Availability Heuristic — We Judge Risk by What Comes to Mind First

## The Mechanism

When you estimate the probability of an event, you often rely on how easily examples come to mind. The more readily you can recall instances, the more probable the event feels.

This is the availability heuristic — first systematically described by Amos Tversky and Daniel Kahneman in 1973.

The problem: ease of recall is driven by *memorability*, not frequency. Dramatic, recent, vivid, emotionally charged events are easier to recall than common but boring ones.

## The Classic Demonstration

Kahneman and Tversky asked: Which is more common — words beginning with *K*, or words where *K* is the third letter?

Most people say words beginning with K. Words with K as the third letter are roughly twice as common. We retrieve words by their first letter, so first-letter words feel more available — and therefore more common.

## Where This Goes Wrong

**Airplane vs. car risk**: Plane crashes are vivid, widely reported. Car accidents are common but quickly forgotten. Per kilometer traveled, flying is approximately 50-100x safer than driving, depending on the metric.

**Investment after media coverage**: Stocks that appear frequently in news receive more investor attention regardless of whether the coverage reflects value. Post-crash coverage causes overweighting of downside risk.

**Security theater**: After high-profile attacks, resources shift to preventing the specific method used, even when the underlying risk profile hasn't changed.

## The Fluency Twist

A subtler version: it's not just what you remember, but *how easily* you retrieve it.

Schwarz et al. (1991): People asked to generate 12 examples of assertive behavior rated themselves as *less assertive* than people asked for only 6 examples. Retrieving 12 is hard, so the difficulty signals "I'm not that assertive." The experience of retrieval — not the content — shapes the judgment.

## Calibration Strategy

Consult actual statistics before forming probability estimates.

For repeated decisions, track your predictions against outcomes. The feedback loop corrects the heuristic over time.

For one-off decisions, ask: "What does the reference class data say?" If starting a restaurant, the relevant question isn't how many successful restaurants you can name — it's the base rate of restaurant survival.

Your memory is a search engine optimized for emotional salience. Use it for retrieval, not probability estimation.

Availability Heuristic — We Judge Risk by What Comes to Mind First

// COMMENTS

ON THIS PAGE