Conjunction fallacy

Home/All Biases/Conjunction fallacy

Cognitive Bias

Conjunction fallacy

The tendency to assume that specific conditions are more probable than a more general version of those same conditions

EstimationAssociation

What it distorts

Biases that distort numerical judgment, risk perception, calibration, and first-pass estimates.

Typical trigger

Situations where estimation is already difficult and the association cue feels easier to trust than a fuller review.

First countermove

Start with the estimation question instead of the first intuitive answer, then check whether the association pattern is doing invisible work.

Best use

Quick reference

Comparison Traps And Choice Architecture

Quick check

Am I calling the narrower story more probable just because it feels more representative?

Mechanism snapshot

In estimation problems, the mind overweights resemblance, vividness, proximity, or intuitive linkage before a fuller check catches up.

Teaching gauges

These are classroom-facing editorial estimates for comparing how the bias behaves in use. They are teaching aids, not measured statistics.

Common in probability talk

Especially active when stories are vivid and categories are abstract.

Rare Frequent

Easy to spot from outside

Clear once the set relation is named explicitly.

Hidden Obvious

Easy to innocently commit

Representativeness feels like a strong cue if the math is not foregrounded.

Low risk Easy slip

Teaching difficulty

One of the best biases for teaching the distance between coherence and probability.

Foundational Advanced

What's happening here.

This comparison makes the hidden pull easier to see before the technical label has to do all the work.

Biased move

This is like thinking a decorated branch is bigger than the tree because it is more interesting to look at.

Clearer comparison

Interesting stories can fit well and still belong inside a larger category that remains more probable. Specificity is not probability.

Caveat

Do not use this label every time someone tells a detailed story. The issue is the mathematical mistake of rating the specific conjunction above the broader set that contains it.

Use the label only when...

Use this label when narrative fit or representativeness makes the richer option feel more likely than the more general one.

How this entry is classified

Estimation: Biases here distort numerical judgment, probability, calibration, and first-pass estimation.
Association: The mind overweights resemblance, vividness, proximity, or intuitive linkage.

Reference use

Use the quick check, caveat, and nearby confusions together. The fastest diagnosis is often the noisiest one.

Bias in the wild

Each example changes the surface context while keeping the same hidden distortion in place.

Everyday life

A description of someone seems so specific that people judge the narrower story as more likely than the more general one that contains it.

Work and teams

A team treats a detailed failure chain as more probable than the simpler broader event of the project merely slipping.

Public discourse

Commentary gives extra probability credit to narratives that feel coherent and representative even when the added conditions should mathematically reduce likelihood.

What it feels like from inside

The richer, more vivid story feels more probable because it seems to fit the person or situation better, even though adding details should not make the event more likely than the broader category.

Teaching note: This page is especially useful for teaching that coherence and probability are not interchangeable, even when the coherent story feels smarter.

Telltale signs

Specificity is being mistaken for probability.
The more story-like option feels more likely mainly because it is easier to picture.
People resist the broader category because it feels too plain despite being more probable.

Repair at three levels

The strongest debiasing moves change the process, not just the label.

Solo move

Translate the story into set language and ask which option is the broader container.

Team move

Force probability judgments to be stated before the narrative embellishment is read aloud.

System move

Use forecasting templates that separate scenario vividness from numerical likelihood.

Practice And Repair

Follow the drift, then interrupt it

Conjunction fallacy is what happens when narrative fit outruns set logic.

Trigger

A specific story feels highly representative of a person or event.

Felt certainty

The richer description seems more likely because it seems more revealing.

Distortion

Specificity gets mistaken for probability, and the narrower event outranks the broader container.

Reset

Ask which option contains the other, then remember that the container cannot be less probable than one of its subsets.

Repair question

Which option is the superset here, and why am I letting the more vivid subset feel stronger?

Spot It

What number, rate, sample, or magnitude is being misread because the mind grabbed an easier proxy?
What feels connected here mainly because it is salient, familiar, or easy to pair mentally?
Compare the current interpretation against the brief source definition before treating the label as settled.

Similar biases and easy confusions

These are nearby labels that can share the same outer appearance while differing in what actually drives the distortion. Use the overlap, the distinction, and the diagnostic question together before settling the call.

Base-rate neglect

Why compare it: Base-rate neglect ignores the background prevalence; conjunction fallacy specifically violates the rule that adding conditions cannot increase probability.

Availability heuristic

Why compare it: Availability uses memorable examples; conjunction fallacy uses representativeness and vivid narrative fit to overrate the narrower story.

Attribute substitution

Why compare it: Attribute substitution can explain part of the mechanism when representativeness or vivid fit is used instead of actual probability.

Reflection questions

These are useful when the label seems roughly right but the process change still feels underspecified.

Does the added detail really add probability, or only vividness?

Which option contains the other as a subset?

Am I grading plausibility of the story or probability of the event?

Case studies

These sourced cases do not prove what was in someone's head with perfect certainty. They are teaching cases for showing where the bias pressure becomes visible in practice.

View related cases

The Linda problem

In the famous Linda problem, many people judge a more specific description of Linda as more probable than the broader one that logically contains it.

Why it fits: Representativeness and vivid fit defeat the simpler probability rule.

1983

Detailed suspect profiles outrank broader categories

In case-style reasoning, richly detailed descriptions can make a narrow conjunction feel more probable than the simpler broader category that contains it.

Why it fits: Coherence and narrative fit are doing the work that probability should be doing.

Modern probability research

Use it in context

These linked tools turn the page into practice instead of leaving it at the level of definition.

Learning paths

This bias appears directly in one guided sequence and also in nearby paths that frame the same judgment problem from a slightly wider angle.

Direct path

Comparison Traps And Choice Architecture

Use this path when you suspect the choice set itself is manufacturing preference rather than merely revealing it.

Same path family · Evidence and explanation

Start Here

Use this path when you want the minimum set of pages that gives the rest of the site immediate traction.

Assessment

These scenarios mix direct and nearby cases so you can practice the label itself and the broader judgment pattern around it.

Direct scenario

The richer story must be more likely

A committee judges a detailed and very representative-sounding failure chain as more probable than the simpler, broader event of the launch just going badly in some way.

Same scenario family · Evidence and explanation

Risk gets replaced by fear

A team discussing cyber risk keeps answering how dangerous a scenario feels instead of how probable it actually is, and the emotional force of the scenario quietly becomes the ver…

Start a focused assessment set

Companion reading

These links widen the frame around the bias without interrupting the core lesson on this page.

One-person stories and big-number cases pull judgment very differently

An article on why identifiable cases, vivid prototypes, and human-scale stories can overpower larger but more abstract evidence and need deliberate rebalancing.

CogBias theory

Related biases

These neighbors were selected from shared categories, shared patterns, and explicit editorial links where available.

Poster illustration for Aesthetic–usability effect