Confirmation Bias — Why Your Brain Filters Out the Truth

## The Basic Claim

Confirmation bias is the tendency to search for, interpret, and recall information in a way that confirms your pre-existing beliefs. Everyone knows this. What people underestimate is how deep the mechanism runs.

It's not just that you notice confirming evidence more. It's that you generate different *quality* questions depending on whether you expect a "yes" or "no" answer.

## The Wason Selection Task

One of the most replicated demonstrations in psychology. Participants are shown four cards:

> A, D, 4, 7

They're told each card has a letter on one side and a number on the other, and given the rule: *"If a card has a vowel on one side, it has an even number on the other."*

Which cards should you flip to test the rule?

Most people say **A** and **4**. The correct answer is **A** and **7**.

Flipping 4 can't falsify the rule — even if there's a consonant on the other side, that doesn't break the rule (the rule only talks about vowels). Flipping 7 can break it, if there's a vowel on the other side.

People test for confirmation, not falsification. This is the default mental posture.

## Why It's Adaptive

Confirmation bias isn't a glitch. It's a feature that became a problem.

In most low-stakes daily decisions, finding confirming evidence quickly is efficient. The bias becomes costly in high-stakes, information-rich environments — investing, medicine, politics, science — where disconfirming evidence carries more diagnostic value than confirming evidence.

## The Hypothesis-Generation Loop

Researchers Snyder and Swann (1978) showed that when people are asked to assess whether someone is an introvert, they generate different interview questions than when assessing extroversion — for the same target person.

If you're looking for introversion, you ask "Do you find large parties draining?" If you're looking for extroversion, you ask "What do you do to liven up a dull party?" Both questions are real, but designed to confirm, not test.

## Motivated Reasoning vs. Cold Confirmation Bias

- **Cold confirmation bias**: purely cognitive, happens without emotional investment
- **Motivated reasoning**: emotionally driven, you *want* a particular answer to be true

When your belief is tied to identity — your political group, professional identity, investment thesis — the cost of being wrong is social and psychological, not just epistemic.

## Practical Countermeasures

**Pre-mortems**: Before committing to a decision, imagine it failed. What caused the failure? This forces disconfirming scenarios while the decision is still open.

**Steel-manning**: Before dismissing an opposing view, reconstruct it in its strongest form. If you can't make the opponent's case better than they can, you don't understand it well enough to rebut it.

**Base-rate anchoring**: Find the base rate before deep-diving into specifics. How often do companies like this succeed? Base rates don't anchor to your expectations the way specific information does.

**Red teams**: Assign someone the explicit role of finding flaws. Not someone casually skeptical, but someone whose job depends on breaking the plan.

## The Meta-Problem

Knowing about confirmation bias doesn't protect you from it. Studies show that teaching people about cognitive biases reduces their bias estimates for *other people* more than for themselves.

The reliable defense isn't awareness — it's process design. Build the countermeasures into the workflow before you need them.

Confirmation Bias — Why Your Brain Filters Out the Truth

// COMMENTS

ON THIS PAGE