Questions: Conjunction Fallacy and Probability Judgment Errors

Question 1

Linda is described as a former philosophy student, politically active, and deeply concerned with social justice. Participants rate whether Linda is more likely to be 'a bank teller' or 'a feminist bank teller.' Why do most people rate the conjunction as more probable?

• A. They correctly apply Bayesian reasoning — the background description raises the prior probability of feminist beliefs
• B. They substitute representativeness (how well the description matches the category) for probability
• C. They assume 'feminist bank teller' is a more common job category than 'bank teller'
• D. They misread the question as asking which description is more coherent, not more probable

Answer: (B)

Participants are not computing probabilities — they are evaluating how well the description *fits* each category. The feminist bank teller is more representative of Linda's description, so it feels more probable. This is heuristic substitution: the hard question (what is the probability?) is replaced by the easier one (how good is the match?). Option A is the key misconception — even if feminist beliefs are likely, the conjunction P(bank teller AND feminist) can never exceed P(bank teller), because every feminist bank teller is also a bank teller.

Question 2

A description says Alex is highly analytical, loves puzzles, and has a PhD in mathematics. Participants are asked: is Alex more likely to be 'a software engineer' or 'a software engineer who volunteers at a food bank'? Which outcome does the conjunction fallacy predict?

• A. Participants rate 'software engineer' as more probable, correctly applying the subset rule
• B. Participants rate the conjunction as equally probable, because both include software engineering
• C. Participants rate the conjunction as more probable because the added detail forms a coherent narrative
• D. Participants avoid rating the conjunction because the description does not mention charity

Answer: (C)

The conjunction fallacy predicts that the added detail ('who volunteers at a food bank') is unlikely to feel like it *reduces* probability — it will either feel irrelevant or, if it makes a vivid coherent story, may even raise the perceived probability. The mathematically correct answer is that P(A and B) ≤ P(A), but the narrative pull makes the conjunction feel plausible. Note that Option B ('equally probable') is also wrong — the conjunction must be strictly less probable unless the two events are perfectly correlated.

Question 3

Adding more vivid details to a description makes the described conjunction more probable.

• T. True
• F. False

Answer: False

This is the exact confusion the conjunction fallacy exploits. Mathematically, adding conditions can only keep probability the same or reduce it — every additional detail narrows the set of outcomes that satisfy all conditions simultaneously. However, vivid details increase *representativeness* and narrative coherence, which feels like higher probability. The fallacy is precisely this divergence: intuitive plausibility rises with detail; mathematical probability falls.

Question 4

The conjunction fallacy is primarily a problem for people with no statistical training; researchers and statisticians who know the probability axioms consistently avoid it.

• T. True
• F. False

Answer: False

One of the most important findings in the conjunction fallacy literature is that it persists even in trained statisticians and probability researchers when scenarios are presented in realistic narrative form. The fallacy is driven by automatic representativeness heuristic substitution, which is not switched off by formal knowledge of probability rules. Explicit frequency formats (e.g., 'out of 100 people like this...') reduce the fallacy, but vivid narrative presentation elicits it even in experts — which is why this bias has such broad practical implications.

Question 5

Why does adding vivid, coherent details to a description make a conjunction feel more probable, even though mathematically it can only make it less probable or equal?

Model answer: Because people substitute representativeness — how well the description fits the category — for probability. A richer description creates a more coherent narrative match with the conjunction, making it feel like a better fit. But representativeness and probability are fundamentally different quantities: more details increase perceived coherence and fit while simultaneously constraining the set of people who could satisfy all conditions at once. The mind is organized as a narrative pattern-matcher, not a frequency tracker, so coherence and fit dominate probability calculation.

The deep lesson is that human cognition evaluates scenarios for story quality and category fit, not for base-rate compliance. This is why adding a plausible biographical detail to a conjunction makes it feel more probable rather than less — the detail 'explains' the person. Understanding this divergence between narrative coherence and statistical probability is the key insight that extends to legal reasoning, political persuasion, and con artistry, all of which exploit vivid detail to manufacture credibility.