Trap/Heuristic/No. 0068

Availability Heuristic

The availability heuristic is a mental shortcut for judging frequency or likelihood by how easily examples come to mind. Named by Amos Tversky and Daniel Kahneman in 1973, it can lead people to treat vivid or recent events as more common than evidence suggests.

Also called Availability Bias

a trap: easy to walk into

Evidence: Well established
Read: 6 min
Links: 14 connections
Useful when: Deciding under uncertainty · Evaluating a claim · Forecasting · Reading data and statistics · Risk and safety

01You've seen this when…

in life
A friend describes a frightening reaction to a medication. At your next appointment, that story comes to mind immediately; the uneventful experiences of thousands of patients don’t.
at work
Two customer complaints dominate the team chat all week. In the planning meeting, everyone treats that problem as the biggest reason customers leave, though nobody has checked the cancellation data.
out in the world
Clips of subway assaults keep appearing in your feed. You start to feel that the subway is becoming more dangerous, without knowing whether assaults are rising or the clips are simply spreading.

02The idea

When you’re asked how often something happens, you rarely have a complete count ready. Instead, you search your memory. If examples arrive quickly, the event feels common. If you struggle to find any, it feels rare.

That’s the availability heuristic: using the ease of bringing examples to mind as a shortcut for estimating frequency or probability. The search itself can supply a feeling of how much evidence there is, whether you count the examples or simply try to recall them.

This often works. If you regularly encounter late trains, late trains probably are common on your route. But an example can be easy to recall for reasons that have little to do with its frequency: it was shocking, happened yesterday, involved someone you love, or appeared in your feed twenty times.

The mistake is treating easy to remember as likely to happen. Your memory contains a limited sample of the world. Attention, emotion, experience, and what other people choose to show you filter that sample.

03Why it happens

Frequent experiences usually leave accessible memories. Repeated encounters give you more opportunities to learn and recall something. That makes accessibility a reasonable, indirect clue to frequency.
Memorable events get extra weight. A dramatic accident can be easier to recall than years of uneventful journeys. Salience describes what grabs attention, which can make an event easier to recall. That ease of recall can later shape a frequency judgment through the availability heuristic.
Repeated coverage can resemble repeated events. Coverage of one incident across several outlets gives you several encounters with the same story. In a quick estimate, those encounters can feel like several separate incidents.
Difficulty can look like scarcity. In a classic experiment, people asked to recall twelve examples of their own assertive behavior rated themselves as less assertive than people asked for six. The longer search felt harder, so more evidence could come with lower confidence.

Emotion can also change a decision without changing your estimate of probability. Avoiding something because it feels frightening is closer to the affect heuristic. Feeling frightened and therefore concluding that it happens often is where availability may enter.

04A worked example

Amos Tversky and Daniel Kahneman’s 1973 paper described a deceptively simple test. It used English words at least three letters long: participants judged whether letters such as K appeared more often in the first position or the third.

People tended to favor the first position. Yet the English frequency data used for the test showed that the selected letters occurred more often in the third position.

What it looks like An informed estimate based on your knowledge of English. Words beginning with K—king, key, kitchen—arrive quickly, so first-position K seems more common.

What’s actually going on Retrieving words by their first letter is easier than searching for a letter tucked inside them. The authors explained the judgments in terms of that unequal accessibility. The search method favored one answer even though the frequency data favored the other.

What would have helped Basing your estimate on a count of occurrences in a representative collection of text. That’s a proposed remedy that the experiment left untested. The practical lesson is to check whether your method of finding examples makes one category easier to retrieve than its competitor.

05How to spot it

A frequency claim rests on a few stories that come to mind immediately.
Your estimate changes sharply after a striking event, but you haven’t learned whether the underlying rate changed.
You can name the failures but have no count of the uneventful cases.
The same incident reaches you through several channels and starts to feel like a pattern.
One category is much easier to search than another: public failures versus quiet successes, for example.

A useful distinction: recency bias gives extra weight to recent information. Availability can favor an old event too, if it’s still exceptionally easy to recall.

06What to do instead

Turn the impression into a countable claim. Specify which event you’re counting in a defined population over a set time period. When you feel that theft is everywhere, a claim about theft per bicycle parked gives you a rate to check.
Ask for the denominator. The significance of three failures depends on whether they occurred among ten attempts or ten thousand. Use a base rate to put memorable cases in context.
Use a broader sample. A random set of customer records, a complete incident log, or a credible population survey can expand the evidence available from memory. Searching for more examples online may simply reproduce the same sampling bias.
Count the events behind the posts. Trace five alarming posts to the incidents they describe: they may cover five separate events or retell one event five times. Also ask which events never get posted.
Use comparable cases when predicting. For a project or investment, inspect outcomes from a relevant group of past cases. Reference-class forecasting supplies a systematic comparison that reaches beyond the examples you happen to remember.
Add a pause before the judgment. Write down your initial estimate and the example that prompted it. Then check one source with actual counts. This is a small cognitive forcing strategy: build that count check into the process.

07When it isn’t a mistake

Easy recall can be good evidence when your experience is broad, relevant, and reasonably representative. A mechanic who repeatedly sees a particular fault may have useful information that hasn’t reached a published database. Dismissing that knowledge just because it comes from memory would be another mistake.

A vivid event can also reveal a new hazard. Discovering that a device can catch fire matters even before you know the exact rate. You may need to investigate or take a temporary precaution before deciding whether the concern reflects bias.

Probability is one part of a decision. Avoiding a rare danger can be reasonable when its consequences are severe and the alternative costs little. Availability becomes a problem when accessibility distorts the estimated likelihood—not whenever someone remembers a frightening event or chooses cautiously.

08Roots

In 1991, psychologists John Anderson and Lael Schooler compared memory with records of everyday information, including newspaper headlines. Their question was practical: which information is likely to be needed again? Frequency and recency helped predict future encounters, much as they help predict what people can retrieve from memory.

That provides an adaptive account of availability. A mind with limited time benefits from keeping recently and repeatedly useful information close at hand. Consulting what’s accessible can then be an economical way to estimate what is common. The fit between environmental patterns and memory is supported by research; the stronger claim that natural selection specifically built the availability heuristic is plausible, not established by those studies. Efficient memory also cannot guarantee accurate judgments when the information reaching it is unrepresentative.

Tversky and Kahneman had given the shortcut its name earlier, in their 1973 paper. Their letter-position tests made the problem visible without relying on fear or dramatic news: ordinary knowledge could produce a wrong estimate because one mental search was easier than another. The idea traveled into explanations of risk perception and everyday judgment. Its lasting insight is that the feeling of having evidence can come partly from how easily the evidence arrives.

09How solid is this?

Experiments support the influence of accessible examples and experienced retrieval difficulty on judgments. Effects depend on the task and whether people treat ease of recall as informative. Establishing that availability caused a particular decision requires evidence beyond a vivid story.

10Connections

Often confused withRecency Bias, Salience Bias, Negativity Bias, Representativeness Heuristic
Countered byReference-Class Forecasting, Cognitive Forcing Strategy
Can lead to Base Rate Fallacy, False Consensus Effect
Part of Heuristic
See also Affect Heuristic, Frequency Illusion, Ecological Rationality, Sampling Bias, Peak-End Rule

+ 4 more in the list

11Origin and sources

Amos Tversky and Daniel Kahneman described and named the heuristic in their 1973 paper, Availability: A heuristic for judging frequency and probability.

Suggest an edit· Updated 2026-10-02