Naive Realism¶

Explain intractable disagreement as the converging conviction that one perceives the world directly, that all rational observers would agree, and that dissenters must be biased — relocating the dispute from the facts to each party's invisible construction of them.

Core Idea¶

Naive realism, as defined in social psychology by Lee Ross and Andrew Ward (1996), is the converging cognitive disposition in which a person (1) believes they perceive the social world as it actually is — directly, without mediation or bias; (2) believes that any other rational, well-informed person exposed to the same information would arrive at the same view; and (3) when others do disagree, attributes that disagreement not to any construction or selection operating on their own perception, but to the others' ignorance, irrationality, or motivated bias. The three limbs together produce a characteristic asymmetry: one's own view is experienced as the neutral reading of reality, and opposing views require explanation in terms of defect.

The mechanism is epistemic ego-centrism: the constructive processes that shape attention, encoding, and interpretation of social information — prior commitments, motivated reasoning, selective exposure, social position — are invisible to the person whose perception they structure, while those same processes are readily attributed as distorting agents when explaining why others see things differently. The asymmetry propagates directly into intergroup conflict: each side experiences its own concessions as principled and the other side's matching concessions as inadequate, because each side's own position appears to be the obvious, unbiased reading and the other's appears to be a distortion. The hostile media effect — in which partisans on both sides of a political controversy judge the same news coverage as biased against their side — is a canonical empirical signature of naive realism in action: each group's naive-realist position is the same (my view is neutral, the media deviates toward the opponent), producing symmetric but opposed complaints about a single piece of content. The term also has an older, distinct meaning in philosophy of perception, where naive realism or direct realism is the pre-critical view that perception delivers the external world without any intermediating mental representation; the psychological usage builds on that philosophical substrate and adds the social-attribution asymmetry that gives it empirical bite.

Structural Signature¶

Sig role-phrases:

the unmediated-perception limb — the perceiver experiences their own construction of the social situation as the situation itself, directly and without bias
the universalization limb — they expect every other rational, well-informed observer given the same information to reach the same view
the attribution limb — when others disagree, the gap is charged to the others' ignorance, irrationality, or bad faith, never to one's own construction
the invisible asymmetry — the self-objective / other-biased split is itself unseen by the perceiver (epistemic ego-centrism)
the source-keyed evaluation — identical content (the same coverage, concession, or proposal) earns "neutral" from one's own side and "biased" from the other, flipping with attribution alone
the conflict signatures — the hostile-media symmetry, the concession trap, and the bias blind spot all follow from the asymmetry
the information-remedy failure — more facts do not converge the parties, because both assimilate them toward the prior; the obstacle is a missing meta-cognitive step
the construction-surfacing intervention — only moves that make one's own view visible as a construction (perspective-taking, blinded attribution, joint fact-finding) gain traction

What It Is Not¶

Not the philosophical doctrine of direct realism. The older epistemological sense — that perception delivers the external world without intermediating representation — is a thesis about perception, not a documented cognitive bias. The social-psychological usage builds on that substrate but adds a third limb: the social-attribution asymmetry by which disagreement is charged to the other's defect. It is that limb, not the perceptual claim, that gives the effect its empirical bite.
Not just confirmation bias. Confirmation bias is selective gathering and weighting of belief-consistent evidence. Naive realism is the broader three-limbed posture — unmediated self-perception plus the expectation that all rational observers share it plus the attribution of disagreement to the other's bias. Confirmation bias may feed it, but it does not capture the universalization or the self-objective/other-biased asymmetry.
Not the false-consensus effect. Overestimating how widely others share one's view is one consequence of naive realism's universalization limb, not the whole disposition. The effect's distinctive content is the attribution asymmetry that follows when the expected consensus fails to materialize and the gap is blamed on the dissenter.
Not a claim that one party really is objective and the other really is biased. The structure is symmetric: both sides are naive realists, each experiencing its own reading as neutral and the other's as distorted. The hostile-media effect makes this vivid — both sides judge the same coverage as slanted against them — so naming the effect does not adjudicate who is right; it explains why each is certain it is.
Not a problem more information can fix. Because both parties assimilate the same new evidence toward their prior, "just give them more facts" predictably fails to converge them. The obstacle is a missing meta-cognitive step — recognizing one's own view as a construction — so traction lies with moves that surface the construction (perspective-taking, blinded attribution), not with settling the facts.
Not the bare map/territory failure of a lone modeler. Stripped of the social-attribution limb, what remains is just map_territory_distinction — mistaking one's model for the world — which recurs anywhere a representation is taken for the represented. Naive realism's distinctive third limb presupposes other minds in disagreement and an attribution process; invoking it for a solitary modeler over-reaches into territory the parent already covers.

Scope of Application¶

Naive realism lives across the social-cognition subfields that study disagreement — social and political psychology, conflict resolution, and lay epistemology — wherever agents under dispute each treat their own construction as the unmediated world; its reach is within that one substrate seen in different arenas, and the bare map-mistaken-for-territory failure of a lone modeler belongs to its map_territory_distinction parent, not here.

Political polarization — the hostile media effect, in which partisans on both sides judge the same news report as biased against them, is the canonical empirical signature.
Conflict resolution and negotiation — disputants treat their own view as obvious and the other's as requiring explanation by bias; the Ross & Stillinger blinded-proposal studies show identical compromise terms judged unfair when labelled as the opponent's.
Marital and interpersonal conflict — each partner experiences the conflict-relevant facts as plainly obvious and the other's resistance as wilfulness or distortion.
Lay theory of bias — the bias blind spot (people detect bias more readily in others than in themselves) is the self-objective/other-biased asymmetry the disposition predicts.
Cross-cultural miscommunication — each culture experiences its own conventions as natural and the other's as arbitrary or irrational, the same source-keyed reading across national lines.
Online and platform-mediated discourse — algorithmic feeds plus naive realism produce the conviction that "everyone reasonable sees what I see" while users in fact inhabit partially distinct information environments.
Policy-relevant scientific controversy — in contested arenas (climate, public-health interventions, nutrition), opposing camps assimilate the same evidence toward their prior and charge the other with bias rather than recognizing a construction.

Clarity¶

Naming naive realism reframes a recurring puzzle in conflict research: why rational, well-informed people handed the same nominal facts disagree so sharply about what is happening. Without the concept the natural diagnosis is that one side simply has the facts wrong, which sends the analyst hunting for who is misinformed. Naive realism replaces that question with a sharper one: not who is right about the facts, but which construction of the facts each party is using, and why that construction is invisible to the very person it structures. The decisive move is to relocate the disagreement from the world to the perceivers' processing of it — and to make legible the asymmetry that the folk view never notices, in which one's own reading is felt as the neutral, unmediated reading of reality and only the opponent's reading registers as a reading, requiring explanation by bias or bad faith.

That asymmetry is exactly what the label makes diagnosable, and it explains why intergroup conflict is so sticky. Because each side experiences itself as objective and the other as distorted, each judges its own concessions as principled and the other's matching concessions as inadequate or manipulative — symmetric content, opposite evaluations, keyed only to source. With the concept in hand, a mediator can predict the failure of "just give them more information" (both sides assimilate the same evidence toward their prior) and can locate the real obstacle precisely: not a shortage of facts but a meta-cognitive step each party resists — recognizing one's own view as a construction rather than as the way things plainly are. The productive question becomes whether an intervention can surface that construction, not whether it can settle the facts.

Manages Complexity¶

Conflict research accumulates a long roster of separately-named phenomena — the hostile media effect, the false-consensus effect, the bias blind spot, biased assimilation of identical evidence, the unfairness rating that flips when a proposal is relabelled with the other side's name, the mutual charge of bad faith in marital and intergroup disputes — and each, studied on its own, invites its own mechanism. Naive realism collapses that roster onto a single generative posture: epistemic ego-centrism, the three-limbed disposition in which one experiences one's own construction of the social world as the unmediated world itself, expects all rational observers to share it, and explains disagreement by the other's bias, ignorance, or bad faith. From that one posture the listed phenomena are not separate findings to be explained but derivations to be read off, and the analyst predicts the qualitative course of a conflict by tracking a few quantities the structure names rather than re-theorising each case. How invisible each party's own construction is to that party sets whether the disagreement will be attributed to the world or to the opponent's defect; how source-keyed the evaluation is sets the signature asymmetry, where identical content — the same coverage, the same concession, the same compromise terms — earns "neutral" from one's own side and "biased" or "self-interested" from the other, so the evaluation flips with attribution alone while the content holds fixed. Those two parameters generate the hostile-media symmetry (both sides see the same report as slanted against them), the concession trap (each side's matching concessions judged principled-by-us, inadequate-by-them), and the bias blind spot (distortion detected readily in others, not in oneself) as one family. The branch structure is equally decisive for intervention: because the obstacle is a missing meta-cognitive step — recognising one's own view as a construction — rather than a shortage of facts, the analyst reads off in advance that "give them more information" will fail (both sides assimilate the same evidence toward their prior) and that traction lies only with moves that surface the construction (perspective-taking, blinded attribution, joint fact-finding). A high-dimensional "why do these many conflict biases each arise, and what fixes each" problem reduces to one posture plus a two-parameter read on construction-invisibility and source-keying, off which the mutual mis-attribution, the stickiness, and the failure mode of information-based remedies all follow.

Abstract Reasoning¶

Naive realism licenses a relocation move that the conflict analyst runs when rational, well-informed people handed the same nominal facts disagree sharply about what is happening. The diagnostic reasons FROM "both sides treat their own reading as the plain truth and the other's as requiring explanation" TO "the disagreement lives in the perceivers' construction of the facts, not in the facts," and so replaces the question "who has the facts wrong?" with "which construction of the facts is each party using, and why is that construction invisible to the very person it structures?" The signature the analyst looks for is source-keyed evaluation: identical content earning opposite verdicts that flip with attribution alone. The cleanest test is a blinded-attribution manipulation — when the same compromise terms are judged unfair labelled as the opponent's and reasonable labelled as one's own, the analyst infers naive realism rather than a genuine disagreement about the terms, because the content held fixed while only the source changed. The hostile media effect is read the same way: when partisans on both sides judge the same report as biased against them, the analyst reasons that each side's own view is felt as neutral and the coverage as deviating toward the opponent — a symmetric, opposed pair of complaints derived from one posture, not two separate biases.

The predictive moves read two parameters the structure names. How invisible each party's own construction is to that party predicts whether disagreement gets attributed to the world or to the opponent's defect — the more invisible the construction, the more the analyst predicts mutual charges of ignorance, irrationality, or bad faith. How source-keyed the evaluation is predicts the signature asymmetries: the concession trap, where each side judges its own matching concessions principled and the other's inadequate or manipulative; and the bias blind spot, where distortion is detected readily in others and not in oneself. From these the analyst forecasts the stickiness of a conflict and, crucially, the failure of information-based remedies: because both sides assimilate the same new evidence toward their prior, the analyst predicts that "just give them more facts" will not converge them and reasons that the obstacle is not a shortage of facts but a missing meta-cognitive step — recognizing one's own view as a construction rather than as the way things plainly are. The interventionist corollary follows directly: traction lies only with moves that surface the construction — perspective-taking that forces articulation of the other's reasoning, blinded attribution that strips the source cue, joint fact-finding that builds a shared information environment — so the analyst reasons that an intervention should target the meta-cognitive step, not the facts, and predicts a remedy attacking the facts alone will fail. The boundary-drawing move scopes the construct to the three-limbed disposition — unmediated self-perception, expectation that all rational observers share it, attribution of disagreement to others' defect — applied to social cognition under disagreement: the analyst predicts the source-keyed asymmetry wherever those limbs converge in an intergroup or interpersonal dispute, and reasons that the third limb, the social-attribution asymmetry, is what gives the effect its bite beyond the general map-is-not-territory failure of treating one's model as the world.

Knowledge Transfer¶

Within social and political psychology, conflict resolution, and lay-epistemology research naive realism transfers as mechanism, because all of these are the same social-cognitive substrate — agents under disagreement — seen in different arenas. The diagnostic carries intact from political polarization (the hostile media effect) to negotiation (the relabelled-proposal asymmetry), to marital and team conflict, to cross-cultural miscommunication, to platform-mediated discourse, to policy-relevant scientific controversy: in each the analyst relocates the disagreement from the facts to the perceivers' construction of them, looks for source-keyed evaluation (identical content earning opposite verdicts that flip with attribution alone), and reads the two parameters the structure names — how invisible each party's construction is to that party, and how source-keyed the evaluation is — to forecast mutual mis-attribution, the concession trap, the bias blind spot, the stickiness of the conflict, and the failure of "give them more information." The intervention family travels with the diagnosis because it follows from it: perspective-taking, blinded attribution, structured dialogue, and joint fact-finding all target the missing meta-cognitive step (recognizing one's own view as a construction) rather than the facts, and they apply in a boardroom, a marriage, or a diplomatic table alike. The vocabulary travels because the substrate does: epistemic ego-centrism, the three limbs, the self-objective/other-biased asymmetry mean the same thing wherever rational people disagree.

Beyond social cognition this entry is a clean case of a general pattern recurring across domains while the named effect's distinctive cargo stays home. Strip the conflict context and the residue is a general failure already in the catalog: an agent treats its own model of a domain as the domain itself, failing to model the model — which is map_territory_distinction (and the broader model_dependency family of map-versus-territory failures). That general pattern genuinely recurs far outside intergroup conflict — in measurement, modeling, perception, any place a representation is mistaken for the represented — and those cross-domain instances are co-instances of the map/territory parent, not analogies to naive realism. So the honest report is that the cross-domain lesson should be carried by the map/territory parent, not by "naive realism."

What makes the named effect more than that parent — and what does not travel — is its third limb: the social-attribution asymmetry in which disagreement is explained by the other's ignorance, irrationality, or bad faith, producing the source-keyed, symmetric-but-opposed signature (both sides see the same coverage as biased against them; each judges its own concessions principled and the other's inadequate). That limb presupposes other minds in disagreement and an attribution process; it has no referent in the bare map/territory failure of a lone modeler mistaking model for world. It is exactly the limb that gives naive realism its empirical bite in conflict research, and exactly the limb that stays home. Loose invocations of "naive realism" for any case of someone mistaking their model for reality therefore over-reach: absent the social-attribution asymmetry, what is being described is just the map/territory parent, and should be named as such. The clean summary: within social cognition under disagreement the effect transfers as mechanism and remedy; beyond it, carry the map/territory (and model-dependency) parent across domains, while naive realism's universalization-plus-attribution machinery — the source-keyed asymmetry, the hostile-media and blinded-proposal paradigms — stays bound to the social arena. See Structural Core vs. Domain Accent.

Examples¶

Canonical¶

The founding demonstration is Vallone, Ross, and Lepper's 1985 study of the "hostile media phenomenon." Pro-Israeli and pro-Arab partisans watched the same television news segments covering the 1982 Beirut massacre (Sabra and Shatila). Each group judged the identical coverage as biased against its own side and favorable to the other, and each believed that neutral viewers would be turned against its side by what they saw. The two groups even "saw" different content, disagreeing about how many pro- and anti-Israel references the footage contained. Because the stimulus was held fixed while the verdicts diverged and mirrored one another, the disagreement could not live in the coverage itself; it lived in each group's conviction that its own reading was the neutral one and any deviation marked hostile bias.

Mapped back: Each group's experience of its own view as the plain, fair reading is the unmediated-perception limb; the expectation that uncommitted viewers would end up seeing it their way is the universalization limb. Judging the identical broadcast as slanted against oneself is the source-keyed evaluation, and the symmetric, mutually opposed complaints are the hostile-media form of the conflict signatures — both derived from one posture, not two separate biases.

Applied / In Practice¶

In a study directly relevant to Middle East peace negotiations, Maoz, Ward, Katz, and Ross (2002) presented Israeli Jewish participants with an actual peace proposal but varied its stated authorship: for half it was attributed to an Israeli source, for half to a Palestinian source. The identical proposal was rated more favorably when labeled Israeli than when labeled Palestinian — participants devalued the very terms they would otherwise endorse, purely because the disfavored side had ostensibly offered them. The content was constant; only the attribution changed. This "reactive devaluation" is exactly the source-keyed evaluation naive realism predicts, and it explains a persistent obstacle at the negotiating table: a concession reads as principled from one's own side and as a trick or as inadequate from the other, so identical terms cannot be judged on their merits.

Mapped back: The verdict flipping with authorship while the proposal held fixed is the source-keyed evaluation in its cleanest, blinded-attribution form. Downgrading the terms simply because the opponent offered them is the attribution limb at work, and the resulting devaluation of matching concessions is the concession-trap member of the conflict signatures — which is why simply tabling better facts or terms (the information-remedy failure) cannot by itself move the parties.

Structural Tensions¶

T1: Relocating the dispute versus dissolving the possibility of being right (the meta-move that can corrode adjudication). Naive realism's decisive gift is to relocate a disagreement from the facts to each party's invisible construction of them, replacing "who has the facts wrong?" with "which construction is each using, and why is it invisible?" This is exactly what exposes the source-keyed asymmetry the folk view never notices. But the same relocation, applied without discipline, threatens to make every disagreement a symptom of construction and none a matter of who is actually correct — a slide into the view that all positions are equally "just constructions." The concept is careful that it "does not adjudicate who is right," only why each is certain; yet a tool that treats conviction-of-objectivity as itself the diagnosis can be turned against any well-grounded claim, dissolving the distinction between a constructed reading and an accurate one. Diagnostic: Is the construction-frame being used to explain the stickiness of a disagreement, or to deny that either party could be substantively right about the facts?

T2: Symmetric structure versus asymmetric reality (both sides are naive realists, but they need not be equally wrong). The structure is symmetric: both parties experience their own reading as neutral and the other's as distorted, which the hostile-media effect makes vivid — both sides see the same coverage as slanted against them. That symmetry is what lets the analyst diagnose the posture without taking sides. But symmetry of posture is not symmetry of accuracy: two disputants can be equally certain of their objectivity while one's construction tracks reality far better than the other's, and the concept's even-handedness can be mistaken for a verdict of equivalence — a "both sides do it" that flattens genuine asymmetries of evidence into mere mirrored bias. The methodological symmetry that makes the diagnosis fair is the same symmetry that can launder a real disparity in warrant into false balance. Diagnostic: Does symmetric naive realism here imply the two constructions are equally warranted, or only that both parties are equally convinced while their evidential standing may differ?

T3: The third limb as the effect's bite versus the parent it rides on (attribution is what travels least and matters most). What makes naive realism more than the generic map/territory failure is its third limb — the social-attribution asymmetry that charges disagreement to the other's ignorance, irrationality, or bad faith, producing the source-keyed, symmetric-but-opposed signature. That limb is exactly what gives the effect empirical bite in conflict research. But it is also precisely the limb that presupposes other minds in disagreement and an attribution process, so it has no referent in the bare map/territory failure of a lone modeler — meaning the feature that constitutes the named effect is the one that cannot be reduced to its parent, while everything reducible (mistaking model for world) is what the parent already covers. The effect's identity lives entirely in the increment its parent lacks. Diagnostic: Does the case involve other minds and an attribution of their disagreement to defect (naive realism proper), or a solitary agent mistaking model for world (just the map/territory parent)?

T4: Information-remedy failure versus the counsel of despair (predicting more facts won't converge can license giving up on facts). A sharp, useful prediction: because both sides assimilate the same new evidence toward their prior, "just give them more information" fails, so the obstacle is a missing meta-cognitive step, not a shortage of facts. This correctly redirects intervention toward surfacing the construction. But the prediction can harden into a counsel of despair — if facts are inert, why establish them at all? — which overreads the claim: information fails to converge the parties by itself, not that facts are worthless or that joint fact-finding (an intervention the concept endorses) is pointless. The very insight that saves a mediator from the naive "more data" move can, taken too far, discredit the shared-information-environment building the concept prescribes as a genuine remedy. Diagnostic: Is "information won't fix it" being used to redirect effort toward surfacing constructions, or to abandon the joint fact-finding the concept actually recommends?

T5: Diagnosing others' naive realism versus one's own (the concept is a bias about bias, and self-application is the hard part). The bias blind spot — detecting distortion readily in others, not in oneself — is one of naive realism's own predicted signatures, and the concept's power is that it names the self-objective/other-biased asymmetry the perceiver cannot see. But this makes the concept unusually prone to reflexive misuse: an analyst can wield "you're being a naive realist" as a rhetorical weapon that itself enacts the third limb (charging the other's disagreement to a cognitive defect), and the tool built to expose the asymmetry becomes a new vehicle for it. The construct that diagnoses the invisibility of one's own construction is the same construct whose invocation most tempts one to exempt oneself from the diagnosis. Diagnostic: Is the naive-realism label being applied to surface one's own construction as construction, or deployed to dismiss an opponent's view as mere bias — reproducing the attribution limb it names?

T6: Autonomy versus reduction (a named social-cognitive effect or the map/territory pattern its parent carries). "Naive realism" is a canonically studied social-psychology disposition with proprietary machinery — the three limbs, the hostile-media and blinded-proposal paradigms, the concession trap, epistemic ego-centrism, the reactive-devaluation and source-keyed signatures — and within social cognition under disagreement it transfers as mechanism and remedy across polarization, negotiation, marital conflict, cross-cultural miscommunication, and platform discourse. But strip the conflict context and the residue is a general failure already in the catalog: an agent treats its own model of a domain as the domain itself, failing to model the model — map_territory_distinction (and the broader model_dependency family). That parent recurs far outside intergroup conflict — measurement, modeling, perception — as genuine co-instances, and it is the parent, not the named effect, that should carry the cross-domain lesson. What stays home is the third limb: universalization-plus-attribution presupposes other minds in dispute and has no referent in a lone modeler's map/territory slip. Diagnostic: Resolve toward the parent (map_territory_distinction / model_dependency) wherever a representation is mistaken for the represented; toward named naive realism only where the social-attribution asymmetry among disagreeing minds is doing the work.

Structural–Framed Character¶

Naive realism sits at the framed-leaning position on the structural–framed spectrum — a real regularity of human social cognition, but one whose distinctive, non-reducible content is constituted by a social practice of attribution among disagreeing minds. The criteria mostly point framed, with a structural thread. On evaluative_weight it leans mildly framed: it is a documented bias, so naming it carries the tincture of a defect-verdict — though the entry is careful that it "does not adjudicate who is right," only why each party is certain, which keeps the weight lighter than a fallacy's. Human_practice_bound is high for what makes the effect itself: the load-bearing third limb — charging disagreement to the other's ignorance, irrationality, or bad faith — presupposes other minds in disagreement and an attribution process, so the distinctive machinery runs only in a social arena and dissolves for a solitary cognizer. Institutional_origin is mixed: the phenomenon is a genuine cognitive regularity (people are naive realists whether or not Ross and Ward named it), but the named construct — the three limbs, epistemic ego-centrism, the hostile-media and blinded-proposal paradigms — is social-psychology apparatus. On vocab_travels it scores low, and on import_vs_recognize it patterns as recognition within social cognition (across polarization, negotiation, marital and cross-cultural conflict) but as the parent recurring beyond it, not naive realism imported.

The portable structural skeleton is the map/territory (model-dependency) failure — an agent treats its own model of a domain as the domain itself, failing to model the model. That skeleton is genuinely substrate-spanning, recurring observer-free wherever a representation is mistaken for the represented (measurement, modeling, perception), which is the structural thread. But it does not lift naive realism off the framed side, because that skeleton is exactly what naive realism instantiates from its umbrella prime — map_territory_distinction (and the broader model_dependency family) — not what makes "naive realism" itself travel: the cross-domain reach belongs to that parent, while the distinctive increment, the social-attribution asymmetry that gives the effect its empirical bite, presupposes disagreeing minds and has no referent in a lone modeler's map/territory slip, so it stays home. Its character: a genuine cognitive regularity whose reducible core is the substrate-neutral map/territory failure but whose distinctive, non-portable third limb is a socially-constituted attribution asymmetry, leaving it framed-leaning rather than a free-floating prime.

Structural Core vs. Domain Accent¶

This is the section that settles why naive realism is a domain-specific abstraction and not a prime — and it carries the case for its domain-specificity, since there is no separate section for that.

What is skeletal (could lift toward a cross-domain prime). Strip the conflict away and a thin relational structure survives: an agent treats its own model of a domain as the domain itself, failing to model the model. The pieces that travel are abstract — a representation, a represented, and a cognizer who runs the first as though it were the second, with no register for the mediation in between. That skeleton is genuinely substrate-spanning: it recurs observer-free wherever a map is mistaken for its territory — in measurement, in modeling, in perception — and its recurrence there is mechanism, not metaphor. It is exactly why naive realism instantiates map_territory_distinction (and the broader model_dependency family) as its parent. But it is the core naive realism shares, not what makes it the named effect.

What is domain-bound. Everything that makes the effect naive realism in particular presupposes a social arena of disagreeing minds. Its load-bearing third limb — charging a dissenter's disagreement to their ignorance, irrationality, or bad faith while one's own reading is felt as the unmediated world — has no referent without other minds in disagreement and an attribution process running between them. The apparatus is social-psychology furniture: epistemic ego-centrism, the three limbs (unmediated perception, universalization, attributed defect), the self-objective / other-biased asymmetry, and the empirical paradigms that give it bite — the hostile-media effect (both sides judging the same coverage as slanted against them), the blinded-proposal / reactive-devaluation studies, the concession trap, the bias blind spot, the source-keyed evaluation that flips with attribution alone. The decisive test: remove the disagreeing minds and the attribution, and what is left — a solitary modeler mistaking model for world — is no longer naive realism at all but the bare map/territory failure. The distinctive increment is constituted by the very social context 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. Naive realism's transfer is bimodal. Within social cognition under disagreement it travels intact — from political polarization to negotiation to marital conflict to cross-cultural miscommunication to platform-mediated discourse — because each arena supplies the one thing it needs: disagreeing parties, each experiencing its own construction as neutral and the other's as distortion. Beyond that arena the named effect does not travel; only the parent recurs, and its cross-domain instances (a mis-calibrated instrument, an over-trusted model, a perceptual illusion) are co-instances of map_territory_distinction, not analogies to naive realism. So when the bare structural lesson is needed across domains, it is already carried, in more general form, by the parent the effect instantiates: the map/territory (and model-dependency) failure. The cross-domain reach belongs to that parent; "naive realism," as named, carries the social-attribution machinery — universalization plus defect-attribution among disagreeing minds — that should stay home.

Relationships to Other Abstractions¶

Current abstraction Naive Realism Domain-specific

Parents (1) — more general patterns this builds on

Naive Realism is a kind of Bias Prime

Naive realism is a bias specialized to a systematic self-objective and other-defective displacement in judgments about mediated social reality.

Children (2) — more specific cases that build on this

Bias Blind Spot Domain-specific is a decomposition of, typical Naive Realism

Bias blind spot typically realizes naive realism's self-objective and other-biased asymmetry, but its exact channel-mismatch mechanism can occur without a live disagreement or universal-consensus expectation.
Hostile Media Effect Domain-specific is a decomposition of, typical Naive Realism

Hostile-media perception typically realizes naive realism when each partisan experiences its own reading as neutral and explains balanced coverage's disagreement with that reading as bias toward the opponent.

Hierarchy path (1) — routes to 1 parentless root

Naive Realism → Bias

Not to Be Confused With¶

Philosophical direct realism (the older namesake). The pre-critical epistemological thesis that perception delivers the external world without any intermediating mental representation — a claim about perception, not a documented cognitive bias. The social-psychological usage builds on this substrate but adds a third limb, the social-attribution asymmetry, that the perceptual doctrine lacks. Tell: is the claim that we perceive reality directly with no mental mediation (philosophical direct realism), or that people charge disagreement to the other's defect while feeling their own view neutral (the psychological effect)?
Confirmation bias. The selective gathering and weighting of belief-consistent evidence. It may feed naive realism, but it does not capture the universalization limb (expecting all rational observers to agree) or the self-objective/other-biased attribution asymmetry. Tell: is the failure in how evidence is sought and weighted (confirmation bias), or in the three-limbed posture that one's own reading is the unmediated world and dissenters are biased (naive realism)?
False-consensus effect. Overestimating how widely others share one's view. This is one consequence of naive realism's universalization limb, not the whole disposition — it lacks the attribution asymmetry that fires when the expected consensus fails and the gap is blamed on the dissenter. Tell: is the claim merely that people overestimate agreement (false-consensus), or that they additionally attribute disagreement to the other's ignorance or bad faith (naive realism)?
Bias blind spot. The tendency to detect distortion more readily in others than in oneself. This is a derived signature of naive realism's self-objective/other-biased asymmetry, not the full disposition — the disposition generates it along with the hostile-media symmetry and the concession trap. Tell: is the observation specifically that one sees bias in others but not oneself (bias blind spot), or the whole generative posture from which that asymmetry follows (naive realism)?
Map/territory distinction & model-dependency (the parent it instantiates). The substrate-neutral failure in which an agent treats its own model of a domain as the domain itself, failing to model the model — which recurs observer-free in measurement, modeling, and perception. Naive realism is the social-cognitive instance that adds disagreeing minds and defect-attribution. Invoking "naive realism" for a solitary modeler mistaking model for world over-reaches into the parent's territory. Tell: is there a lone agent conflating representation with represented (the parent), or disagreeing parties each charging the other's dissent to bias (naive realism)? (Treated fully in earlier sections.)

Neighborhood in Abstraction Space¶

Naive Realism sits in a crowded region of the domain-specific corpus (39^th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.

Family — Rhetorical Framing & Loaded Discourse (24 abstractions)

Nearest neighbors

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