Cognitive biases
The availability heuristic: why vivid and recent things seem common
A neighbor tells you that a friend of hers had a car window smashed and a bag stolen in a nearby neighborhood. A couple of days later a similar story flashes by in the news. And now you are sure that “people steal things all the time” in that district. You have not seen a single figure, but the picture seems clear.
This is the availability heuristic: we judge how common or how likely something is by how easily examples come to mind. Usually it is a decent rule, because common things really are easier to recall. But vivid, recent, and loud things are easy to recall too. Then the estimate slides off.
Below, we look at the classic experiments, where this trap appears in ordinary life, and how to check estimates that come “from the top of your head.” In the middle of the article there is a short experiment with a list of names.
Contents
In short: how the availability heuristic works
What the availability heuristic is
A heuristic is a quick rule that saves effort. Availability is how easily an example comes to mind. Amos Tversky and Daniel Kahneman described this rule in a 1973 paper, and a year later included it in a well-known review together with the anchoring effect.
One of their experiments is very simple. Participants listened to a list of names. One gender had slightly fewer names in the list, but all of them belonged to famous people. The other gender had one more name, but almost nobody knew those people. Then some participants were asked to recall as many names as they could, and others were asked which group the list had more of, men or women. The famous names were recalled better. And most people answered that the list had more of the gender with the famous names. In other words, they were wrong.
This experiment was later repeated. In a 1997 replication the effect held up, and it was not weak: moderate or large by psychology’s standards.
You can see something similar in more serious questions. In a 1978 study, people were asked to estimate how many people die each year from different causes. The answers of different people were remarkably similar to one another, but they were biased in the same way: rare causes were overestimated and common causes underestimated. And among causes of roughly the same frequency, some were overestimated and others underestimated. The authors linked this to how much is written and said about a cause, how memorable it is, and how easy it is to imagine. Asking people to avoid these errors on purpose hardly helped.
Try it yourself: a list of names
The experiment takes about a minute. Here is what will happen.
Step 1. Nineteen names will appear on the screen one at a time, about a second for each. Just read them. You do not need to count or write anything down.
Step 2. We will ask which group the list had more of: men or women.
Step 3. You will estimate how many there were of each.
Step 4. We will sort the names into two columns, and you will see how it really was.
Why names? A famous name works like a loud piece of news: it catches your attention and later comes to mind more easily. And the question “which group was bigger?” is a question about frequency. It is the very question we answer from memory in daily life: do people steal a lot, do people divorce a lot, do people get this illness a lot. The names flash by quickly on purpose. If you had time to count, the task would test your carefulness, not your memory.
What usually happens. Most people answer that the group with the famous names was larger. Even though it had one name fewer. Famous names are easy to recall, and the feeling “a lot comes to mind” quietly turns into “there were a lot.”
If it came out differently for you, that does not mean the effect is not real. The difference is only one name, and in a single answer chance decides a lot: you can simply guess. The effect shows up on average across many people. And if you started counting, you were solving a different task, and that is a useful lesson too: counting protects you better than impressions do.
Why easy seems common
Imagine that memory works like a search engine. You type in the query “thefts in our district.” If results appear at once, it seems there are many. If you have to scroll for a long time, it seems there are few. This is an image, not a description of how memory is built. But it is close to what was tested in experiments.
The best-known such experiment was run by Norbert Schwarz and colleagues in 1991. Some participants were asked to recall 6 occasions when they had behaved with confidence, others 12. Six occasions come to mind easily. Twelve come with difficulty. And those who recalled twelve then rated themselves as less confident people than those who recalled six. Even though they had more examples in their heads. According to the authors, what mattered was not the number of examples but the feeling: “this is hard to recall, so maybe I’m not that confident.”
That experiment had small groups of participants, so what came afterward matters. A meta-analysis of 263 studies showed that the ease-of-retrieval effect is real and of medium size. But published studies most likely overstate it, by up to a third by the authors’ estimate: weak results get into journals less often. And the feeling of ease itself explains only part of the effect, roughly a third to a half.
Why do we reason this way at all? Because most of the time it works. What happens often, we also see more often and recall faster. The failure comes when an example is easy to recall for another reason: it is vivid, recent, frightening, touched you personally, or everyone is discussing it.
Where it gets in the way
Relationships. “I do everything around the house, and you do nothing.” You saw all of your own chores: washing the dishes, calling the repairman, remembering your grandmother’s birthday. You did not see all of your partner’s. In a study by Michael Ross and Fiore Sicoly, people, including married couples, recalled their own contribution to a joint effort more easily and credited themselves with more than others credited them with. This does not mean the reproach is always wrong. But the feeling “I do more” is poor evidence.
Work and money. In a conversation about the year’s results, the last few weeks and a couple of vivid episodes come to mind: a loud failure or a loud success. Money decisions lean on stories in a similar way: an acquaintance tells you how well an investment worked out, and the investment seems safe. People who failed usually do not tell their stories.
News and the feed. The news reports what happened, not what did not happen. Rare and dramatic events make the headlines and are then easy to recall. Common and boring ones do not. In the 1978 study, the authors linked errors in judging risks, among other things, to what gets written about more. That is why after a run of alarming news the world can seem more dangerous even if the figures have not changed. How strongly this affects a particular person depends on the topic and on what they already know.
Health and how you see yourself. A friend’s story about a rare illness can make it seem “likely” for you too for weeks. And Schwarz’s experiment shows that we sometimes build our opinion of ourselves from how easily examples come to mind. If it is hard to recall your successes, that does not mean you had none.
How to check estimates that come from your head
The availability heuristic is one of the biases on the cognitive bias map, where you can filter them by area of life. With news and the feed, it is often joined by the negativity bias: bad headlines grab attention harder.
Frequently asked questions
Is the availability heuristic a thinking error?
Not in itself. It is a quick rule, and most of the time it gives a reasonable answer, because common things really are easier to recall. It becomes an error when what comes to mind easily is not the common but the vivid, the recent, or the loud.
How is it different from confirmation bias?
With the availability heuristic, we judge frequency by what came to mind first. With confirmation bias, we already hold an opinion, and we look for and notice what supports it. In life they often work together: a vivid case produces an opinion, and then the opinion picks out new cases.
If I know about this trap, does it stop working?
Not really. In the 1978 study, people were asked outright not to fall for such errors, and their estimates hardly improved. What helps is not knowledge but action: finding figures instead of stories.
Why can recalling more examples make a feeling weaker, not stronger?
In the 1991 experiment, people who had to recall 12 occasions of confident behavior rated themselves as less confident than those who recalled 6. According to the authors, the difficulty of recalling mattered more than the number of examples. A later meta-analysis found that this ease-of-retrieval effect is real and of medium size, but probably overstated in published studies, and that the feeling of ease explains only part of it.
