Preference Falsification¶
Core Idea¶
Preference falsification is the structural pattern in which an agent's publicly stated preference systematically diverges from their privately held preference under asymmetric cost on expression. The agent holds an actual ordering over states of the world (or a belief, opinion, or judgement) and faces a cost structure in which publicly voicing that ordering imposes social, political, professional, or material penalties that voicing a different, typically conformist, ordering does not. Under that cost asymmetry, the agent voices the cheaper preference while continuing to hold the costlier one privately. The structural fact is the gap between the held distribution and the visible distribution, sustained by the cost asymmetry.
Three commitments fix the pattern. First, an internal preference (or belief or judgement) that the agent has with some stability — it is not the act of public expression that constitutes the preference. Second, an expression cost asymmetry — voicing the private preference is more costly than voicing some alternative (silence, an opposing opinion, the orthodox line), where the cost can be reputational, professional, social, political, or material. Third, an expressed preference that differs systematically from the held one in the direction the cost asymmetry favours. The pattern is not ordinary lying, persuasion, or opinion change: the agent is not deceived and has not changed their mind, but faces a private problem of strategic expression and resolves it in favour of the cheaper public stance. The aggregate consequences are non-trivial: when many agents falsify in the same direction, the observed distribution diverges systematically from the held distribution, and that observed distribution becomes the input other agents use to estimate the held distribution — a self-referential loop in which misperception cascades downward, deepening falsification, until the resulting large gap becomes structurally fragile and a small change in the cost structure can trigger a sudden preference cascade.
How would you explain it like I'm…
Pretending to Like It
Hiding What You Think
The Silent Majority's Mask
Structural Signature¶
the privately held preference — the expression-cost asymmetry across voicings — the strategically expressed preference — the held-versus-expressed gap — the perception loop reading the visible distribution — the cost-structure-shock cascade fragility
A setting exhibits preference falsification when each of the following holds:
- A privately held preference. Each agent has a stable internal ordering, belief, or judgement that exists independently of any act of public expression — the agent is not deceived and has not changed their mind.
- An expression-cost asymmetry. Voicing the held preference carries higher cost — reputational, professional, social, political, or material — than voicing some alternative (silence, the orthodox line, the conformist position).
- A strategically expressed preference. Under that asymmetry the agent voices the cheaper preference while continuing to hold the costlier one, resolving a private problem of expression rather than lying or being persuaded.
- A held-versus-expressed gap. Across many co-directional falsifiers, the visible distribution diverges systematically from the held distribution in the direction the cost asymmetry favours. This gap is the load-bearing structural object.
- A perception loop. The visible distribution is the only estimator other agents have of the held distribution, and each agent's expression decision depends on that estimate — a self-referential loop in which misperception cascades and deepens the falsification.
- A cascade-fragility invariant. A large sustained gap is structurally fragile: a small change in the cost structure (a visible defection, a permission signal) can trigger a sudden preference cascade in which the held distribution surfaces rapidly while itself remaining unchanged.
The pattern is distinct from conformity (the held preference is preserved, not adopted), from information cascades (no cost asymmetry), and from costly signalling (it hides rather than conveys private information). Its variables — held distribution, cost structure, observed distribution, perception loop, cost-shock — make sudden public reorientations predictable and yield a channel-targeted intervention set (anonymity, aggregated revelation, pioneer dissent, cost-structure auditing).
What It Is Not¶
- Not
conformity. Conformity is adoption of the group view — the agent comes to actually hold the majority preference. Falsification preserves the held preference while distorting only the visible one; remove the expression cost and a different view surfaces. - Not an
information_cascade. An information cascade is sequential rational copying of others' visible actions under inference about their private signals, with no cost-asymmetry component. Falsification is driven by expression cost, not by Bayesian updating on others' choices. - Not
signaling. Costly signalling conveys private information through costly action; falsification is its inverse — it hides private information by avoiding the costly act of voicing the held view. - Not
preference_heterogeneity_and_conflict. That names genuinely differing held preferences in tension. Falsification can occur even under near-universal agreement in held preference, hidden behind uniform public conformity. - Not
spiral_of_silenceas a static state. The spiral of silence (and pluralistic ignorance) is the cognitive climate falsification produces; this prime is the expressive process under cost asymmetry that produces and sustains it. - Common misclassification. Reading sudden public reorientation as mass opinion change ("they were persuaded"). The held distribution may never have moved — only expression became safe; the diagnostic question is "what made expression safe?", not "what changed their minds?"
Broad Use¶
The held-versus-expressed gap under expression-cost asymmetry recurs across human-social and political substrates. In authoritarian and totalitarian polities, its original substrate, citizens publicly express loyalty while privately holding opposition, and the apparent stability masks a large gap; the 1989 collapses of Eastern European communist regimes were preference cascades in this exact pattern, the falsifying majority surfacing in days once defection became visible. In public opinion in democracies, spiral-of-silence dynamics and pre-election polling surprises (the shy-voter effects) reflect falsification under perceived social-desirability cost. The pattern recurs in religious orthodoxy (private doubt within enforced orthodoxies, released as rapid reorientation when permission arrives), in workplaces (silent dissent in meetings where junior staff face reputational cost, revealed in the post-meeting hallway conversation), in academic orthodoxy (junior researchers voicing the orthodox line until tenure or a paradigm shift makes the held view safe), in harassment underreporting (targets silencing themselves under expected retaliation, with the public cascade a textbook release of long-falsified preferences), in markets (analysts shading reports toward the dominant narrative, with "everyone secretly knew it was overvalued" recognition after a crash), in social-media content (users self-censoring under platform-norm pressure), in voting under perceived judgement (stated intent for socially-disfavoured candidates underrepresenting actual voting), and in pre-cascade conditions in social movements (civil-rights, suffrage, and gay-rights movements exhibiting long gaps followed by rapid reorientation). The cross-substrate fit is strong because the mechanism — asymmetric cost on expression — is generic to any system in which agents can voice a preference and face differential cost for different voicings, though all the substrates are human-social.
Clarity¶
Preference falsification clarifies by separating held preference from expressed preference and naming the gap between them as a structural object with its own dynamics. Ordinary discourse treats publicly visible opinion as a more-or-less reliable signal of held opinion; the prime makes the cost-driven substitution explicit, so the gap can be reasoned about, measured, and managed rather than assumed away. This converts a hidden assumption — that the visible distribution estimates the held one — into an inspectable claim that holds only when the expression cost asymmetry is small.
The clarifying force extends to a recurring class of explanation errors. When a regime collapses suddenly, when an organisation pivots overnight on a settled policy, or when a market reverses once the contrarian view becomes safe, the conventional explanation reaches for "people changed their minds." The prime points instead to a stable held distribution previously suppressed and now released by a cost-structure change, so the diagnostic question is not "what made them change?" but "what made expression safe?" The prime also distinguishes itself carefully from neighbours. It is distinct from conformity, which is adoption of the group view (the agent comes to actually hold the majority preference), whereas falsification preserves the held distribution but distorts the visible one. It is distinct from an information cascade, which is sequential rational copying of visible actions under inference about private signals, with no cost-asymmetry component. It is the inverse of costly signalling, which conveys private information through costly action, whereas falsification hides it by avoiding costly action. It is the structural divergence that self-censorship is one means of producing, and it is the expressive process that produces and sustains the static cognitive state of pluralistic ignorance. Holding these apart keeps the prime from being mistaken for genuine opinion change, rational copying, signalling, or a mere cognitive state.
Manages Complexity¶
The pattern compresses an unmanageable space of "why did the regime collapse, the seminar reverse, the boardroom unanimity turn out to be unanimous skepticism, the bubble pop, the movement break through" into a small set of structural pieces: the held distribution, the expression-cost structure, the observed distribution, the perception-of-distribution feedback, and the cost-structure shock. Once these are named, sudden public reorientations become tractable as predictable consequences of shocks to the cost structure rather than mysterious mass opinion changes. A bewildering variety of historical surprises reduces to one mechanism operating on a few variables.
The compression is operational because the intervention catalogue compresses in the same way. Wherever falsification is suspected, the same intervention family applies, each move targeting a specific variable. Anonymity — secret ballots, anonymous surveys, anonymous suggestion mechanisms, blind peer review — reduces the expression cost asymmetry directly. Aggregated revelation — poll sampling that conceals individual responses, list experiments that elicit sensitive preferences statistically — gives population-level signal without individual-level cost. Pioneer dissent — a small number of visible defectors — lowers the perceived isolation cost for everyone behind them, shifting the cost structure. And cost-structure auditing — asking what would happen to a dissenter before treating a unanimous public position as evidence of a unanimous held position — is the diagnostic move. Because each intervention is recognisable as a move on a named variable (the reputational channel, the professional channel, the perceived climate), the prime turns the open-ended problem of surfacing true opinion into a bounded set of channel-targeted interventions.
Abstract Reasoning¶
Preference falsification trains a reasoner to interrogate any apparent consensus through the cost structure on expression. The reasoner asks: whose held preference is at issue? What is the cost of voicing the held preference relative to alternatives? Who has the held preference but is voicing the cheaper alternative? And what would the observed distribution look like if the cost asymmetry were removed? Because these questions reference only the abstract roles — held preference, expression cost asymmetry, observed distribution, perception loop — they apply to a polity, a market, an organisation, or a seminar without translation, though the substrates remain human-social.
Several reusable moves follow. The cost-asymmetry diagnostic estimates the cost of voicing each available position; large asymmetries predict large held-versus-expressed gaps, so the reasoner characterises the cost structure before treating the expressed distribution as a measurement of the held one. The cascade-fragility prediction treats settings with large gaps as structurally unstable, so the reasoner reads apparent stability as evidence of either genuine consensus or fragile suppression, with the structural fragility itself diagnostic of which. The cost-channel decomposition treats expression cost as decomposable into reputational, professional, social, political, and material sources, so interventions can target individual channels. The held-distribution-inference move treats the expressed distribution as a biased estimator and triangulates across alternative access channels — secret ballots, list experiments, randomised response — rather than trusting expressed-distribution data. And the endogeneity move recognises that each agent's expression decision uses an estimate of others' held distribution whose only available estimator is the expressed distribution, which is downstream of every agent's decision, so the perceived climate can drift far from the held distribution and must be broken through external measurement. The same reasoning that tells a pollster to use a list experiment tells a manager to use an anonymous survey, because both are reasoning about a biased expressed distribution under a cost asymmetry.
Knowledge Transfer¶
The transferable content of the prime is a sharp set of diagnostic questions and a substrate-portable intervention catalogue. A practitioner who has internalised preference falsification looks at any group, polity, market, or organisation and asks: whose held preference is at issue, what is the cost of voicing it relative to alternatives, who is voicing the cheaper alternative, and what would the observed distribution look like if the cost asymmetry were removed? The intervention catalogue — anonymity, list experiments, secret ballots, pioneer dissent, blind review, randomised response, whistleblower channels — transfers identically across substrates, because each targets the expression cost asymmetry that drives the gap in all of them.
The transfer is deep because the same structure licenses both historical reasoning and design. Sudden mass reorientations — the 1989 Eastern European cascades, the rapid collapse of de jure segregation, the rapid mainstreaming of same-sex-marriage acceptance, the harassment cascades — are recognisable as preference-cascade releases of long-falsified preferences rather than mass opinion change, marked by the diagnostic signature of rapid public reorientation following a small visible defection, with reports of long private dissent emerging from many participants. Romania in December 1989 makes this concrete: four decades of apparent public consensus masked an enormous held-versus-expressed gap sustained by severe expression cost, and the falsifying majority had no way to discover itself until a small section of a regime-organised crowd began to boo, the visible defection broke the cost structure, and the booing spread through the crowd in minutes as those who had been falsifying saw that defection was now possible and that they were not alone — the held distribution unchanged in those four days, only the expression cost structure. The same structure recurs in the post-meeting hallway where a junior engineer voiced no concern in the all-hands but sharp concern privately, and in the post-crash recognition that "everyone secretly thought the market was overvalued." On the design side, anyone building a mechanism for surfacing held preferences — an organisational survey, electoral system, peer-review process, board meeting, or market for dissent — can use the prime to diagnose where the cost asymmetry sits and which intervention reduces it, and because the intervention family is substrate-neutral the same moves apply whether the substrate is a totalitarian polity or an engineering all-hands. The strip-the-jargon form ("if people face a cost for saying what they really think, they say something cheaper, and the public distribution drifts away from the private distribution") does load-bearing work across authoritarian politics, corporate dynamics, academic orthodoxy, markets, voting behaviour, and social movements.
Examples¶
Formal/abstract¶
Cast as a threshold model of the perception loop, preference falsification yields a precise account of why suppressed consensus is fragile. Each agent \(i\) privately opposes the regime but will voice that opposition publicly only if the perceived number of fellow dissenters already visible exceeds a personal threshold \(\theta_i\) — the point at which the safety of numbers outweighs the expression cost. The held distribution is the full set of private oppositions; the expressed distribution is only those agents whose threshold has been crossed. The decisive structural fact is the perception loop: each agent's threshold is evaluated against the visible count, which is itself the output of everyone's threshold decisions, so the visible count can sit stably at near-zero even when the held opposition is near-universal — a self-consistent equilibrium of mutual misperception. This equilibrium is fragile in a way the model makes exact: a single visible defection raises the observed count past the lowest threshold, that agent's defection raises it past the next, and the system can unravel in a cascade if the threshold distribution is dense enough that each newly-revealed dissenter unlocks the next. The held distribution never changes during the cascade — only the cost structure, perturbed by the first visible defection, shifts. The model thereby distinguishes the prime sharply from its neighbours: it is not conformity (no private preference is altered), not an information cascade (the driver is expression cost, not Bayesian updating on others' signals), and not costly signalling (the agents are hiding, not conveying, private information).
Mapped back: Private opposition is the held preference, the safety-in-numbers threshold encodes the expression-cost asymmetry, the visible dissenter count is the expressed distribution feeding the perception loop, and the cascade triggered by one defection is the cost-structure-shock fragility — preference falsification rendered as a threshold model whose equilibrium is consensus-masking and whose instability is exact.
Applied/industry¶
Romania in December 1989 and a corporate all-hands run the identical structure at vastly different scales. Under the Ceaușescu regime, four decades of apparent public consensus masked an enormous held-versus-expressed gap sustained by severe expression cost — the secret police made voicing opposition career-ending or worse. The falsifying majority had no way to discover itself, because the only available estimator of held opinion was the uniformly loyal expressed distribution. At a regime-organised rally, a small section of the crowd began to boo; this visible defection broke the perceived cost structure, and the booing spread through tens of thousands in minutes as each person who had been falsifying saw both that defection was now possible and that they were not alone. The held distribution did not change in those minutes — only the expression cost did — which is the prime's signature of a preference cascade rather than mass opinion change. The same structure, with the cost dialled down, governs a corporate all-hands: a junior engineer voices no concern about a flawed product decision in the meeting (where dissent carries reputational and career cost in front of senior leadership) but voices sharp concern in the post-meeting hallway, and a settled "unanimous" boardroom position turns out to be unanimous private skepticism. The matched intervention family is substrate-neutral and channel-targeted: anonymity (the secret ballot, the anonymous engineering survey) removes the cost asymmetry directly, aggregated revelation (list experiments, anonymized polls) yields population signal without individual exposure, and pioneer dissent (one visible defector, one leader who voices doubt first) lowers the isolation cost for everyone behind them. A reformer auditing a regime's apparent stability and a manager designing an all-hands feedback channel are reasoning about the same biased expressed distribution under a cost asymmetry.
Mapped back: Loyal public conduct and meeting silence are the strategically expressed preferences; private opposition and private skepticism are the held preferences; the secret police and the career-risk-in-front-of-leadership are the expression-cost asymmetries; the spreading boos and the hallway conversation are the cost-shock releases — the same prime in authoritarian politics and corporate decision-making.
Structural Tensions¶
T1 — Held Distribution versus Expressed Distribution (measurement). The load-bearing object is the gap between what agents hold and what they voice, and the visible distribution is a biased estimator of the held one whenever expression cost is asymmetric. The characteristic failure is reading public consensus as held consensus — treating the expressed distribution as a measurement of private opinion. The diagnostic is to estimate the expression-cost asymmetry first: where voicing the held view is costly, the visible distribution measures the cost structure as much as the preferences, and held opinion must be triangulated through anonymity, list experiments, or randomised response rather than trusted directly.
T2 — Apparent Stability versus Cascade Fragility (temporal). A large suppressed gap looks maximally stable — uniform public conformity — while being structurally fragile, since a small cost-structure shock can release the held distribution in days. The failure is inferring durability from surface unanimity, mistaking suppression for genuine consensus right up to the sudden cascade. The diagnostic is to ask what would happen to a visible defector: where the answer is "severe cost," apparent stability is evidence of fragile suppression, not agreement, and the very uniformity that looks reassuring is the signature of a gap poised to unravel on the first permission signal.
T3 — Falsification versus Genuine Opinion Change (sign/direction). When a consensus reverses overnight, two explanations compete: people changed their minds, or a stable held distribution was released by a cost-structure change. They point to opposite interventions. The failure is reaching for "they were persuaded" when the held distribution never moved — only expression became safe — misdiagnosing a cascade as mass conversion. The diagnostic is to ask not "what changed their minds?" but "what made expression safe?": the cascade signature (rapid reorientation following one visible defection, with reports of long-private dissent) marks released falsification, distinct from the slow drift of genuine persuasion.
T4 — Falsification versus Conformity (scopal). Falsification preserves the held preference while distorting the visible one; conformity adopts the group view so the held preference actually changes. The failure is collapsing the two — treating public agreement as either always genuine (missing falsification) or always fake (missing real conformity), which prescribes the wrong remedy. The diagnostic is to ask whether the private ordering is intact behind the public one: if removing the expression cost would surface a different view, it is falsification (intervene on the cost channel); if the agent has internalised the majority view, it is conformity, and anonymity reveals nothing because nothing is hidden.
T5 — Self-Referential Perception Loop (coupling). Each agent's expression decision uses an estimate of others' held distribution whose only available estimator is the expressed distribution — which is itself the output of every agent's decision, so misperception couples back on itself and the perceived climate drifts far from the held one. The failure is trusting the perceived climate as exogenous information, when it is downstream of the very falsification it appears to justify. The diagnostic is to break the loop with external measurement: a perceived consensus that everyone privately doubts is self-sustaining, and only an out-of-band channel (anonymous poll, secret ballot) can reveal that the climate is an artefact of mutual misreading.
T6 — Cost-Channel Decomposition versus Aggregate Cost (scalar / local-global). Expression cost is not a single quantity but decomposes into reputational, professional, social, political, and material channels, and an intervention that lowers one may leave the binding one untouched. The failure is treating "the cost of speaking up" as monolithic — installing an anonymous survey (removing reputational cost) while professional retaliation (the actual binding channel) remains, so falsification persists. The diagnostic is to identify which channel dominates for this population and target it specifically: aggregate cost is the wrong unit, and a remedy aimed at a non-binding channel produces the appearance of openness without the substance.
Structural–Framed Character¶
Preference falsification sits at the far framed end of the structural–framed spectrum: it is fully framed, with an aggregate of 1.0 and every one of the five criteria scored at 1.0. It is among the most thoroughly human-social primes in the catalog, and every diagnostic points the same way — toward framed.
There is, to be sure, an abstract skeleton: a held distribution, an expression-cost asymmetry, an expressed distribution, and a self-referential perception loop that can cascade. But that skeleton cannot be lifted out of human society at any point. It is fully human_practice_bound (1.0): "preference," "public expression," and "social cost" presuppose agents embedded in a polity, a workplace, a congregation, or an academy — there is no physical or biological substrate where this runs. Its home vocabulary travels with it intact (vocab_travels 1.0): preference, falsification, conformity, reputational cost, and preference cascade are imported wholesale wherever the pattern is invoked, with no domain telling it purely in its own words. It carries heavy evaluative weight (evaluative_weight 1.0): "falsification," conformity under penalty, and the implicit critique of repressive cost structures are normatively loaded throughout. Its origin is squarely institutional (institutional_origin 1.0), rooted in political science and the sociology of public opinion and collective action — authoritarian polities, public-opinion dynamics, religious conformity, workplace and academic speech. And invoking it imports a full interpretive frame (import_vs_recognize 1.0): to call a silence "preference falsification" is to read a private held preference, a cost asymmetry, and strategic misrepresentation into observed behaviour, not merely to recognise a pattern already wired into a physical system. The substrate clustering confirms it — every instance is social-political, and the substrate-independence score is correspondingly low. This is a paradigm framed prime, and the prose makes no attempt to inflate it toward structural.
Substrate Independence¶
Preference falsification is a low-substrate-independence prime — composite 2 / 5 on the substrate-independence scale. Its domain breadth is narrow and tightly clustered: every substrate is human social-political — authoritarian and totalitarian polities (its original home, with the 1989 Eastern European collapses as textbook preference cascades), public opinion in democracies (spiral-of-silence, shy-voter effects), religious orthodoxy, workplaces, academic orthodoxy, harassment underreporting, markets (analysts shading toward the dominant narrative), social-media self-censorship, and pre-cascade conditions in social movements — with no reach whatsoever into physical, biological, or formal domains. Its structural abstraction is correspondingly low: there is an abstract skeleton (a held preference distribution, an expression-cost asymmetry, an expressed distribution, and a self-referential perception loop that can cascade), but it cannot be lifted out of human society at any point — "preference," "public expression," and "social cost" presuppose agents embedded in a polity, congregation, or academy, and the home vocabulary (falsification, conformity, reputational cost, preference cascade) travels wholesale rather than being re-stated in each field's own terms. Its transfer evidence is the only component above the floor (3): the mechanism — asymmetric cost on expression — is generic enough that the same held-versus-expressed gap, and the same rapid-cascade release when the cost drops, is documented across the social substrates with recognisable structure (the Eastern-bloc collapse, post-meeting hallway candour, post-tenure reorientation, post-crash "everyone knew"). But because that transfer never leaves the human-social band, and the prime imports a full interpretive frame to apply, the composite is held to a low 2.
- Composite substrate independence — 2 / 5
- Domain breadth — 2 / 5
- Structural abstraction — 2 / 5
- Transfer evidence — 3 / 5
Relationships to Other Primes¶
Parents (2) — more general patterns this builds on
-
Preference Falsification is a kind of Private-Public Preference Divergence
child of emergent private_public_preference_divergence
-
Preference Falsification presupposes, typical Conformity
Operates against a conformist baseline (the cheaper public stance) under expression-cost asymmetry; but the file is emphatic it is NOT conformity (held preference preserved, not adopted). Recorded as a tentative composition-adjacency, low conviction; owner may drop.
Children (1) — more specific cases that build on this
-
Spiral Of Silence In Publics is a kind of Preference Falsification
preference_falsification (cand) is the expressive PROCESS (public preference diverges from private under expression-cost asymmetry); its own file states "spiral_of_silence (and pluralistic_ignorance) is the cognitive climate falsification PRODUCES — this prime is the expressive process that produces and sustains it." spiral_of_silence is candidate-valid and giant-connected. parent_of spiral_of_silence (process -> resulting state) is directionally sound and bridges the cluster (conformity/groupthink are co-members). Medium: the produces-relation is a generative tie rather than a taxonomic is-a, but direction is unambiguous. (Its file explicitly rejects conformity, information_cascade, signaling, and preference_heterogeneity as parents.)
Path to root: Preference Falsification → Conformity
Neighborhood in Abstraction Space¶
Preference Falsification sits among the more crowded primes in the catalog (28th percentile for distinctiveness): several abstractions describe nearly the same structure, so a description that fits it will tend to fit its neighbors too — transporting it usually means disambiguating within this family rather than landing on it exactly.
Family — Public-Private Belief Divergence (13 primes)
Nearest neighbors
- Private-Public Preference Divergence — 0.80
- Spiral Of Silence In Publics — 0.73
- Revealed Preference — 0.72
- Abilene Paradox — 0.72
- Public vs. Private Contexts — 0.71
Computed from structural-signature embeddings · 2026-06-14
Not to Be Confused With¶
The most consequential confusion is with conformity, because both end in a person publicly agreeing with a group they might privately resist. The distinction is what happens to the held preference. Conformity is genuine adoption: under social influence the agent comes to actually hold the majority view, internalising it, so there is no private dissent left behind the public agreement. Preference falsification preserves the held preference and distorts only the expressed one: the agent still privately holds the costlier view and voices the cheaper one strategically, so a private/public gap persists. This difference dictates opposite remedies and opposite predictions. Against falsification, anonymity reveals the hidden view because there is a hidden view to reveal; against conformity, anonymity reveals nothing because nothing is hidden — the agent has actually changed. And only falsification carries cascade fragility: because the held distribution is intact and merely suppressed, a small cost-structure shock can release it suddenly, whereas a genuinely conformed population has no suppressed distribution to surface. The diagnostic is to ask whether removing the expression cost would surface a different view: if yes, it is falsification (intervene on the cost channel); if the agent has internalised the majority view, it is conformity.
A second genuine confusion is with the information_cascade, since both can produce uniform public behaviour that diverges from private information. An information cascade is Bayesian: agents observe others' visible actions, rationally infer that those actions reveal private signals, and copy them, so a sequence of observable choices drives a herd through inference. Preference falsification is cost-driven: agents voice the cheaper preference not because they have updated their beliefs by observing others but because voicing the held view is penalised. The drivers are entirely different — inference about others' signals versus asymmetric cost on one's own expression — and so are the fixes. An information cascade is broken by revealing private signals or by introducing independent information; falsification is broken by removing the expression-cost asymmetry (anonymity, aggregated revelation). Conflating them leads to treating a suppression problem as an information problem, supplying more information to a population that already privately disagrees and is merely afraid to say so.
A third confusion is with signaling, and here the relationship is inversion. Costly signalling conveys private information: an agent takes a costly action precisely because the cost makes the signal credible, deliberately revealing a type or quality that cheap talk could not. Preference falsification does the opposite — it hides private information by avoiding the costly act of voicing the held view, substituting a cheaper, conformist expression that conceals the agent's true ordering. Signalling pays a cost to be believed; falsification avoids a cost to stay hidden. Reading falsification as a signalling phenomenon inverts its informational sign and would lead an analyst to treat the expressed distribution as an honest (if costly) revelation of held preference, when it is in fact a deliberate concealment of it — the precise opposite of what signalling delivers.
These distinctions matter because each mis-framing prescribes the wrong intervention: a conformity framing expects anonymity to reveal nothing and abandons it, an information-cascade framing supplies information to a population that already privately disagrees, and a signalling framing trusts the expressed distribution as honest revelation — whereas the prime directs the analyst to the expression-cost asymmetry and its channels (reputational, professional, social, political, material), the variables that anonymity, list experiments, and pioneer dissent actually move.
Solution Archetypes¶
No catalogued solution archetypes reference this prime yet.