Persona¶
Core Idea¶
A persona is a synthesized, named, concrete archetype that stands in for an unwieldy population during reasoning under uncertainty about that population. A designer, strategist, teacher, policy author, or security analyst cannot reason directly about millions of heterogeneous users, students, citizens, or attackers — the distribution is too broad and too sparsely known to drive concrete decisions. So they construct a small set of representative composite individuals — Sarah, 34, single mother, mobile-only, mid-trust in institutions — and reason about Sarah making a choice rather than about the population making it. The persona is a deliberate reduction in dimensionality: from a population to a handful of named exemplars chosen to span exactly the parts of the distribution to which the decision is most sensitive.
The essential commitment is that the archetype must be more concrete than the population is — and more concrete even than an "average user," which is a statistical artifact that no real individual instantiates. The persona is a fictional individual constructed to be plausible, internally consistent, and cognitively manipulable: a stand-in the reasoner can put questions to. The trade is coverage for traction. The persona is only a few points in a high-dimensional space, so it covers the population poorly, but it gives the reasoner concrete purchase on needs, motivations, frictions, and trade-offs in a way that no statistical summary supports. The structural skeleton has five parts: a target population, real, broad, heterogeneous, and partially unknown; a parameter selection, naming which dimensions of variation matter for this decision; a persona construction, one or a small set of composite individuals named and given enough texture to be reasoned about; a reasoning-via-archetype discipline, framing each decision as "what does this persona need, want, do, or fail at" rather than as a population question; and a re-grounding step, in which the conclusions reached through the persona are checked back against the actual population before they drift into free-floating fiction. That last step is load-bearing: a persona that cannot be falsified by research about the real population has decayed into a stereotype.
How would you explain it like I'm…
The Make-Believe Kid
Stand-In Character
Composite Archetype Stand-In
Structural Signature¶
a broad, heterogeneous, partially-unknown target population — a selection of the dimensions to which the decision is sensitive — a small set of named, concrete composite archetypes — a reasoning-via-archetype discipline — a re-grounding check against the actual population — a coverage-for-traction trade-off
The pattern is present when each of the following holds:
- A target population. A real, broad, heterogeneous body — users, learners, citizens, attackers, patients — too large and sparsely-known to reason about directly under uncertainty.
- A sensitivity-dimension selection. A naming of which dimensions of variation actually matter for the decision at hand, so the archetype spans exactly those axes.
- A concrete composite archetype. One or a few fictional individuals, named and given enough texture to be cognitively manipulable — more concrete than the population, and more concrete even than its statistical average.
- A reasoning-via-archetype discipline. Each decision is framed as "what does this archetype need, want, do, or fail at" rather than as a population question, converting an intractable distribution into a queryable stand-in.
- A re-grounding invariant. Conclusions reached through the archetype must be checked back against the real population; an archetype that cannot be falsified by research about the population has decayed into a stereotype. This step is load-bearing.
- A coverage-for-traction trade. The archetype covers the population poorly (few points in a high-dimensional space) in exchange for concrete cognitive purchase the statistics cannot give.
These compose into a deliberate, lossy dimensionality reduction whose validity depends on the stand-in remaining falsifiable against the population it represents.
What It Is Not¶
- Not a representative sample. See
sampling_representativeness. A representative sample is drawn so its statistics mirror the population's; a persona is a constructed fictional individual chosen for cognitive traction, deliberately more concrete than any real or average member. One optimizes for unbiased coverage; the other sacrifices coverage for queryability. - Not a stereotype. See
stereotyping. A stereotype is a fixed, unfalsifiable generalization applied to a group. A persona is built to be re-grounded — checked against and revised by research on the real population. A persona that loses falsifiability has decayed into a stereotype; the re-grounding invariant is exactly what separates them. - Not a mental model. See
mental_model. A mental model is the reasoner's internal representation of how a system works. A persona is an externalized, named artifact representing a population of people, deliberately shared and contestable rather than private. - Not prototype theory. See
prototype_theory. Prototype theory describes how the mind naturally organizes categories around central exemplars (a robin as prototypical bird). A persona is a deliberately engineered design instrument, constructed and re-grounded on purpose, not an automatic feature of cognition. - Not user-centered design. See
user_centered_design. That is the broad methodology of grounding design in user needs; the persona is one instrument within it — the named composite stand-in — not the whole practice. - Common misclassification. Treating a persona as a factual claim about a typical user. The catch: ask whether it can be falsified by population research. If the archetype is never re-grounded and merely encodes the team's assumptions (often resembling the team itself — the I-methodology trap), it is a stereotype wearing a persona's name.
Broad Use¶
- Human-computer interaction and product design. User personas drive interface and feature decisions; mock-ups are tested against a small persona set rather than against an abstract "everyone."
- Marketing. Buyer personas and ideal-customer profiles shape segmentation, messaging, and channel choice, letting a team write copy concretely for a named decision-maker.
- Pedagogy. Learner personas — the novice with no background, the second-career learner with rusty prerequisites, the heritage speaker with high oral comprehension but low literacy — surface scaffolds that an "average learner" assumption hides.
- Public-service design. Citizen personas (the rural elderly user, the recent immigrant, the shift worker) surface access friction that a median-citizen assumption conceals.
- Security engineering. Threat-actor personas — the script-kid, the financially motivated criminal, the nation-state actor, the malicious insider — let defenders reason concretely about attacker capability, motivation, and likely attack paths.
- Clinical practice and medical education. Patient archetypes (the uncomplicated case, the comorbidity-laden elderly patient, the medication-noncompliant patient) ensure a protocol works for each, not just for an average patient.
- Strategic and adversary planning. Adversary personas reason about specific decision-makers with known patterns, risk tolerance, and domestic constraints rather than about a featureless monolith.
Clarity¶
Naming the persona move surfaces a specific cognitive cheat that reasoners otherwise smuggle in unmarked: they reason about some imagined user without naming who that user is, and the imagined user almost always resembles the reasoner — the I-methodology failure, in which a designer unconsciously designs for themselves. Promoting the imagined user to an explicit, named persona makes the assumption inspectable, and the inspecting question — whose decision have we actually been optimizing for? — can be asked equally of marketing copy, a curriculum, a policy form, or a security control. Many failures that present as execution failures are revealed, once the persona is made explicit, as persona-mismatch failures: the artifact was quietly built for someone other than the population it was meant to serve. The clarity also runs negatively, and this is where good persona work earns its keep. Making the persona set explicit makes visible the persona's shadow — the populations the set does not cover and that the design will therefore quietly underserve. A persona set that names its own blind spots is doing the clarifying work; one that pretends to cover everyone has hidden its exclusions rather than removed them. The structural payoff of naming is that an implicit, unaccountable choice of whom to serve becomes an explicit, contestable one.
Manages Complexity¶
The persona compresses a high-dimensional, sparsely-known population into a handful of named, internally consistent exemplars. The compression is deliberate and lossy: the persona is easier to reason about than the population precisely because it has been stripped of variation. Several disciplines manage the resulting trade-off. Persona-set size governs the coverage-traction balance — more personas mean more coverage but less cognitive grip, and standard practice settles on a few, enough to span the sensitivity dimensions while remaining holdable in working memory at once. Persona texture governs realism without over-specification — enough biographical and motivational detail to be reasoned about, but not so much that incidental detail hardens into a constraint masquerading as data. Anti-personas name explicitly whom the design is not for, converting an unspoken exclusion into an accepted, accountable decision. And a re-grounding cadence of periodic checks against actual research detects drift from the population the persona is supposed to represent. The complexity-management is asymmetric in cost. Building the persona set is moderately expensive — it takes research, synthesis, and calibration — but once built, every subsequent decision becomes cheaper, because the otherwise paralyzing population question has been pre-collapsed into a tractable form that can be queried directly. The expensive analysis is paid once and amortized across many downstream decisions.
Abstract Reasoning¶
The persona structure supports reasoning about the three-way distinction between population, archetype, and individual, each with a different epistemic status: the population is real but unknowable in detail, the persona is fictional but cognitively manipulable, and the individual is real and knowable but not generalizable. Sound persona work navigates among the three rather than collapsing them. It supports reasoning about the I-methodology trap — in the absence of explicit personas, reasoners default to themselves — and forces the diagnostic question of whether the persona secretly is the reasoner. It supports reasoning about coverage versus traction as a tunable trade-off with a sweet spot rather than a fixed quantity. It supports reasoning about persona drift, the failure in which a persona never re-grounded against its population becomes a free-floating fiction that the team begins to optimize against directly, with no remaining feedback from reality. It supports reasoning about negative-space coverage through explicit anti-personas that name the populations the design will fail. And it supports recognizing the symmetry between designing-for and designing-against: the same construction discipline that powers user personas powers threat-actor personas, and the adversarial case inherits the identical risks — drift, I-methodology, blind spots — and the identical benefits of concreteness and sensitivity-dimension naming. That symmetry is part of what marks the move as structural rather than merely a UI convention: the cognitive instrument is indifferent to whether the modeled population is one the reasoner serves or one the reasoner opposes.
Knowledge Transfer¶
The persona's construction discipline transfers across substrates as a single reusable procedure — interview or observe, synthesize, name, texture, drive decisions, re-ground — and the cleanest demonstration of its substrate-independence is its symmetric transfer from "the person we are designing for" to "the person we are designing against." Threat-actor archetypes in security engineering are persona work done in the adversarial register: name the actor, name their motivation, name their capability, and design defenses that hold up against each named actor in turn, with the same anti-patterns (the threat actor that suspiciously resembles the security team's own preconceptions) and the same corrective (re-grounding against actual incident data). The HCI discipline of user research synthesized into named user personas translates directly into the instructional-design discipline of learner research synthesized into named learner archetypes, carrying both the shared anti-pattern — designing for the designer rather than the learner — and the shared fix of an explicit named archetype textured along the dimensions to which the lesson is sensitive. Fiction's accumulated craft of constructing internally consistent, manipulable individuals transfers to persona design without modification, because the persona that "feels real" is doing the same cognitive work as a well-drawn character; the design literature traces the persona move's origin explicitly to dramaturgy.
The role mappings carry across every substrate. The target population is the user base, the market, the cohort, the citizenry, the threat landscape, the patient population. The sensitivity dimensions are the capability, motivation, context, and risk tolerance that the particular decision turns on. The composite archetype is Sarah the mobile-only filer, the CTO under board pressure, the second-career learner, the rural elderly citizen, the nation-state actor, the comorbidity-laden patient. The anti-persona is the tax professional explicitly routed elsewhere, the market segment explicitly not served, the attacker explicitly out of scope. Because the skeleton is shared, the failure modes and their defenses transfer as a family: the I-methodology trap, persona drift, and collapse-into-stereotype recur in every substrate, and the same defenses — explicit construction, intersectional persona sets, anti-personas, and a re-grounding cadence — answer them in every substrate. The shift from "the form must be usable" to "the form must work for Aisha, who is bilingual, mobile-only, and has filed once before" is the same move whether the artifact is a government form, a marketing campaign, a curriculum, or a security architecture, and it surfaces the same class of exclusion failures the abstract framing hides. What transfers is not a template to be filled in but a discipline of reasoning: construct a falsifiable stand-in, reason through it concretely, and keep it honest by repeatedly testing it against the population it claims to represent.
Examples¶
Formal/abstract¶
Threat modeling in security engineering is the persona move executed in the adversarial register, and it shows the structure operating where its "designing-against" symmetry is sharpest. The target population is the threat landscape — every entity that might attack a system, broad, heterogeneous, and only partially knowable. Reasoning about "attackers in general" yields no concrete defenses, so the analyst performs a sensitivity-dimension selection: the dimensions that actually drive defensive decisions are the attacker's motivation, capability, resources, and risk tolerance. From these, the analyst constructs a small set of concrete composite archetypes: the script-kiddie running off-the-shelf tools with low skill and low patience; the financially-motivated criminal who will invest effort proportional to payoff; the nation-state actor with deep resources, long time horizons, and tolerance for burning expensive zero-days; the malicious insider with legitimate access but limited stealth. The reasoning-via-archetype discipline then reframes each design question as "how would this actor approach this surface" — the nation-state persona justifies defense-in-depth against zero-days that would be irrational against the script-kiddie. The re-grounding invariant is load-bearing exactly as in user personas: the threat-actor set must be checked against actual incident data and threat intelligence, or it decays into a stereotype that secretly resembles the security team's own preconceptions (the adversarial form of the I-methodology trap). The coverage-for-traction trade is explicit — four archetypes cannot enumerate every adversary, but they give concrete purchase that an undifferentiated "hacker" never could, and the populations they fail to cover are the anti-personas the model must name.
Mapped back: The threat landscape is the target population, attacker motivation and capability are the sensitivity dimensions, the script-kiddie and nation-state actor are the concrete archetypes, and checking against incident data is the re-grounding step that keeps the model from collapsing into stereotype — the persona structure operating in reverse polarity.
Applied/industry¶
In public-service design, citizen personas surface access failures that a "median citizen" assumption structurally conceals. Consider a government agency redesigning an online benefits-application form. The target population is the entire eligible citizenry — vast, heterogeneous, and, crucially, skewed toward exactly the people least like the civil servants and contractors building the form. Designing for an imagined average user triggers the I-methodology trap: the team builds for someone with a stable home computer, a reliable connection, fluent literacy, and high institutional trust, because that describes the team. The persona discipline intervenes through sensitivity-dimension selection — for a benefits form, the decisive dimensions are device access, connectivity, literacy and language, and trust in government. The team constructs concrete archetypes spanning those axes: Aisha, mobile-only, bilingual, who has filed once before and abandoned the process at the document-upload step; an elderly rural applicant on intermittent connectivity with low digital literacy; a shift worker who can only file at 2 a.m. and needs the session to persist. Reasoning-via-archetype then exposes concrete frictions — Aisha's abandonment reveals that the upload step assumes a desktop file system she does not have, prescribing a mobile-camera capture path. The re-grounding invariant is satisfied by usability research with actual applicants matching each archetype, and anti-personas (the tax professional, explicitly routed to a different channel) name whom the form is deliberately not for, converting a silent exclusion into an accountable decision. The same construction discipline — interview, synthesize, name, texture, drive decisions, re-ground — runs identically in product design (user personas), marketing (buyer personas), and pedagogy (learner archetypes), which is why a designer fluent in one can ask the load-bearing question in any of them: whose decision have we actually been optimizing for?
Mapped back: The citizenry is the target population, device and literacy and trust are the sensitivity dimensions, Aisha is the concrete archetype that makes the upload friction visible, usability testing is the re-grounding check, and the routed-elsewhere tax professional is the anti-persona — the same lossy-but-tractable dimensionality reduction that drives user, buyer, and learner personas.
Structural Tensions¶
T1 — Coverage versus Traction (the Core Trade). The persona buys cognitive grip by sacrificing coverage — a handful of named points cannot represent a high-dimensional population. More personas extend coverage but dilute traction; fewer concentrate traction but widen the blind spots. The failure mode is treating the persona set as if it covered the population, optimizing confidently for the named few while the unnamed many are silently underserved. The diagnostic is to ask what fraction of the real distribution sits outside every persona's neighborhood: a set that cannot name its own uncovered regions has hidden its exclusions rather than removed them. The competing move — full statistical representation — gives coverage but no traction, which is exactly why the persona exists.
T2 — Concrete Archetype versus Real Individual (Generalizability Boundary). The persona must be more concrete than the population, yet it is fictional; a real, knowable individual is concrete too but does not generalize. These are three distinct epistemic statuses — population (real, unknowable in detail), persona (fictional, manipulable), individual (real, knowable, non-generalizable) — easily collapsed. The failure mode is promoting an actual interviewed user to stand for the population, importing their idiosyncrasies as if representative. The diagnostic is to ask whether each persona trait was selected for its coverage of a sensitivity dimension or merely inherited from one vivid informant: traits that trace to a single real person, not to the distribution, are individual noise masquerading as archetype.
T3 — Persona versus Reasoner (the I-Methodology Trap). Absent an explicit persona, reasoners default to themselves — the imagined user silently becomes the designer. The persona's whole point is to displace that default, but a carelessly built persona can simply re-encode the reasoner's own profile under a fictional name. The failure mode is a persona that suspiciously resembles the team, producing artifacts that work for the builders and fail for the served population — the access-friction failures a median-user assumption conceals. The diagnostic is to ask, of each persona, how it differs from the people in the room: a persona that shares the team's device access, literacy, and institutional trust is the reasoner in disguise, not a corrective to them.
T4 — Construction Cost versus Re-Grounding Cost (Temporal Drift). Building the persona set is paid once and amortized across many downstream decisions — but the cheapness of reuse is exactly what tempts teams to keep querying a stale persona indefinitely. Re-grounding is a recurring cost the amortization model quietly omits. The failure mode is persona drift: a persona never re-checked against the population becomes a free-floating fiction the team optimizes against directly, with no remaining feedback from reality, decaying into stereotype. The diagnostic is to date the last re-grounding: a persona driving decisions on research older than the population's rate of change has lost falsifiability, which is the property that separated it from a stereotype in the first place.
T5 — Designing-For versus Designing-Against (Polarity Symmetry). The same construction discipline powers user personas and threat-actor personas, but the adversarial register inverts the stakes of error. For a user persona, a blind spot underserves someone; for a threat persona, a blind spot is an unguarded attack path an adversary actively seeks. The failure mode is carrying for-persona habits into the against-case unchanged — assuming the modeled adversary is static and cooperative rather than adaptive and hostile, so the persona becomes the team's preconception of an attacker rather than the attacker. The diagnostic is to ask whether the modeled population would change its behavior in response to the design: a served population mostly will not; an adversary will, so the against-persona must anticipate its own evasion.
T6 — Texture versus Over-Specification (Detail Calibration). A persona needs enough texture to be cognitively manipulable, but incidental biographical detail can harden into a constraint masquerading as data — "Sarah drives a sedan" quietly becomes a design assumption no research supports. There is a sweet spot between a persona too thin to reason about and one so richly specified that its decorative details are mistaken for findings. The failure mode is reasoning from texture that was invented for plausibility as though it carried evidential weight. The diagnostic is to separate, for each persona attribute, what was measured from what was added for realism: only the sensitivity-dimension traits are load-bearing, and a decision that turns on a decorative detail is resting on fiction.
Structural–Framed Character¶
Persona sits on the framed side of the structural–framed spectrum — a framed prime with a 0.5 aggregate sitting right at the boundary, framed because the single diagnostic that maxes out is decisive. Underneath there is a genuine relational move — reduce an unwieldy high-dimensional population to a small set of named, concrete exemplars chosen to span the axes a decision is sensitive to, then re-ground the conclusions against the real population — and that lossy dimensionality reduction is structurally clean. But the prime is constitutively a cognitive aid built by and for people, and that pins it to the framed side.
The decisive diagnostic is human_practice_bound, which reads a full 1. A persona is not a pattern running in an indifferent physical or biological substrate; it exists only as a deliberate construction that a designer, strategist, teacher, policy author, or analyst makes in order to reason — every one of its canonical homes (HCI, marketing, pedagogy, policy, threat modeling, clinical archetypes, military planning) is a human-design context, and there is no version of "Sarah, 34, mobile-only" that instantiates itself without a reasoner building it. That single fact is what the rationale calls out: the practice is the carrier. The remaining diagnostics sit at the middle or below. Vocab_travels is 0.5 because the home lexicon (persona, archetype, user) translates with only mild friction into "threat actor," "buyer profile," or "representative learner," so the vocabulary travels but carries a faint design-practice residue. Institutional_origin is 0.5 because the prime was codified in user-centered-design practice, a real but soft institutional origin rather than a hard doctrinal one. Import_vs_recognize is 0.5 because invoking a persona half-imports a design methodology (construct, name, re-ground) and half-recognizes the underlying stand-in-for-a-population move. And evaluative_weight is flatly 0 — a persona is neither good nor bad until you say what decision it serves; a stereotype is just a persona that failed its re-grounding check, a structural failure mode, not an inherent evaluative charge. The genuine relational skeleton is real, but because the pattern cannot exist outside a human reasoning practice, the framed label with its 0.5 boundary aggregate is the faithful placement.
Substrate Independence¶
Persona is a moderately substrate-independent prime — composite 3 / 5 on the substrate-independence scale. Its breadth is genuinely good: the named-concrete-archetype-standing-in-for-a-population move recurs across HCI and product design, marketing, pedagogy, public-service and policy design, security threat-modeling, clinical and medical-education archetypes, and adversary planning — and its symmetric transfer from "the person we design for" to "the person we design against" is striking evidence of structural portability. Transfer within that band is concrete and documented: the construction discipline (interview, synthesize, name, texture, drive decisions, re-ground) carries intact across substrates with a stable set of role mappings (target population, sensitivity dimensions, composite archetype, anti-persona) and a shared family of failure modes (I-methodology, persona drift, collapse into stereotype). What pins it to the middle, and caps structural abstraction at 3, is that every one of those substrates is a human-design context: a persona is constitutively a cognitive aid built by and for a reasoner — there is no version of "Sarah, 34, mobile-only" that instantiates itself in an indifferent physical or biological substrate, and the practice, not just the vocabulary, is the carrier. The strong, formally-disciplined cross-domain transfer holds it at a 3; the human-practice ceiling keeps it from climbing.
- Composite substrate independence — 3 / 5
- Domain breadth — 4 / 5
- Structural abstraction — 3 / 5
- Transfer evidence — 4 / 5
Relationships to Other Primes¶
Parents (1) — more general patterns this builds on
-
Persona is part of, typical User-Centered Design
The file: the persona is 'ONE instrument WITHIN' user-centered design (the broad methodology of grounding design in user needs) — the named composite stand-in, not the whole practice. Part-of, not is-a.
Path to root: Persona → User-Centered Design
Neighborhood in Abstraction Space¶
Persona sits in a sparse region of abstraction space (99th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely rather than landing on a neighbor.
Family — Representation & Mental Models (7 primes)
Nearest neighbors
- Sampling (Representativeness) — 0.67
- Shared Mental Model — 0.65
- Mental Model — 0.65
- Archetype — 0.65
- Population Coding — 0.64
Computed from structural-signature embeddings · 2026-06-14
Not to Be Confused With¶
The nearest confusion is with sampling_representativeness, the embedding-nearest neighbor, because both are strategies for letting a small, tractable object stand in for an unwieldy population. The structural commitments are opposite. Representative sampling aims for statistical fidelity: a sample is good to the degree its distributions of the relevant variables match the population's, so that estimates computed on the sample are unbiased for the whole. Its load-bearing invariant is coverage without distortion — every subgroup represented in proportion, no systematic skew. A persona makes the reverse bet. It deliberately abandons coverage — a handful of named points cannot mirror a high-dimensional distribution — in exchange for cognitive traction, a stand-in concrete enough to put questions to. A representative sample is precisely what a persona is not: it is many, anonymous, and proportioned; a persona is few, named, and textured. What sampling captures that personas cannot is quantitative generalizability — you can compute an estimate and bound its error. What personas capture that sampling cannot is concrete reasoning purchase — you can ask "what does Aisha do at the upload step?" in a way no proportioned sample supports. The two are complementary instruments, not rivals: sampling tells you how much of the population looks a certain way; the persona lets you reason concretely about what that looking-a-certain-way implies for a decision.
The most dangerous confusion is with stereotyping, because a persona that goes wrong becomes a stereotype, and the two can be indistinguishable in a snapshot. Both are concrete, generalized, named images of a kind of person. The distinguishing invariant is falsifiability against the population. A persona is constructed from research and, crucially, kept honest by a re-grounding discipline: it is periodically tested against fresh data about the real population and revised when it is wrong, so it remains a falsifiable hypothesis about whom the design serves. A stereotype is a fixed generalization that is not answerable to evidence — it persists regardless of disconfirming cases and often encodes the holder's preconceptions rather than the population's reality. The persona's tensions name exactly the slide between them: persona drift (T4), in which an un-re-grounded persona becomes a free-floating fiction, and the I-methodology trap (T3), in which the persona silently re-encodes the team itself. The structural claim is sharp — a persona that cannot be falsified by population research has already decayed into a stereotype. The difference is not in the artifact at any instant but in whether a living feedback loop to reality is attached to it.
A third confusion worth drawing is with prototype_theory. Both center on a concrete exemplar standing for a category, and both exploit the cognitive fact that people reason more fluently about a vivid central case than about a definition or a distribution. But prototype theory is a descriptive account of how human categorization spontaneously works — categories have graded structure organized around natural prototypes (a robin is a better bird than a penguin), and this happens automatically, below deliberation. A persona is a prescriptive, engineered instrument — deliberately built, named, textured, and re-grounded by a design team for the purpose of driving decisions. Prototype theory explains why personas work as cognitive aids (the mind is already prototype-organized, so a well-drawn persona slots into native machinery), but the persona adds the entire discipline prototype theory lacks: explicit sensitivity-dimension selection, anti-personas naming the uncovered, and a re-grounding cadence. Mistaking the persona for a mere prototype loses the engineering discipline; mistaking the prototype for a persona over-credits an automatic cognitive bias with the falsifiability a persona must earn.
For a practitioner, the distinctions assign different tools to different jobs. When the question is how much of the population has a property, reach for a representative sample, not a persona. When you need concrete purchase for a design decision, build a persona — and attach the re-grounding loop that keeps it from sliding into a stereotype. And remember that the persona borrows its cognitive power from the mind's native prototype machinery, which is exactly why it must be disciplined: the same fluency that makes a persona useful makes an un-grounded one a comfortable, persuasive falsehood.
Solution Archetypes¶
No catalogued solution archetypes reference this prime yet.