitude — and that the corrective lever is matched issuance: a front-page error corrected on the front page, a push-notified claim corrected with a push to the same list, socially distributed content re-distributed through its original sharing graph. The amplitude gap, not the act of correcting, becomes the thing to instrument and close.

Manages Complexity¶

The sprawl this tames is the catalogue of ways a published error fails to get unwound — the buried corrections column, the editor's note appended without a re-push, the walk-back segment that draws a sliver of the original's audience, the quiet search re-ranking that never reaches the query population, the moderation label served to a fraction of the viral content's reach. Each looks like its own editorial pathology with its own remedy (write the correction more forcefully, verify it harder, run it sooner, word the retraction more plainly), and a desk that treats them separately is forever debugging one channel at a time. The concept collapses that whole list to a single comparison on one axis: the issuance amplitude of the original against the issuance amplitude of the correction on the same publisher-controlled channel. Reach is a monotone function of amplitude, so the population-level outcome reads straight off the gap — where the two amplitudes are matched, audience belief tracks the corrected record; where the correction runs fainter, belief ratchets toward the now-incorrect original by roughly the shortfall in receivers reached, no matter how complete the formal retraction. The analyst stops tracking the content, prominence, timing, and tone of each correction as separate variables and tracks one quantity per error: the fraction of the original story's audience the correction reached at comparable prominence. That single scalar also sorts every intervention. Moves that change wording, verification, or speed leave the amplitude gap untouched and so cannot move public belief; the only corrective lever is matched issuance — front-page error on the front page, push-notified claim re-pushed to the same subscriber list, socially distributed content re-distributed through its original sharing graph, search snippet re-surfaced to the same query population. A high-dimensional "why don't our corrections land" problem becomes a one-parameter diagnostic with a branch structure (matched → belief corrected; under-amplified → belief stuck) and a fix that follows directly from which branch the error sits on.

Abstract Reasoning¶

The concept licenses a tight family of inferences, all keyed to the single amplitude-gap scalar. Diagnostic: when an editor observes that a subject's reputation has not recovered, that polls show the original belief still dominant, or that the cited rationale in ongoing fallout (donor flight, cancelled contracts, persisting accusations) is the now-incorrect original story despite a correction being on the record, the move is to infer that the correction under-reached — to reason from un-moved population belief back to an issuance-amplitude shortfall, rather than to the receiver's stubbornness. The signature to look for is the divergence between two records: the publisher's log says "corrected," the audience's beliefs say "uncorrected." That divergence is the fingerprint of low-amplitude issuance, and it is distinguishable from its downstream cousin because the audience here was never reached, not reached-and-unpersuaded. The test that settles the diagnosis is to estimate the fraction of original-receivers the correction actually reached at comparable prominence; a fraction near zero confirms the amplitude branch, a fraction near one redirects the post-mortem to continued-influence territory.

Interventionist: the corrective lever is matched issuance, and each fix carries a sharp prediction. Re-pushing a corrected push-notified claim to the same subscriber list, re-running a front-page error's correction on the front page, re-distributing socially-shared content's correction through its original sharing graph, or re-surfacing a corrected search snippet to the same query population — each predicts that population belief should move toward the corrected record in proportion to the receivers newly reached, and how far it moves measures how much of the original failure was amplitude rather than persistence. The contrapuntal prediction is just as load-bearing and is the concept's main warning: interventions that strengthen the correction's wording, verification, or speed while leaving the amplitude gap untouched should move public belief essentially not at all — because they operate on a channel condition that was never the binding constraint. An editor who keeps rewording a correction that still runs at a fraction of the original's reach is predicted to keep failing, and the concept says exactly why.

Boundary-drawing: the pattern applies only where a single publisher controls a genuinely amplitude-variable channel and the audience updates on what reaches it rather than what is formally on the record; remove the single controllable channel (peer-to-peer rumor, distributed price discovery, face-to-face talk) and there is no amplitude to match, so the diagnosis does not even pose. Within its regime, it draws a second internal boundary: it is upstream of the continued-influence effect and must be cleared first. If the correction reached comparable amplitude and belief still did not move, the explanation has crossed out of this concept and into the cognitive one — so the order of investigation is fixed, channel-reach before receiver-persistence.

Predictive / order-of-events: because reach is monotone in amplitude and the shortfall is built in at issuance rather than accruing over time, the model predicts the belief gap is present at the correction's moment of publication, not something that only emerges as memory fades — a correction audited at t=0 should already show the deficit. Across repeated cycles it predicts a ratchet: each under-amplified original-correction pair adds a fresh cohort of audiences holding now-incorrect beliefs, a stock that accumulates and that the publisher's record-keeping will systematically fail to reflect, so the gap between formal-record accuracy and population belief widens monotonically with the number of uncorrected-in-practice cycles.

Knowledge Transfer¶

Within its home domain — the family of publisher-channel-receiver substrates where one source controls a genuinely amplitude-variable channel — the concept transfers as mechanism, intact: the diagnostic (compare original-issuance amplitude against correction-issuance amplitude on the same channel; estimate the fraction of original-receivers the correction reached at comparable prominence), the intervention (matched issuance — front-page error corrected on the front page, push-notified claim re-pushed to the same subscriber list, socially distributed content re-distributed through its original sharing graph, search snippet re-surfaced to the same query population), and the vocabulary (issuance discipline vs. retraction policy, amplitude gap, belief ratchet) all carry without translation. An analyst who has internalized the newsroom corrections-column case recognizes the same structure with the same fix in scientific publishing (an indexed, cited, pushed original versus a retraction notice published with no comparable push and often poorly linked back), in public-health communication (prominent initial guidance, quieter walk-back when the science shifts), in legal-record publication (a publicized conviction, an appellate reversal with a fraction of the original footprint), in audit and accountability reporting (a press-released headline finding, a later substantive amendment buried in an appendix), in software changelogs (marketing-grade launch notes, low-traffic regression acknowledgments), and in algorithmic distribution (a viral item reaching audience N, a moderation label or correction the recommender never re-routes through the original distribution graph). Across all of these the channel, the players, and the currency of "amplitude" change, but the binding constraint, the divergence between the publisher's "corrected" log and the audience's un-moved belief, and the corrective lever do not — this is one mechanism wearing many institutional costumes, not a set of look-alikes.

Beyond that family the transfer is best characterized carefully, because two different things are going on. First, where there is no single publisher controlling a single amplitude-variable channel — peer-to-peer rumor, informal social talk, markets where distributed price discovery substitutes for explicit correction — the pattern does not merely weaken; it fails to pose, because the amplitude to be matched does not exist. There is nothing for the diagnostic to measure and no issuance lever to pull. So the honest statement is not "it transfers as analogy here" but "the precondition is absent, so the concept does not apply." Second, when people do reach for the idea outside the publisher substrate — saying an organization "buried" a retraction, or that bad news got "louder" than the fix — they are borrowing the shape (a loud emission, a quiet correction) while dropping the named machinery (a controllable channel whose amplitude parameters set reach); that is analogy, and it should be marked as such.

But the deeper and more useful point is the third case: the cross-domain lesson that genuinely recurs is not "asymmetric-amplitude error correction" — it is the more general patterns this concept instantiates. Wherever a correcting signal lands more faintly than the error it corrects and a population's state ratchets toward the stale value, what is actually recurring is signal_amplification run asymmetrically on two related messages, the competition for bounded receiver attention (attention_economics, salience), and a ratchet-style accumulation that formal record-keeping fails to reflect. Those parents travel across substrates on their own terms; the cross-domain reasoner should carry them, not the named concept. What stays home-bound is precisely the cargo that makes this a domain-specific abstraction rather than a prime: the requirement of a single publisher, an amplitude notion that applies symmetrically to original and correction, an audience that updates on what reaches it rather than what is formally on the record, and the institutional asymmetry that systematically under-amplifies corrections. Strip those and only the parent mechanisms remain — which is exactly the boundary between this concept (load-bearing within journalism, scientific publishing, public-health communication, and audit) and the primes it composes (see Structural Core vs. Domain Accent).

One internal boundary travels with the concept and is worth restating as part of its transfer: it is upstream of the cognitive continued-influence effect and must be cleared first. In any substrate where it applies, the order of investigation is fixed — channel-reach before receiver-persistence. If the correction reached comparable amplitude and belief still did not move, the explanation has crossed out of this concept and into the cognitive one. That ordering is not a metaphor that gets renamed across domains; it is a structural fact about where this failure mode sits in the causal chain, and it holds wherever the concept's preconditions hold.

Examples¶

Canonical¶

The defining instance is the newspaper front-page error answered by a buried corrections column. A daily runs a damaging story — say, naming a local official as the target of a fraud investigation — as its lead, above the fold on page one, where it reaches the paper's full readership, is picked up by wire services, and dominates the day's coverage. Weeks later the paper learns the official was never a target; it issues a correction, honestly and promptly, as a two-line entry in the standard corrections column on page A2 or deep in the section, with no front-page flag and no re-push. The correction is formally complete and on the record, yet only a small fraction of those who saw the accusation ever encounter the retraction. The official's reputation does not recover, because the population's belief was fixed by the high-amplitude original and never reached by the faint correction.

Mapped back: The paper is the single publisher operating the amplitude-variable channel whose placement parameters set reach. The page-one accusation is the high-amplitude original; the corrections-column line is the low-amplitude correction, its shortfall built in at publication — the issuance-time asymmetry leaving the under-reached receiver population. The paper's "corrected" log against the public's unchanged belief is the record–belief divergence.

Applied / In Practice¶

Scientific publishing shows the pattern with measurable, harmful consequences. Andrew Wakefield's 1998 Lancet paper linking the MMR vaccine to autism was published in a prominent journal, amplified worldwide by press coverage, and drove a lasting collapse in vaccination confidence. The Lancet fully retracted it in 2010, but the retraction notice — an indexed correction with none of the original's press amplification or citation footprint — reached a tiny fraction of those who had absorbed the original claim. Studies of retracted literature more broadly find that retracted papers keep accruing citations for years afterward, because the retraction is issued far more faintly than the paper it withdraws. The record shows "retracted"; population and even scholarly belief stay partly anchored to the discredited original.

Mapped back: The journal is the single publisher; the indexed, cited, press-amplified paper is the high-amplitude original, and the retraction notice is the low-amplitude correction. Continued citation and persisting vaccine fear are the belief ratchet — a cohort left holding the now-incorrect claim. The gap between the "retracted" record and unmoved belief is the record–belief divergence, and the fix would be amplitude-matched: retract with the reach the original enjoyed.

Structural Tensions¶

T1: Record-keeping standard versus population-belief standard (the log that documents good faith hides the failure). A correction can be fully, honestly, and promptly issued — satisfying every record-keeping and ethics requirement — and still leave audience belief untouched, because most receivers were never reached. The two standards come apart, and the record-keeping one is the trap: it is easy to meet (append an editor's note, print a corrections line), it is the standard editors are actually held to, and passing it confers false assurance that the matter is closed. The tension is that the very artifact documenting the publisher's diligence — the "corrected" log — is what obscures that nothing moved at the population level, so the more scrupulously a desk satisfies the record standard, the more confidently it can miss the belief standard entirely. Diagnostic: Is success being judged by whether a correction is on the record, or by what fraction of the original's audience the correction reached at comparable prominence?

T2: Channel-reach versus receiver-persistence (the same symptom, opposite remedies). Un-moved belief despite a correction on record has two entirely different causes: the audience was never reached (this pattern, upstream) or was reached and unpersuaded (the continued-influence effect, downstream). They present identically — reputation not recovering, the stale belief still dominant — yet demand opposite interventions: amplify the issuance versus address cognitive persistence. The concept fixes the order of investigation, channel-reach before receiver-persistence, precisely because misreading under-reach as reader stubbornness sends a desk into endless rewording of a correction that was never seen. The tension is that the correct remedy hinges on a distinction invisible in the symptom, resolvable only by estimating the fraction of original-receivers the correction actually reached — a fraction near zero indicts amplitude, near one hands the case to continued influence. Diagnostic: What fraction of the original's audience did the correction reach at comparable amplitude — near zero (fix the channel) or near one with belief still stuck (a cognitive-persistence problem)?

T3: Matched issuance versus attention cost and re-amplification risk (the clean fix has a price). The prescribed lever is amplitude symmetry: correct a front-page error on the front page, re-push to the same subscriber list, re-distribute through the original sharing graph. But full matching is not free. Re-issuing a correction at the original's amplitude re-exposes the whole population to the very claim being corrected, raising the error's familiarity even as it broadcasts the fix; it consumes scarce prominence (front pages, push slots, attention budget) that competes with current reporting; and applied indiscriminately it would over-amplify trivial corrections. The tension is that the structurally correct fix collides with attention scarcity and with the hazard of amplifying the falsehood by the same act meant to bury it, so "match the amplitude" is right in principle yet must be rationed and shaped against costs the clean diagnosis abstracts away. Diagnostic: Does re-issuing this correction at the original's amplitude buy enough belief-movement to justify the attention it consumes and the risk of re-exposing the audience to the error?

T4: Structural default versus deliberate burying (the same signature, different accountability). The concept is intent-neutral by design: the amplitude gap arises from ordinary default routing — corrections column versus front page, no-push versus push — even with no wish to conceal, which is analytically clean and avoids moralizing every faint correction. But deliberate suppression produces the identical signature, an under-amplified correction and an unmoved audience, so the diagnostic that detects the gap cannot by itself distinguish an innocent default from intentional burying. The tension is that the very intent-neutrality that makes the pattern a general structural diagnosis also makes it blind to the accountability question that often matters most — whether the shortfall was a routing default or a choice — and the two are told apart only by evidence outside the amplitude comparison. Diagnostic: Does the amplitude shortfall trace to standard default correction routing, or to a choice to withhold the prominence the original received — and what evidence beyond the gap itself distinguishes them?

T5: Autonomy versus reduction (publisher-substrate concept or the composition of amplification, attention, and ratchet). Asymmetric-amplitude error correction is a load-bearing domain concept wherever a single publisher controls an amplitude-variable channel and an audience updates on what reaches it — across journalism, scientific publishing, public-health, legal records, and algorithmic distribution it transfers as mechanism with the amplitude-gap diagnostic intact. But its portable cross-domain content is the parents it composes: signal_amplification run asymmetrically on two related messages, competition for bounded attention (attention_economics, salience), and a ratchet accumulation that formal record-keeping fails to reflect. Notably, where the precondition is absent — peer-to-peer rumor, distributed price discovery, face-to-face talk — the concept does not merely weaken into analogy; it fails to pose, because there is no single controllable amplitude to match. The tension is that a named, institutionally specific failure mode sits atop substrate-neutral primes, with a hard precondition (single publisher, matchable amplitude, reach-based updating) that both defines its home and marks exactly where only the parents survive. Diagnostic: Resolve toward the parents (signal_amplification, attention_economics/salience, ratchet) when carrying the lesson to any substrate; and note that without a single publisher controlling a matchable-amplitude channel the named concept does not apply at all — reserve it for the publisher–channel–receiver case.

Structural–Framed Character¶

Asymmetric-amplitude error correction sits at mixed on the structural–framed spectrum — off both poles, balancing a real cross-substrate amplification-and-ratchet mechanism against a tight binding to the publisher-channel-receiver substrate and a failure-mode framing. The criteria pull in both directions. Evaluative_weight leans mildly framed: the concept is cast as a failure mode, a defect that leaves belief stuck on a now-incorrect original — a negative valence — but the entry deliberately holds it intent-neutral (T4: it "arises from default channel routing even with no intent to bury"), so the underlying phenomenon of two messages emitted at different amplitudes is value-neutral and the hazard charge is a stance the analyst brings. Human_practice_bound points framed and is unusually sharp: the pattern requires a single publisher controlling an amplitude-variable channel and an audience that updates on what reaches it, and where that precondition is absent (peer-to-peer rumor, distributed price discovery) the concept does not merely weaken but "fails to pose" — so it is constitutively bound to publisher-mediated human communication. Institutional_origin is mixed: the named failure mode and its issuance-discipline apparatus are editorial-practice furniture, yet the mechanism it instantiates (asymmetric amplification feeding a ratchet) is a real structural regularity, not one field's invention. Vocab_travels is bimodal: the amplitude gap, issuance discipline vs. retraction policy, belief ratchet, record–belief divergence idiom is pinned to the publisher substrate, while the parent mechanisms travel under their own names. Import_vs_recognize points structural within the substrate — the entry insists journalism, scientific publishing, public health, legal records, and algorithmic distribution are "one mechanism wearing many institutional costumes, not a set of look-alikes," so those are genuine mechanism-recognitions — while beyond the substrate the move is either "fails to pose" (no publisher) or loose analogy.

The portable structural skeleton is a single compound idea — a correcting signal emitted at lower amplitude than the error it corrects, under competition for bounded receiver attention, ratcheting a population's state toward the stale value in a way formal record-keeping fails to reflect — which no one prime carries alone; it is composed from signal_amplification (run asymmetrically on two related messages), attention_economics/salience (the bounded-attention competition), and ratchet (the accumulating uncorrected cohort). That cluster genuinely travels across substrates on its own terms, which gives the concept real structural substance and holds it at mixed. But it does not reach the structural side proper, because that cluster is exactly what the concept instantiates and composes, not what makes the named publisher-substrate failure mode travel: the cross-domain reach belongs to signal-amplification, attention-economics, and ratchet, while the distinctive cargo — the single-publisher requirement, the matchable-amplitude notion, the reach-based-updating audience, the issuance-discipline-versus-retraction-policy distinction, and the upstream-of-continued-influence ordering — is the media-practice furniture that stays home (and whose hard precondition marks exactly where only the parents survive). Its character: a real, cross-substrate-recurring asymmetric-amplification-and-ratchet mechanism bound to a single publisher's channel and framed as a correction failure mode — structural in the signal-amplification/attention/ratchet skeleton it composes from its parents, but wearing an issuance-discipline apparatus and a hard publisher precondition that do not leave media communication.

Structural Core vs. Domain Accent¶

This section decides why asymmetric-amplitude error correction is a domain-specific abstraction and not a prime, and it carries the case for its domain-specificity — there is no separate section for that.

What is skeletal (could lift toward a cross-domain prime). Strip the publisher furniture and a thin compound structure survives: a corrective signal emitted at systematically lower gain than the signal it corrects, under competition for a bounded receiver channel, so a population's state ratchets toward the stale value and never fully unwinds even though the corrective was formally emitted. The portable pieces are abstract — two related signals injected into the same channel at unequal gain, a monotone reach-as-a-function-of-amplitude relation, an accumulating stock that resets in only one direction, and a divergence between what the source has on record and what the population actually holds. This skeleton is genuinely substrate-portable, which is exactly why no single prime carries it alone and the entry composes it from three: signal_amplification run asymmetrically on two messages, the bounded-attention competition of attention_economics / salience, and the one-way accumulation of ratchet. But that composed cluster is the core the concept shares with any asymmetric-signal-plus-ratchet situation, not what makes it this named failure mode.

What is domain-bound. Almost everything distinctive is media-practice furniture that does not survive extraction. The pattern requires a single publisher controlling an amplitude-variable channel and an audience that updates on what reaches it rather than what is formally on the record — a hard precondition, not a soft tendency. Its worked vocabulary is home-bound: issuance discipline versus retraction policy, the amplitude gap, the record–belief divergence, the belief ratchet, matched-issuance levers (front page on the front page, re-push to the same subscriber list, re-distribute through the original sharing graph, re-surface to the query population). And its internal ordering — that it sits upstream of the cognitive continued-influence effect and must be cleared first — is a fact about where the failure sits in the publisher-mediated causal chain. The decisive test is unusually strict: remove the single controllable channel (peer-to-peer rumor, distributed price discovery, face-to-face talk) and the concept does not merely loosen into analogy — it fails to pose, because there is no matchable amplitude to compare and no issuance lever to pull. The named machinery is constituted by the very substrate the prime bar asks it to shed.

Why this does not clear the prime bar. A prime is a relational structure whose vocabulary travels and whose cross-domain transfer is recognition of the same mechanism, not analogy. This concept's transfer is bimodal. Within the publisher–channel–receiver family it travels intact — journalism, scientific publishing, public-health communication, legal-record publication, audit reporting, software changelogs, algorithmic distribution are "one mechanism wearing many institutional costumes," the amplitude-gap diagnostic and matched-issuance fix reading the same in each. Beyond it, two things happen, and neither clears the bar: where the single-publisher precondition is absent the concept fails to pose outright, and where people reach for it anyway ("they buried the retraction," "the bad news got louder than the fix") they borrow the loud-then-quiet shape while dropping the controllable-channel machinery — analogy, not mechanism. Crucially, when the bare cross-domain lesson is needed — a correcting signal landing fainter than the error, a population ratcheting toward a stale value — it is already carried, in more general form, by the parents the concept composes: signal_amplification, attention_economics / salience, and ratchet each travel across substrates on their own terms. The cross-domain reach belongs to those parents; asymmetric-amplitude error correction, as named, carries the single-publisher requirement, the matchable-amplitude notion, and the issuance-discipline apparatus as domain baggage that should stay home.

Relationships to Other Abstractions¶

Current abstraction Asymmetric-Amplitude Error Correction Domain-specific

Parents (1) — more general patterns this builds on

Asymmetric-Amplitude Error Correction is a decomposition of Emphasis Prime

Removing the publisher and correction vocabulary leaves two related signals receiving unequal gain in the same channel, with the original amplified relative to its repair.

Children (1) — more specific cases that build on this

Correction Burial Domain-specific is a kind of Asymmetric-Amplitude Error Correction

Correction Burial is the newsroom and mass-publishing species of Asymmetric-Amplitude Error Correction, fixing the channel to editorial placement and distribution.

Hierarchy paths (2) — routes to 2 parentless roots

Asymmetric-Amplitude Error Correction → Emphasis → Attention

Show alternative path (1)

Not to Be Confused With¶

Continued-influence effect. The cognitive cousin, and the one this concept is most often conflated with. It sits downstream: it describes a receiver who persists in a belief even after seeing the correction. This pattern is upstream — its receivers were never reached at all. The two present identically (reputation not recovering, the stale belief still dominant) but demand opposite fixes: amplify issuance versus address cognitive persistence. Tell: estimate the fraction of the original's audience the correction reached at comparable prominence — near zero indicts amplitude (this concept); near one with belief still stuck hands the case to continued influence.
Information asymmetry. A market/strategic condition of differential knowledge between parties, where one holds private information the other lacks. This concept involves no private information at all — the correction is fully public and on the record; what differs is the reach of two related messages from one source. Differential knowledge versus differential distribution. Tell: does one party secretly know something the other cannot (information asymmetry), or is the corrected fact public but under-broadcast so most receivers never encountered it (asymmetric-amplitude correction)?
Backfire / familiarity-backfire effect. The finding that re-stating a correction can strengthen the false belief by boosting the error's familiarity or triggering reactance. It is a hazard of the prescribed matched-issuance fix (T3), not the failure mode itself: this concept diagnoses corrections that never reached the audience, whereas backfire concerns corrections that reached them and rebounded. Tell: is the problem that the audience never saw the correction (this concept), or that seeing it entrenched the error further (backfire)?
Streisand effect. The near-inverse dynamic: an attempt to suppress information inadvertently amplifies its reach. Here the loud message (the original) already dominates and the quiet message (the correction) under-reaches — suppression, if any, runs against the correction, not toward accidental amplification. Tell: did an act of suppression backfire into wider spread (Streisand), or did a correction simply go out fainter than the error it addresses, leaving belief stuck (asymmetric-amplitude correction)?
The signal-amplification / attention-economics / ratchet cluster (the parents). The substrate-neutral patterns this concept composes — signal_amplification run asymmetrically on two related messages, competition for bounded attention (attention_economics / salience), and a one-way ratchet accumulation that formal record-keeping fails to reflect. These are not confusable peers but the umbrella that owns the cross-domain reach; where there is no single publisher controlling a matchable-amplitude channel, only these parents survive and the named concept "fails to pose." Tell: absent a single publisher and a matchable amplitude to compare, carry signal_amplification / attention_economics / ratchet, not this concept.

Neighborhood in Abstraction Space¶

Asymmetric-Amplitude Error Correction sits in a crowded region of the domain-specific corpus (20^th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.

Family — Media Systems & Agenda Control (24 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-07-12