Why Coherent Stories Blind Us to Logic

Probability is a branch of mathematics that governs the rules of risk, statistics, and logical decision-making. However, human beings did not evolve to calculate mathematical probabilities; we evolved to interpret reality through narratives and stories.

In Thinking, Fast and Slow, Daniel Kahneman demonstrates that when a story is highly descriptive and coherent, our minds will choose it over basic mathematical logic. This specific cognitive error is known as The Conjunction Fallacy, famously illustrated by an experiment known simply as The Linda Problem.

This article analyzes the cognitive architecture behind this fallacy and explains why descriptive prose easily blinds us to mathematical truth.

1. The Anatomy of the Linda Problem

In one of the most controversial and widely debated experiments in cognitive science, Daniel Kahneman and Amos Tversky presented participants with a psychological profile of a fictional woman named Linda:

“Linda is 31 years old, single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and also participated in anti-nuclear demonstrations.”

After reading this short biography, participants were asked to choose which of the following two scenarios was more statistically probable:

• Scenario A: Linda is a bank teller.
• Scenario B: Linda is a bank teller and is active in the feminist movement.

+-----------------------------------------------------------------+
|                    THE MATHEMATICAL REALITY                     |
|                                                                 |
|   All Bank Tellers (Scenario A)                                 |
|   +---------------------------------------------------------+   |
|   |                                                         |   |
|   |   Feminist Bank Tellers (Scenario B)                    |   |
|   |   [XXXXXXXXXXXXXXXXXXXXXXXX]                            |   |
|   |                                                         |   |
|   +---------------------------------------------------------+   |
|                                                                 |
|   Mathematical Law: Scenario A must ALWAYS be larger than B.   |
+-----------------------------------------------------------------+

When this experiment was run across elite university students—including individuals with extensive training in probability and statistics—an overwhelming 85% to 90% of participants chose Scenario B as more probable.

2. The Mathematical Law of Conjunction

To understand why this choice is a major cognitive fallacy, one must review the basic set theory of probability. The probability of two events occurring together (in conjunction) is always less than or equal to the probability of either event occurring alone.

Mathematically, this relationship is expressed as:

$$P(A \cap B) \le P(A)$$

For Scenario B to be true, Linda must satisfy two separate criteria simultaneously: she must belong to the group of all bank tellers, and she must belong to the group of all active feminists.

Therefore, the population of “feminist bank tellers” is a strict sub-set of the population of “all bank tellers.” It is a mathematical impossibility for a sub-set to be larger than the overarching set to which it belongs. Linda is infinitely more likely to be a bank teller (which includes every type of bank teller on Earth) than a feminist bank teller.

3. Why the Brain Fails: Representativeness vs. Probability

Why do highly educated minds consistently fail this simple logical test? The answer lies in System 1’s reliance on the Representativeness Heuristic.

System 1 does not calculate mathematical probabilities; it measures similarity and matches prototypes. When reading Linda’s profile, the details (philosophy major, social justice advocate) perfectly match our mental prototype of an active feminist. The description of a simple bank teller does not match that prototype at all.

[ Read Profile ] ---> [ Match to Prototype (Feminist) ] ---> [ System 1 Substitutes Probability for Similarity ]

When evaluating the choices, System 1 seamlessly substitutes the difficult question (“Which is mathematically more probable?”) with an easier question (“Which scenario looks more like the description of Linda?”).

Because Scenario B fits her description beautifully, the brain experiences a wave of cognitive ease and flags it as correct, completely blinding System 2 to the obvious logical contradiction starring it in the face.

4. Coherence vs. Plausibility: The Danger of Detailed Scenarios

The Linda Problem reveals a profound vulnerability in human forecasting, legal judgments, and corporate risk management: adding specific, highly realistic details to a scenario makes it sound more persuasive, but mathematically makes it less likely to happen.

Consider these two geopolitical predictions:

1. “A massive war will break out in Western Europe next year.”
2. “A massive war will break out in Western Europe next year, triggered by a sudden cyberattack on the electrical grid of a major nation.”

To a human jury, policy board, or corporate leadership team, Scenario 2 sounds far more believable, coherent, and realistic. The specific details paint a clear picture that creates cognitive ease.

In reality, however, Scenario 2 is significantly less probable than Scenario 1, because it requires two independent events to happen in a specific sequence.

5. Conclusion

The Conjunction Fallacy demonstrates that a good story will almost always defeat abstract logic inside the human mind. We are narrative-driven creatures who value coherence over statistical facts.

To protect your investments, business strategies, and long-term decisions from this bias, you must systematically separate stories from data. When reviewing any plan or prediction, remember to strip away the descriptive adjectives and evaluate the baseline probabilities of the individual events on their own.