No personality type has a higher IQ than the others. There is no credible dataset ranking the 16 four-letter types by measured intelligence — the tables that circulate online come from small, self-selected samples that were never peer-reviewed. Type describes how you prefer to think. An IQ test measures how well you reason. Only one of those is a measurement, and it is not the one with four letters.

That answer is less satisfying than a ranking, so this page does the harder thing: it shows you exactly where the rankings came from, what the research genuinely establishes about personality and ability, and what your type does and doesn't tell you about your mind.

Where the "smartest type" rankings come from

Search for the smartest personality type and you will find confident, specific answers — and they disagree with each other. One list puts INTJ first. Another puts INTP first. A widely shared blog post argues for ENFP. A forum thread from 2012 still ranking on the first page of Google settles on INTJ; a Q&A answer directly beneath it settles on INTP.

That disagreement is the finding. If any of these rankings rested on real data, they would broadly agree with each other. Instead, nearly all of them trace back to the same handful of self-published data tables that have been passed around since the 2000s, reproduced without their original context, and treated as authoritative through sheer repetition.

Those tables share a specific set of problems:

  • No sampling frame. The respondents were whoever happened to take both tests in a particular setting — usually people who sought out a personality instrument, which is not a cross-section of anybody.
  • Cells far too small to rank. Splitting a modest sample across 16 types leaves very few people in each one — and the rarer types, the very ones these lists put at the top, get the smallest cells of all. Ranking 16 groups requires far more people per group than these tables ever had.
  • No peer review. None of the commonly cited tables appeared in a peer-reviewed journal, where the sample-size problem would have been caught immediately.
  • An unstable input. Four-letter type itself is not fully stable — a substantial share of people retest as a different type within weeks, particularly when they score near the middle of a dimension. Ranking a variable that moves is difficult even with good data.
  • Self-selection at both ends. The types that top these lists are also the types most drawn to discussing intelligence online. That inflates their presence in informal samples without saying anything about ability.

What the research actually shows

Serious work on personality and intelligence almost never uses four-letter types. It uses the Big Five traits, which are measured continuously rather than sorted into boxes. And there, one finding does hold up.

Openness to Experience shows a small positive correlation with measured intelligence — typically around r = .2 to .3 across meta-analyses, including Ackerman and Heggestad's 1997 review in Psychological Bulletin. The other four Big Five traits correlate close to zero with general reasoning ability.

A correlation of .2 to .3 is real, and it is also weak. It means Openness accounts for something in the region of 4–9% of the variation in test scores, leaving over 90% unexplained. It is enough to say something about groups. It is nowhere near enough to predict an individual's score, and it certainly cannot support a league table of 16 types.

The deeper reason is a distinction psychologists draw between two kinds of measurement. Personality instruments measure typical performance — what you tend to do. Ability tests measure maximal performance — the best you can manage when you are trying. A preference for abstract thinking makes someone more likely to enjoy a reasoning puzzle. It does not make them better at solving it. This is also why the publishers of four-letter type instruments state directly that their tools are not aptitude measures and should not be used to assess ability or select candidates.

The one letter with any connection at all

If you want the most that can honestly be said: the Intuition–Sensing (N/S) dimension overlaps conceptually with Openness to Experience — both describe a pull toward abstraction, patterns and possibility over concrete detail. So to whatever small extent any part of a four-letter type tracks measured ability, it is this letter and no other.

Note carefully what that does notmean. It does not mean an Intuitive person will out-score a Sensing one; the association is between broad traits across large groups, and the overlap between the two distributions is nearly total. It does explain a quirk of the folklore, though: every "smartest type" list is topped by N types, because the trait they share is the only one with even a faint statistical connection — and because N types write most of the lists.

What your type does tell you about how you think

Type is not useless here — it is just answering a different question. Below is all 16 types with the thinking style each one describes, and the kinds of reasoning that style suits and doesn't. There is deliberately no IQ column, because no honest one can be written. Every row spans the full ability range.

TypeThinking style it describesWhere that style helpsWhere it costs
ISTJThe AnchorSequential, evidence-first reasoningDetail accuracy, procedure, verificationOpen-ended problems with no precedent
ISFJThe GuardianConcrete recall anchored in people and contextPractical judgement, remembering specificsAbstract or deliberately artificial puzzles
INFJThe CounselorPattern-finding applied to people and meaningSynthesis, reading complex situationsProblems needing one narrow literal answer
INTJThe ArchitectLong-range analysis of systemsStructural reasoning, second-order consequencesSpeed under interruption; details deemed trivial
ISTPThe CraftsmanHands-on diagnosis of how things workMechanical and spatial reasoning, troubleshootingLong verbal or theoretical stretches
ISFPThe ComposerDirect, sensory, present-tense attentionPerceptual detail, aesthetic discriminationTimed abstract testing formats
INFPThe IdealistAssociative thinking organised around valuesVerbal fluency, original framingTasks rewarding speed over consideration
INTPThe LogicianAnalysis of abstract structure for its own sakeLogic, formal systems, spotting inconsistencyClosure — the analysis can outlast the deadline
ESTPThe DynamoFast reading of the immediate situationReal-time problem solving, improvisationSlow, solitary, abstract work
ESFPThe EntertainerImmediate engagement with people and environmentPractical social reasoning, quick adaptationSustained silent concentration
ENFPThe SparkWide-ranging idea generationDivergent thinking, unexpected connectionsConvergent tasks with one correct route
ENTPThe ChallengerArgument-testing and rapid reframingFinding flaws, generating alternativesFollowing a fixed method to the end
ESTJThe DirectorOrganising facts into workable orderExecution logic, prioritisationAmbiguity with no decidable answer
ESFJThe HostPractical reasoning about people and obligationsCoordination, applied social judgementImpersonal abstraction
ENFJThe MentorReading people and organising them around a goalPersuasive and verbal reasoningCold analysis that ignores the human cost
ENTJThe CommanderDecisive structuring toward an objectiveStrategic reasoning under time pressureReconsidering a call once it's made

Read the table as a description of preference, not capacity. An ISTJ who prefers sequential, evidence-first reasoning is not worse at abstract problems than an INTP — they are less drawn to them. Preference shapes which problems you choose to spend your life on, which over decades shapes what you get good at. That is a genuine and interesting effect, and it is not the same as a difference in ability.

How to actually find out

If the question underneath "which type is smartest" is really "how do I compare?", no type chart will ever answer it. A reasoning test will.

A properly scored test places you on the IQ scale relative to a representative sample, where 100 is the average by definition and about two-thirds of people fall between 85 and 115. That comparison is meaningful in a way a four-letter code never can be, because it comes from your answers to problems rather than your answers about yourself. It is worth understanding how IQ tests work and how accurate they actually are before you read too much into any single number — online tests are useful indicators, not clinical instruments.

The IQ Revealed assessment is 40 matrix-reasoning questions and takes about 15 minutes. It is free to take; membership unlocks your full report with your score, percentile and cognitive profile. If you want the other half of the picture, the 16 type profilesdescribe the preference side — the part your IQ score genuinely doesn't cover.

IQ Revealed is not affiliated with, endorsed by or connected to the Myers-Briggs Company, NERIS Analytics or any other personality-test publisher. The four-letter codes are widely used descriptive labels; the type descriptions and analysis on this page are our own original work.