The availability heuristic: Why your brain confuses “easy” with “true”

It’s spring break, and all your friends traveled somewhere warm.

And they’re all posting pictures on Facebook.

Your life sucks.

…Only to an extent.

What you won’t see is that most of your friends are stuck in Michigan, just like you are. They’re just not posting pictures of themselves on Facebook.

You get the wrong impression, because pictures of people on the beach have a disproportionate influence on your perception of how everyone else is spending spring break.

Now let’s say you’re evaluating travel options for a flight to someplace warm.

Nobody likes missing a connection, so you’ll choose airports that give you the best chance of making your connecting flight. And how do you do this? If you’re like most travelers, you’ll think about airports where you’ve missed a connection in the past. You’ll think about when you’ve been stuck on the runway in a snowstorm. Or you’ll remember the time you slept on the floor because you couldn’t get a flight until the next morning. Those are vivid, memorable, and often negative experiences that leave a lasting impression. And those experiences—not the dozens of times you’ve connected without incident—will have an outsized influence on where you decide to fly next.

(I’ve avoided flying through Minneapolis this year for the sole reason that I spent last Christmas Eve there, and I don’t want to repeat that experience. Never mind that in the previous twenty years I’ve flown through Minneapolis nearly a hundred times without incident.)

What do these experiences have in common?

This: the events, memories, experiences, topics, and ideas that come to mind most easily are believed to be the truest.

It’s obvious everyone else is on the beach during spring break, so my life must suck. It’s easy to remember horrible airport experiences, so I’d better not connect through the airport where I misconnected last time.

Easy equals true.

(Or: hard equals difficult.)

This is the availability heuristic.

What is the availability heuristic?

Your brain needs to process more data than it can handle. You need to make decisions quickly and correctly. But since you can’t do both, your brain is forced to optimize between efficiency and accuracy. It’s a constant tradeoff. The way it does this is by taking mental shortcuts that are reliable most of the time, but not all of the time. These shortcuts are called heuristics.

The availability heuristic simply refers to a specific mental shortcut: what comes to mind the easiest—what’s most available—is true.

Let’s use this as our working definition of the availability heuristic:

The availability heuristic is a shortcut that confuses easy with true when you make a decision. You rely on the ease with which something comes to mind instead of the content of what comes to mind when making a decision. As a result, you are more likely to make bad decisions, miscalculate and overreact to risks, hold inaccurate perceptions about yourself and others, and behave in ways that aren’t in your best interests.

In this post, we are going to unpack what the availability heuristic is, why it matters, how it works, and how to overcome it.

To begin, let’s take a closer look at what it is.

Think about how many words start with the letter k compared to the number of words with k as the third letter. Does the letter k occur more often at the beginning of a word or as the third letter?

In a now-famous study, two Israeli psychologists named Amos Tversky and Daniel Kahneman found that twice as many people thought k occurred at the beginning of words than in the third position of words.[1]

K in first position 105 (69.1%)
K in third position 47 (30.9%)

But does it?

Actually, k appears more often in the third position in a word.

Why do most people get it wrong?

Because words where k is at the beginning, such as kite, keep, or kayak, come to mind easily when you’re thinking of the letter k. They are readily available in your memory.

But you’re less likely to think about words like take, makeshift, ankle, inkjet, or acknowledge. These words are less availability in memory.

Availability influences influence your perceived frequency. Your perceived frequency is based on how easily instances of words that start with k come to mind compared to words where k is in the third position. Because words where k is the first letter come to mind more easily than words where k is the third letter, and because you are more likely to believe things that come to mind easily are true, you get it wrong.

In another one of Tversky and Kahneman’s thought experiments, they asked people to guess the likelihood of an imaginary couple getting a divorce.

How would you go about making such a guess? You’ll probably start by doing one of two things:

First, you’ll scan your memory of similar couples, perhaps thinking of couples with similar personalities or couples who have the same number of children. If, in the instances you bring to mind, there are lots of divorces, then it will be easy for you to predict divorce for this couple, too.

Second, you’ll put together a story or scenario that may lead to divorce—something she said, something he did. If you find it easy to come up with a story, then you’ll use the ease of creating the story as a shortcut for predicting divorce.

In neither actual—comparing this couple to other couples, and coming up with a story—are you using actual data to make your prediction.

The problem with the first is that the couples that come to mind most easily are not representative of all couples that could come to mind.

The problem with the second is that it’s easier to come up with some kinds of stories than others.

You’re basing your prediction on the ease with which you can bring to mind just enough data to answer the question: will this couple get a divorce? If it’s easy to bring the data to mind, then the couple will divorce. Easy equals true. If it’s not easy to bring data to mind, then the couple will stick together.

These examples—the positions of letters in words, a prediction about a hypothetical couple—are interesting thought exercises that reveal a quirk in our cognitive processes.

Next, let’s extend beyond thought exercises and enter the real world.

Get more articles about cognitive biases, consumer psychology, and human behavior.

Sign up (it’s free!)

Examples of the availability heuristic

Climate change

No matter what you think about climate change, when it’s hot out, people blame global warming.

The opposite happens during the winter. When it’s cold, climate change skeptics wonder why anyone would believe the earth is warming when it’s so cold out. The President of the United States recently fell into this trap:

Trump global warming tweet

At the exact moment Trump sent that Tweet, the United States was experiencing temperatures 15 to 30 degrees colder than average. The data most available to him was cold temperatures, not all temperatures averaged out. If it were, he would have realized global temperatures were actually .9 degrees warmer than average:

north america global warming tweet

Shark attacks

When there’s a shark attack, people avoid the beach and stay out of the water, even though they are no more or less likely to be attacked by a shark. It only seems like it.

Death-by-shark-attack is a vivid, gruesome, scary way to die. It’s the kind of thing that makes you decide not to go to the beach, or stay out of the water once you get there.

What you might not realize is other dangers lurk at the beach.

Take coconuts.

Did you know you are twice as likely to be killed by a coconut at the beach than a shark?[2] Coconuts are not only more dangerous than sharks. They also strike at random: you can see a shark coming, but it’s impossible to predict when a coconut might fall. You can also provoke a shark and expect a response. But if you provoke a coconut, it is, statistically no more or less likely to respond with rage.

Because it’s easier and scarier to imagine being killed by a shark than a coconut, death-by-shark comes to mind more easily than death-by-coconut, so it’s more likely to affect your behavior.

Buying insurance

Insurance protects you from the slight chance of a massive loss. You buy life insurance, medical insurance, homeowners insurance, flood insurance, or earthquake insurance because there’s a small chance you’ll need it.

The problem is consumers buy insurance based not on actual risk but on perceived risk.

If there hasn’t been a flood in awhile, you’re likely to let your insurance lapse. That’s because floods don’t occur very often. Between 2001 and 2009, the number of flood insurance policies grew between 0% and 4% per year, with the exception of the year 2006. During 2006, the number of policies grew 14.3%—more than three-fold the norm.[3]

What happened in 2006?

Hurricane Katrina.

After Katrina, the concept of flooding and its consequences was more available. People saw pictures of New Orleans underwater, heard about failed relief efforts, and often witnessed flooding themselves. As a result, people perceived their risk to be greater than before, which made them more likely to purchase flood insurance.

The median length of a flood insurance policy is between two and four years. This rate holds whether there’s a natural disaster or not. After one year, 73.2% of flood insurance policies are still in place. By the second year, this drops to 49.5%. Even Hurricane Katrina didn’t change the average policy length over the long term.[4]

Flood insurance policies in place after nine years
from Michel-Kerjan, E., de Forges, S., and Kunreuther, H. (2012). “Policy Tenure Under the U.S. National Flood Insurance Program (NFIP).” Risk Analysis, Vol. 32, No. 4, 2012.

The same phenomenon occurred after the 1989 Loma Prieta earthquake. In 1989, 63.1% of respondents estimated a 1 in 10 probability of a major earthquake damaging their community within the next 10 years. In 1990, 68.6% of the same respondents gave a 1 in 10 chance, and in 1993 it rose to 75.7%.[5]

The availability heuristic influences your decision to purchase insurance for up to nine years. One study found that a flood results in an 8% increase of insurance policies above normal that same year. This number goes up to 9% the year after and remains noticeably above normal for nine years. But by the tenth year, the number of insurance policies returns to the normal, predicted amount. It’s as if the flood had never occurred.[6]

People cancel their insurance policies if they don’t feel these policies protect them. If you pay for flood insurance year after year without experiencing a flood, you’ll question whether flood insurance is a necessary expense. If it’s easy to remember a flood, you’ll buy insurance, because it seems floods occur rather often. But as it becomes harder to remember a flood, it will begin to seem floods don’t occur very often after all.

However, the presence or absence of a natural disaster in one year does not change the statistical likelihood of the same event occurring (or not occurring) the following year. If a tornado barrels through your town next year, the tornado doesn’t remember that a different tornado did the same thing last year, or two years ago, or ten years ago. Only you remember.

(This phenomenon explains why, after the 1989 Loma Prieta earthquake, the stock price of major insurance companies climbed even as the companies took a financial hit from insurance payouts. Savvy investors correctly predicted that the earthquake would cause people to temporarily purchase earthquake insurance, which would more than offset the costs of the insurance payouts.[7])

Advertisements for antidepressants

Frequent ads for antidepressants cause you to believe you have a high risk of becoming depressed. One study indicates depression affects 6.5% of adults in a given year, and the lifetime risk for experiencing depression is 13% for males and 20 to 25% for females.[8] Another study found that 10% of adults experience depression each year.[9] These surveys give quite a range: 6.5% on the low side, to 25% on the high side.

However, when people are exposed to frequent ads to antidepressants, they estimate that 38% of adults experience depression each year.[10] When it’s easy to think about depression, people are more likely to think more people are depressed. As a result, people believe they are at greater risk for depression than they really are.

This causes unnecessary stress. It prompts people to get unnecessary medical care or be prescribed medication they don’t need. This drives up medical costs for themselves and others.

The decline of violence

The world is a safer, more peaceful place than it ever has been. Virtually all forms of violence are on the decline—some very rapidly—and have been for quite some time.

This trend continues right up until the present.

Yet you probably think the opposite: that hate crimes, terrorist acts, school shootings, homicides, and other kinds of violence are on the rise. According to a 2011 Gallup poll, 68% of people say there is more crime this year than last year and that crime is getting worse.

public perceptions of crime

This trend—the belief that violence is increasing, even as the world becomes safer—is covered in exhausting detail by Steven Pinker in his excellent book, The Better Angels of Our Nature: Why Violence Has Declined. Here are some of Pinker’s data:

Homicide rate in western europe
source: ourworldindata.org
Homicide rates in the northeastern United States
source: Steven Pinker, The Better Angels of Our Nature

Pinker’s thesis received a strong reaction. That’s because everything about how we experience the world seems to point to one thing: that things are bad and getting worse.

Why do we think this?

In Enlightenment Now: The Case for Reason, Science, Humanism, and Progress, Pinker’s follow-up to Better Angels, Pinker writes:

“It’s easy to see how the Availability heuristic, stoked by the news policy, ‘If it bleeds, it leads,’ could induce a sense of gloom about the state of the world. Media scholars who tally news stories of different kinds, or present editors with a menu of possible stories and see which they pick and how they display them, have confirmed that the gatekeepers prefer negative to positive coverage, holding the events constant. That, in turn, provides an easy formula for pessimists on the editorial page: make a list of all the worst things that are happening anywhere on the planet that week, and you have an impressive-sounding case that civilization has never faced greater peril. . . . Heavy news watchers can become miscalibrated.”[11]

We are biased toward what’s most available to us. It’s easy to think of violent acts we’ve experienced ourselves or seen on the news.

It’s less easy to think of normal, everyday, average acts because they are less available in your mind.

Investment decisions

The availability heuristic also affects whether, where, and how you invest your money.

Even professional traders invest in companies they tend to hear the most about. When a company is in the news, people are more likely to purchase stock in that company. The same is true when a company’s stock experiences high trading volume or posts bigger- or smaller-than-normal one-day gains or losses.[12] People are also more likely to buy stock after a large positive earnings surprise and they are more likely to sell stock after a large negative earnings surprise, compared to smaller earnings surprises.[13]

Product recalls have the same effect. When a company issues a product recall, investors sell their stock, driving the share price down. A product recall not only drags down one company’s share price, it drags down the share price of the entire sector. That’s because the product category—not just a single company’s product recall—becomes more available in the mind of both consumers who buy that kind of product and in the minds of investors who buy the company’s stock.[14]

In perhaps the oddest example of the effect of the availability heuristic on a company’s stock performance is the Anne Hathaway effect. Dan Mirvish from the Huffington Post noticed a correlation between the increase in share price of Berkshire Hathaway and release dates for movies starring Anne Hathaway:

Date Anne Hathaway-related event Berkshire Hathaway stock price
October 3, 2008 Rachel Getting Married opens up .44%
January 5, 2009 Bride Wars opens up 2.61%
February 8, 2010 Valentine’s Day opens up 1.01%
March 5, 2010 Alice in Wonderland opens up .74%
November 24, 2010 Love and Other Drugs opens up 1.62%
November 29, 2010 Anne Hathaway announced as co-host of the Oscars up .25%

News about Anne Hathaway makes people slightly more likely to think about Berkshire Hathaway, which drives up demand for Berkshire Hathaway stock.

What’s at stake when we fall for the availability heuristic

The availability heuristic, like any heuristic, is, by definition, a shortcut. Shortcuts are often good. They get you from point A to point B more quickly. But sometimes they cause real harm.

They cause the most harm when you’re required to make judgments or predictions for low probability, extreme outcomes. For these kinds of events, you’re more likely to use availability data than frequency data. You overcorrect.

Arsenic in drinking water

Consider the controversy in 2001 over how much arsenic should be allowed in drinking water. Before Bill Clinton left office, he mandated that the standard for arsenic in drinking water be reduced from 50 parts per billion to 10 parts per billion. Arsenic increases cancer risk. According to a 1999 National Academy of Sciences study, at 50 parts per billion, your risk is 1 in 100. At 10 parts per billion, it’s 1 in 500.

The new mandate didn’t come without a cost. For a water district with a million households, the new standards cost 86¢ per household per year. For water districts with 500 households, the costs rose to $163 each. And the numbers went up for smaller water districts. Because of the higher costs, George Bush changed the mandate back to 50 parts per billion.

The response was universally critical. Bush’s critics argued he had enacted regulation that would put people at a greater risk for cancer. He caved to the public outcry and reversed his decision, returning the standard acceptable rate of arsenic in drinking water to 10 parts per billion.

Was this the right decision? To know for sure, you would need to weigh the benefit of lowering cancer risk against the high costs of meeting the new standards. In some cases, it would be worth it. In others, not.

The larger point is not that such calculations weren’t done (they were). The larger point is the final decision was not based on actual frequency data—it was made on data that came to mind most easily.[15]

Increase in traffic accidents after September 11 attacks

The same thing happened after the terrorist attacks on September 11, 2001. After seeing images of planes crashing into buildings, and hearing about chaos at U.S. airports, people opted to drive instead.

As a result, an additional 2,170 people who would have otherwise flown lost their lives in car accidents. Some of these people were forced to drive because U.S. airspace was closed for three days after the attacks. Others decided to avoid the hassle of extra security at the airports. But most decided to drive out of fear of dying in another attack—even though the possibility was remote. Two economists detected the change in traveler behavior as late as January, 2002.[16]

Because people have trouble calculating the real risk of low-probability, extreme events—like dying in an aviation-related terrorist attack, or dying in a car accident—they rely on non-frequency cues to determine how they should travel.

Excessive media consumption

Watching the news makes you depressed. Much of what’s found on TV news is negative, extreme, and low probability—and this distorts your view of the world.

A team of British psychologists found heavy news consumption “raised self-reported measures of anxious and sad mood, and subsequently led to the enhanced catastrophizing of personal worries,” and “it can exacerbate a range of personal concerns.”[17] When people devote outsized attention to what’s most available, they come stressed, worried, indifferent, or apathetic.

Heavy Facebook use leads to the same effect, but for opposite reasons. People represent themselves positively on Facebook. When you scroll through your newsfeed, you only see others’ positive representations of themselves. It’s a skewed sample; you’ll rarely see neutral or negative representations of your friends’ lives. (You’re bound to see more pictures of people windsurfing, eating an incredible meal, or having fun with their kinds than sitting on their couch on a Friday night bored out of their minds.)

This gives you the impression that your friends are happier and have better lives than you do.

But offline, real-world interactions with your friends contain no such selection bias. Two psychologists have found that people who have more than the average number of Facebook friends rely on the availability heuristic when they compare themselves to their friends, and as a result, they believe their friends are happier and have better lives than they do. The same study found when people have more offline interactions with their friends, they are less likely to believe their friends have better lives and are happier than they are. That’s because “when people have more offline interactions with their friends, knowing more stories about others’ lives, both positive and negative, they are less persuaded that others are happier than themselves.”[18]

What makes something more availability: ease of recall

So far, we have shown how the availability heuristic is the shortcut that confuses easy with true.

But what makes something easy? And how much does easiness contribute to your recall?

If you were an assertive person, it would be easy to think of examples of your assertiveness. I would ask you to tell me about times you acted assertively, and you would have no problem coming up with a long list.

Now, suppose, after you made this list, I asked you whether you thought you were an assertive person.

Most people assume the response would be yes. After all, you’ve got this giant list of examples of your assertive paper right in front of you.

But that’s not the case. If I asked you, the assertive person, holding a list of examples of assertive behavior, if you really are an assertive person, you would be more likely to say no.

In the experiment that proved this, people’s perception of how assertive they perceived themselves was directly correlated with how difficult it was to think of examples of assertive behavior.[19]

This sounds backward. People who have compiled extensive evidence of their assertive behavior should believe they, themselves, are assertive.

The reason they don’t believe they are assertive is that coming up with a long list of examples of assertive behavior is hard. If I’m an assertive person, it should be easy for me to come up with a bunch of examples of assertive behavior. When it’s easy to come up with a list of assertive behavior–i.e. a short list–then you’ll think you’re assertive. When it’s hard to come up with a list of assertive behavior–i.e a long list–then you’ll think you’re less assertive.

Whether or not you think you’re assertive has nothing to do with how many examples you can generate to support your position. It has everything to do with how easy or hard it is to generate those examples.

The important distinction is between 1) the content of what they are thinking, and 2) how easy it is for them to think it.

Ease of recalled content affects judgment far more than the actual content of what people recall.

Brace yourself, because things are about to get really strange.

Ease of recall can produce unexpected behavior

Brace yourself, because things are about to get really strange.

In one study, people were asked how often they rode their bicycles. Those who could think of lots of examples of riding their bikes believed they rode infrequently. Those who thought of only a few examples believed they rode frequently. Non-riders viewed themselves as avid cyclists, while frequent riders kicked themselves for never riding.[20]

Be careful when you’re trying to persuade someone to take your position, buy your product, or support your argument. Each new reason provides a diminishing return, and at a certain point, giving additional reasons will actually be counterproductive. People who are asked to give many reasons for a choice they have made instead of a few reasons tend to be less confident in the choice they have made.[21]

This was found to be true in a study on views of public transportation. People were asked to give reasons either in support of or against the use of public transportation. Those who gave seven reasons to use public transportation reported more negative views of public transportation than those who gave three reasons.[22]

If you’re an instructor and you want your students to give you a positive rating, ask them for so much negative feedback that they have a hard time thinking of more. Because easy equals true and hard equals false, the difficulty of thinking of negative feedback will cause students to discount the actual content of the negative feedback. As a result, you get a positive rating.

This makes sense if you think about it: If it’s so hard to think of negative feedback, this must be a pretty great course.

One professor who tried this found that students who offered more criticism rated the course 12% more positively than everyone else.[23]

What’s even more strange is that it’s simply enough to expect thinking of examples to be difficult. In one study, two versions of ads for BMWs were shown. One ad read: “There are many reasons to choose a BMW. Can you name one?” The other ad read: “There are many reasons to choose a BMW. Can you name ten?” Without actually thinking of reasons, simply knowing that it’s easier to think of one reason than ten reasons made the first ad 27% more effective, even though it implies there are fewer reasons to buy a BMW.[24]

Ease of recall can override other inputs, like recency. People were asked to recall the Oklahoma City bombing. When they recalled two details from the bombing (easy!), they believed the bombing happened recently. But when they recalled ten details (not so easy), they believed it happened further in the past. When we estimate a date, an important clue is how easy to remember something happened: if it’s easy to remember, it must have happened recently. If it’s difficult to remember, it must have happened in the more distant past. But if we take something recent and frame it in a way that makes it difficult to recall details about, then we believe it happened earlier than it actually did.[25]

Get more articles about cognitive biases, consumer psychology, and human behavior.

Sign up (it’s free!)

What makes something easier to recall?

As we’ve seen, ease-of-recall can override the actual content of what you’re recalling when you’re making judgment.

But what makes something easy to recall?

Things that are easier to recall have several attributes in common:

  1. Frequent is easier to recall than infrequent
  2. Extreme is easier to recall than ordinary
  3. Negative is easier to recall than positive
  4. Recent is easier to recall than the distant past
  5. Vivid is easier to recall than non-vivid

availability heuristic

Let’s take a look at each of these.

1. Frequent is easier to recall than infrequent

Although the availability heuristic works by substituting frequency data for other inputs that come to mind more easily.

At the same time, we should acknowledge that even though our frequency isn’t perfect, it’s actually not too bad, either. There are more pigeons than orioles in urban areas. There are more lawyers than tailors in your town. The word bacon occurs more often than the word pastrami. Iowa produces more corn than Wyoming.

You can extrapolate from a limited amount of data and, on average, still make a correct guess. One psychologist found that when people were asked to guess the frequency of the occurrence of letters in a set of words, the guesses correlated at .79 with the actual frequency.[26] (Think of correlation on a scale of -1 to 1, with 1 being perfect correlation.)

In most cases, our intuition is correct. We recall things more easily because we correctly judge they occur more frequently.

Frequent things come to mind more often than infrequent things.

2. Extreme is easier to recall than ordinary

In one study, participants were presented with lists of two causes of death and asked to do two things:

  1. First, they were to estimate which of the two causes of death was more likely.
  2. Second, they were to give a ratio to indicate how much more likely one cause was over the other. For example, if they thought dying in a car accident was more likely than from a car accident, then the ratio might be three-to-one: three people die in car accidents for every one person killed by lightning.[27]

Here are the results of the study:

Judged frequency of lethal events
Lichtenstein, Slovic, Fischhoff, Layman, & Combs, 1978. “Judged frequency of lethal events.” Journal of Experimental Psychology: Human Learning and Memory, 4 (6), 551–578.
Judged frequency of lethal events
Lichtenstein, Slovic, Fischhoff, Layman, & Combs, 1978. “Judged frequency of lethal events.” Journal of Experimental Psychology: Human Learning and Memory, 4 (6), 551–578.

As you can see, people made lots of misjudgments. Here are a few examples:

  1. Stroke causes 85% more deaths than accidents, but only 20% of students and 23% of the general population thought stroke was more likely than accidents at all. And students as a whole thought accidents were 25 times more likely than strokes.
  2. People also thought tornados killed more people than asthma attacks, even though asthma attacks kill 20 times more people than tornadoes every year.
  3. People thought death by lightning was unlikely compared to death from botulism, even though 52 times more people die from botulism every year.
  4. People also thought botulism and asthma kill roughly the same number of people each year, even though asthma kills 900 times more people.

When the researchers took a closer look at the data, they noticed when were presented with two causes of death, people usually guessed which cause was more likely, even though they rarely guessed the correct ratio.

Third, when a cause of death had a 2:1 ratio or more–or, when one cause of death was at least twice as likely as the other–people could identify it as being more frequent. But when a cause of death became fifty times more likely than another cause of death, people overestimated the less likely cause. There seems to be a range between 2:1 and 50:1 where our estimates are close. We are not good at predicting how often extreme, but rare, causes of death actually occur. The most overestimated causes of death are botulism, tornado, flood, homicide, car accidents, other accidents of all kinds, and cancer. The most underestimated causes of death are asthma, tuberculosis, diabetes, stomach cancer, stroke, and heart disease.

Causes of death
Lichtenstein, Slovic, Fischhoff, Layman, & Combs, 1978. “Judged frequency of lethal events.” Journal of Experimental Psychology: Human Learning and Memory, 4 (6), 551–578.

Here’s another way to look at it. If you hear thunder, the availability heuristic makes you run inside to avoid being struck by lightning, but, in your attempt to flee to safety, it won’t prevent you from falling down the stairs, which is far more likely to kill you.

Negative is easier to recall than positive

Is the glass half empty or half full?

Half empty, it seems:

  • We are more likely to notice a threatening face in a crowd of people than a neutral or positive face.[28]
  • We feel the need to explain why negative events occur, but not positive events.[29]
  • Negative events such as losing money or receiving criticism elicit a greater physiological and cognitive reaction than their corresponding positive events: making money or receiving praise.[30]
  • When you have a good day, this doesn’t affect how you feel the following day. But when you have a bad day, your negative mood carries over into the next day.[31]
  • Negative interactions have an effect five times stronger than positive interactions. One researcher who studies happiness in marriages found that the number of positive interactions needs to outnumber negative interactions by five to one. If the ratio falls below five-to-one, the relationship is likely to fail.[32]
  • People recall negative events more often than positive events by a ratio of four-to-one. Negative events leave a stronger impression and are easier to remember.[33]
  • People spend more time looking at photos depicting negative events than photos depicting positive events, which indicates people pay more attention to bad events than good events when forming an overall impression.[34]

However, the glass is actually half full.

Even though negative events command more cognitive attention, numerous studies have shown people tend to experience more positive events than negative events.

And as time passes, the glass becomes even fuller.

While we are far more likely to notice negative events as they happen, we are far more likely to remember positive events as time passes. In a review of 52 memory studies, researchers found a much higher proportion of positive events over negative events.[35] The experience of the present tilts negative, but our memory of the past tilts positive. Things get better with the passage of time. Even though negative events have an outsized impact as they are experienced, they have a diminished impact as they are remembered.[36]

Additionally, people’s memory of a negative event becomes less negative over time at a fairly measurable rate. Three psychologists asked people to evaluate experiences they had both immediately after they occurred, and again at intervals of 3 months, 1 year, and 4 and a half years. As time passed, memories of the same experiences became more positive.[37]

What does this mean for ease of recall?

Let’s summarize what we know so far:

  1. People experience more positive things than negative things.
  2. People remember more negative things than positive things in the short term.
  3. People remember more positive things than negative things in the long term.
  4. As time passes, the negative memories become less negative.

If you’re making a judgment about the present or the recent past, negative experiences will come to mind more easily. But if you’re making a judgment about the more distant past, positive events will come to mind more easily.

However, as we’ll see next, recent events tend to come to mind more easily, regardless of whether they are positive or negative.

Recent is easier to recall than the distant past

You can probably tell me what you had for dinner last night, but you can’t tell me what you had for dinner 30 days or a year ago. Things that happened a long time ago are not as easy to recall as things that happened recently.

Before we discover why, let’s think about how recency interacts with frequency.

If I ask you how often you brushed your teeth in the last month, you can do the math: twice per day multiplied by thirty days is sixty brushings.

Now suppose I ask you how often you drove in the last month. Once again, you can make a good guess, but you’re probably not quite as sure. In this case, your guess at the overall frequency in the last month will be influenced by how often you’ve driven in the past week compared to how often you drove four weeks ago. What comes to mind easily–recency–becomes a substitute for calculating the overall frequency.

In an experiment, researchers split people into two groups. The first group was asked to give two instances when they ate out at a sit-down restaurant and two instances when they at a fast food restaurant in the previous four months. The second group was asked to give ten (not two) instances.[38]

Next, each group was asked to estimate on a scale of 1 to 7 how often they ate out. Note that they were not asked how many times they out. They were asked on a scale of one to seven, how often do you eat out? Or, did they perceive themselves as the kind of people who ate out a lot?

In the second part of the experiment, researchers assembled two groups and asked the same set of questions. Except for this time, they didn’t ask how often people ate out in the previous four months. Instead, they asked how often people ate out the previous spring.

Researchers then compared the responses of the groups asked about the previous four months and the previous spring.

When they compared the numbers, the researchers found that when people thought of how often they ate out in the recent past, the people who thought of two instances of eating out were more likely to perceive themselves as the kinds of people who eat out compared to the people who thought of ten instances of eating out (5.5 vs. 4.4 on a scale of 1 to 7). These results are similar to the experience earlier about assertive people who produced many examples of assertive behavior perceived themselves to be less assertive than people who produced only a few examples of assertive behavior.

But when people were asked about the previous spring—the more distant past—there was no difference. People who were asked to recall two instances of eating out in the past four months reported roughly the same subjective frequency as the people who were asked to recall ten instances (5 vs. 5.1 on a scale of 1 to 7).

This indicates that recency affects ease of recall. Recalling two events in the recent past is much easier than recalling ten events in the recent past, so people who recall two events think they eat out a lot, while people who recall ten events think they rarely eat out because the former is easy and the latter is hard. But recalling two events in the distant past is barely easier than recalling ten events in the distant past, which means the easy-it-true bias doesn’t have the same effect.

Vivid is easier to recall than non-vivid

We’ve already seen how shark attacks are more likely to prevent us from going to the beach than risks from falling coconuts because they come to mind more easily. You can probably imagine getting hit on the head with a coconut hurts. But imagine trying to outswim a hungry shark and your palms start to sweat.

Things that are vivid and emotionally jarring come to mind easily, which makes them more likely to influence your behavior. Compare two versions of an ad:

  1. Version 1: “When you drink don’t drive; you have one chance in 1,000,000 of getting into a fatal car crash, a much higher probability than most people believe.”
  2. Version 2: an image of a car wrapped around a tree accompanied by a statement by the grieving family not to drink and drive.

This kind of highly emotional, vivid imagery overrides probability calculations because it comes to mind more easily.

In a study, students were asked how much they would be willing to pay to avoid all risk of arsenic in drinking water.[39] Here are the results:

Probability Unemotional description Emotional description
1/100,000 $241.25 $250.00
1/1,000,000 $59.21 $211.67

You would expect that when arsenic risks went up by a factor of ten, willingness to pay to eliminate those risks would go up by the same amount. But that doesn’t happen. Willingness to pay goes up when descriptions of those risks are emotional, even if the probability doesn’t change.

Oddly, each time someone is killed by a shark, the lives of ten people is saved who would otherwise have drowned from riptides. Shark attacks are more vivid and more likely to affect your behavior, even though riptides may be just as likely to kill you.[40]

Get more articles about cognitive biases, consumer psychology, and human behavior.

Sign up (it’s free!)

When you’re most likely to rely on ease-of-recall

We’ve seen that a variety of factors make something easier to recall: frequent, extreme, negative, recent, and vivid events come to mind easily, which means they have an outsized influence on your judgments and decisions.

However, there are certain conditions where you’re far more likely to rely on ease of recall.

  • when you’re multitasking instead of focused,
  • when you’re a novice instead of an expert,
  • when you’re in a position of leadership or feel powerful, instead of in a submissive role,
  • when you’re happy instead of sad.

Or, to put it differently, if you’re a happy multitasking novice in a position of leadership, you’re more likely to be afraid of sharks than coconuts than if you’re a sad focused expert in a submissive role. Everyone is susceptible to substituting easy for true–it’s just that certain conditions make you more likely to do so.

Let’s take a look at why each of these conditions leads you to rely on ease-of-recall and make you more likely to use the availability heuristic.

You are likely to rely on ease of recall when you’re multitasking

In one study, participants were given lists of 15 words and later asked to recall them. Some participants were distracted while they reviewed the word lists.

People who were not distracted remembered 9.44 of the 15 words, and it took them 420 milliseconds to retrieve each word from memory. But when people were multitasking during retrieval, performance dropped. They remembered 8.22 of the 15 words, and it took 566 milliseconds. Recall took 34.7% longer and accuracy fell by 12.9%.

The researchers made another surprising discovery about recall. The brain places ease of recall in a privileged position when you’re multitasking. So if your brain is performing a recall task while doing something else, even though your recall will suffer (by 12.9%, it seems), the something else you’re doing will suffer even more.[41]

You are more likely to rely on ease of recall if you’re not an expert

When you are an expert are asked to make a judgment, you have a wider pool of experience to draw from, which means the recall process will be easier. But when you’re not an expert, you’ll have a limited number of available instances, which will make recall more difficult.

Non-experts are more affected by ease-of-recall bias. Having a limited number of experiences produces a small sample size, so it feels easier for a non-expert to identify a solution to a problem because the pool of problems is smaller, to begin with. Think up some possible problems is easier for a non-expert than an expert.

(This is the same underlying mechanism that makes infrequent bicyclists believe they ride their bikes a lot. They have only a few instances of bike-riding, so as a category, instances of bike-riding come to mind easily. And the fact that it comes to mind easily becomes the shortcut for making a judgment about how often a person rides their bike.)

A study compared 138 auto mechanics (experts) with 68 people who knew little about cars (non-experts).[42] It was more difficult for mechanics to diagnose a problem with the car than non-expert drivers. When non-experts were asked to think of a few reasons why their car wouldn’t start, they had little difficulty. But when they were asked to produce many reasons why their car wouldn’t start, they found it difficult. This showed they relied on ease of recall as a shortcut for diagnosing the problem with their car.

But this wasn’t the case with the auto mechanics—the experts. Not only did they experience a similar level of difficulty in thinking of many reasons instead of a few reasons, but they were also able to diagnose the correct problem from a pool of many possible reasons why a car might not start.

If you’re a non-expert, then you’ll rely on ease-of-recall if you possess just enough knowledge in a domain to retrieve information in your memory about it. When your car won’t start, you might guess your car has a dead battery or a bad starter. It’s easy for anyone to think of two things, expert or not. The ease-of-recall bias kicks in, and you misdiagnose the problem. Experts don’t have the same problem. It’s still harder for them to think of ten reasons than two reasons why a car won’t start; they just don’t let the ease or difficulty of the decision affect their diagnosis.

The researchers conclude:

“It is, therefore, reasonable to assume that people with a higher level of domain knowledge, who possess a more developed and complete cognitive representation of the focal domains, will be less prone to the bias than people with a lower level of domain knowledge.”

However, if you’re aware you’re not an expert, then you won’t rely on ease-of-recall. In the words Norbert Schwarz and his colleagues: “Not being able to name famous Spanish matadors, for example, doesn’t imply there aren’t any; it only implies one doesn’t know them.”[43]

Number of matadors you can think of Expertise level Subject to ease-of-recall bias?
0 low no
2 low yes
20 high no

If you can’t think of a single Spanish matador, then you won’t be subject to the bias. You can’t pretend—to yourself—that you have enough domain knowledge to make any kind of judgment. But if you know one or two matadors, watch out.

The lesson here is that if you have just enough information to make a dangerous decision, you probably will.

You are more likely to rely on ease of recall if you’re in a position of leadership or feel powerful

People in power (or who feel powerful) are more likely to rely on ease-of-recall when making a decision.

In one experiment, researchers asked business travelers passing through a busy airport to generate either two or six arguments for sending people to Mars. They were also asked to rate the difficulty of making the arguments on a scale of 1 (not difficult) to 10 (very difficult).

People who were not in positions of power rated making two arguments instead of six arguments slightly easier: 5.77 vs. 5.36.

But people in power “reported a more favorable attitude toward sending humans to Mars after generating few as opposed to many arguments.”[44]

The same researchers asked 83 managers and subordinates to recall either two or ten moments of leisure time from the previous two weeks. They were then asked to rank, on a 9-point scale, how happy they were with the leisure time they had taken. Managers who recalled two moments of leisure time indicated greater satisfaction than managers who recalled ten moments of leisure time (4.95 vs. 3.82). The same distinction wasn’t true for subordinates: when subordinates recalled two moments of leisure from the previous two weeks, they gave it a happiness rating of 3.85. But when subordinates offered ten moments of leisure time, they rated it at 4.92.

What’s clear is that managers relied on the ease with which they could remember as a shortcut for judging how happy they were, while subordinates relied on the actual content of the experiences themselves to judge how happy they were.

The lesson? If you are in a position of leadership, then you are far more likely to allow biases and fallible human intuition to guide your decisions instead of sound judgment and reasoning.

Ironically, the people least able to make a decision are also the people best equipped to make the most unbiased one.

You are more likely to rely on ease of recall when you’re happy

Your mood is a significant predictor of whether ease-of-recall will affect your judgment.

People who are in a good mood are more likely to depend on how easy something is to recall, while people who are sad are more likely to think about the actual content of what they are recalling–not how easy it is to recall.

In a study, people were grouped by their mood: a happy group and a sad group. Then, each group was split again. The groups were asked to produce either two arguments or ten arguments for reducing the number of years in school from thirteen to twelve.

As expected, producing more arguments made both groups less likely to agree that the number of years of schooling should be reduced. Happy or sad, thinking of ten arguments is more difficult. It’s not the actual arguments that make people disagree, it’s the fact that producing them is hard.

What was surprising, however, was that this tendency was more pronounced when people were happy. While everyone found producing more arguments harder than producing few arguments, happier people found it harder—which made happy people less likely than sad people to agree that the number of years of schooling should be reduced. People who were sad or depressed were less likely to be as adamant in their disagreement that reducing the number of years of schooling was a good idea.[45]

We saw earlier that negative events are more likely to be recalled than positive events. Now, we see that a good mood makes us more likely to rely on ease-of-recall. In a twisted way, being happy makes us more susceptible to the ease-of-recall bias, and this bias predisposes us to recall negative events over positive events—which changes our mood from happy to sad.

This is why you can go on Facebook in a good mood, where you’re likely to use the availability heuristic to determine all your friends are having more fun without you, which then puts you in a bad mood.

How to overcome the availability heuristic

Let’s review what we’ve covered so far.

If you’re a happy multitasking novice in a position of leadership, you’re more likely to recall events that are frequent, extreme, negative, recent, and vivid. You’re more likely to rely on the availability heuristic. You look out for sharks when you should be running from coconuts.

But if you’re a sad, focused expert in a submissive role, then you are more likely to rely solely on frequency data when making decisions. You’re less likely to rely on the availability heuristic. You wear a helmet to the beach and swim with confidence.

Remember, the availability heuristic replaces frequency data with data that come to mind more easily. It compromises your judgment. As a result, you overreact and overcorrect, and you suffer more from the consequences of misjudgment.

How do we avoid falling for the availability heuristic?

There are two ways.

The first is to be aware of—and avoid—the conditions that make you more likely to rely on the availability heuristic. If you’ve just received great news that puts you in a good mood, avoid making a big decision. Were you just promoted? Make sure you evaluate input from your direct reports as much as possible. Are you an expert in your field? If not, make your decision slowly and carefully—or better yet, go find some experts to lend a hand.

Our brain is wired to find the optimal balance between speed and accuracy, which means you’ll never be able to completely overcome the availability heuristic. It’s part of human nature.

There’s another solution—a better, faster, easier one.

The solution is to count.

If the availability heuristic compromises your ability to use frequency data when making a judgment, then the solution isn’t too difficult: rely on frequency data.

How many sharks attack people each year? Are you more likely to die from a terrorist attack while flying, or are you more likely to die if you drive to grandma’s for Thanksgiving? What is your risk for cancer? Illness? Acts of violence? What’s your risk for experiencing a natural disaster, like a flood, tornado, or earthquake—and should you buy insurance as a safeguard? Are you the kind of person who is assertive or passive? Do you eat out a lot or a little? Should you worry about arsenic levels in your drinking water?

The answers to all of these questions are knowable and accessible.

Moreover, when you’re making decisions on behalf of others, you have a responsibility to get it right. Steven Pinker writes in Enlightenment Now that “a quantitative mindset, despite its nerdy aura, is, in fact, the morally enlightened one, because it treats every human life as having equal value rather than privileging the people who are closest to us or most photogenic.”[46]

You owe it to yourself and to others to avoid making decisions informed by how easy it is to call to mind relevant factors for making those decisions. Adopting a “quantitative mindset” is the way to do this.

Get more articles about cognitive biases, consumer psychology, and human behavior.

Sign up (it’s free!)


[1] Tversky, A. and Kahneman, D. (1973). “Availability: A Heuristic for Judging Frequency and Probability.” Cognitive Psychology 5, 207–232.

[2] Barass, P. (1984). “Injuries due to falling coconuts.” The Journal of Trauma 24(11):990-1.

[3] Michel-Kerjan, E., de Forges, S., and Kunreuther, H. (2012). “Policy Tenure Under the U.S. National Flood Insurance Program (NFIP).” Risk Analysis, Vol. 32, No. 4, 2012.

[4] Michel-Kerjan, E., de Forges, S., and Kunreuther, H. (2012). “Policy Tenure Under the U.S. National Flood Insurance Program (NFIP).” Risk Analysis, Vol. 32, No. 4, 2012.

[5] Palm, R. (1995). Earthquake Insurance: A Longitudinal Study of California Homeowners. Boulder, Colorado: Westview Press, 111.

[6] Gallagher (2014). “Learning about an Infrequent Event: Evidence from Flood Insurance Take-up in the US.” American Economic Journal: Applied Economics, vol 6, no. 3, pp. 206-33.

[7] Shelor, R., Anderson, D., and Cross, M. (1992). “Gaining from Loss: Property-Liability Insurer Stock Values in the Aftermath of the 1989 California Earthquake.” The Journal of Risk and Insurance, Vol. 59, No. 3 pp. 476-488

[8] National Comorbidity Survey (NCA): 2003, ‘Lifetime Prevalence of DSM-IV/WMH-CIDI Disorders by Sex and Cohort.

[9] Regier, D. A., W. E. Narrow, D. S. Rae, R. W. Manderscheid, B. Z. Locke and F. K. Goodwin: 1993, “Epidemiologic Catchment Area Prospective 1-Year Prevalence Rates of Disorders and Services,” Archives of General Psychiatry 50(2), 85–94.

[10] Regier, D. A., W. E. Narrow, D. S. Rae, R. W. Manderscheid, B. Z. Locke and F. K. Goodwin: 1993, “Epidemiologic Catchment Area Prospective 1-Year Prevalence Rates of Disorders and Services,” Archives of General Psychiatry 50(2), 85–94.

[11] Pinker, S. (2018). Enlightenment Now: The Case for Reason, Science, Humanism, and Progress. New York, NY: Viking.

[12] Barber, B. and Terrance O. (2007). “All that Glitters: The Effect of Attention and News on the Buying Behavior of Individual and Institutional Investors.” Review of Financial Studies.

[13] Frieder, Laura, 2003, “Evidence of Behavioral Biases in Trading Activity,” Unpublished Paper, UCLA, The Anderson School.

[14] Jarell, Gregg and Sam Peltzman, 1985, “The Impact of Product Recalls on the Wealth of Sellers,” The Journal of Political Economy, 93, 512-536.

[15] Sunstein, C. R., & Zeckhauser, R. (2011). “Overreaction to fearsome risks.” Environmental and Resource Economics, 48 (3), 435–449.

[16] Garrick Blalock, Vrinda Kadiyali & Daniel H. Simon (2011). “Driving fatalities after 9/11: a hidden cost of terrorism.” Applied Economics, 41:14, 1717-1729.

[17] Johnston, W. M., & Davey, G. C. (1997). “The psychological impact of negative TV news bulletins: The catastrophizing of personal worries.” British Journal of Psychology, 88(1), 85-91.

[18] Chou, H. and Edge, N. (2012), “‘They are happier and having better lives than I am’: The impact of using Facebook on perceptions of others’ lives.” Cyberpsychology, Behavior, and Social Networking 15(2) 117-121.

[19] Schwarz, N., Bless, H., Strack, F. Simons, A. (1991). “Ease of retrieval as information: Another look at the availability heuristic.” Journal of Personality and Social Psychology 61(2).

[20] Kahneman, D. (2011). Thinking, Fast and Slow. New York: Farrar, Straus and Giroux.

[21] Kahneman, D. (2011). Thinking, Fast and Slow. New York: Farrar, Straus and Giroux.

[22] Winke, M., Bless, H., & Biller, B. (1996). “Subjective experience versus content of information in the construction of attitude judgments.” Personality and Social Psychology Bulletin.

[23] Fox, C.R. (2006). “The availability heuristic in the classroom: How soliciting more criticism can boost your course
ratings.” Judgment and Decision Making, 1, 86-90

[24] Wanke, M., Bohner, G., & Jurkowitsch, A. (1997). “There Are Many Reasons to Drive a BMW: Does Imagined Ease of Argument Generation Influence Attitudes?” Journal of Consumer Research vol. 24.

[25] Xu, J. & Schwarz, N. (2005). “Was it long ago or unimportant? Diverging inferences from
difficulty of recall”. Paper presented at Annual Conference of Society for Consumer Psychology, St.
Pete’s Beach, Florida.

[26] Attneave, F. (1953). “Psychological probability as a function of experienced frequency.” Journal of Experimental Psychology 46 (2):81.

[27] Lichtenstein, Slovic, Fischhoff, Layman, & Combs, 1978. “Judged frequency of lethal events.” Journal of Experimental Psychology: Human Learning and Memory, 4 (6), 551–578.

[28] Hansen and Hansen (1988). “Finding the face in the crowd: an anger superiority effect.” Journal of Personality and Social Psychology, 54(6), 917-924.

[29] Peeters, G. & Czapinksi, J. (1990). “Positive-Negative Asymmetry in Evaluations: The Distinction Between Affective and Informational Negativity Effects.” European Review of Social Psychology 1(1).

[30] Baumeister, R., Bratslavsky, E., Finkenauer, C., &Vohs, K. (2001). “Bad is stronger than good.” Review of General Psychology, Vol 5(4), Dec 2001, 323-370.

[31] Baumeister, R., Bratslavsky, E., Finkenauer, C., &Vohs, K. (2001). “Bad is stronger than good.” Review of General Psychology, Vol 5(4), Dec 2001, 323-370.

[32] Gottman, J.M. (1994). What predicts divorce? The relationship between marital processes and marital outcomes. New York: Psychology Press.

[33] Finkenauer, C. and Rimé, B. (1998). “Socially shared emotional experiences vs. emotional experiences kept secret: Differential characteristics and consequences.” Journal of Social and Clinical Psychology Vol. 17, No. 3, pp. 295-318.

[34] Baumeister, R., Bratslavsky, E., Finkenauer, C., &Vohs, K. (2001). “Bad is stronger than good.” Review of General Psychology, Vol 5(4), Dec 2001, 323-370.

[35] Matlin, M. W. , & Stang, D. J. (1978). The Pollyanna Principle: Selectivity in language, memory, and thought. Cambridge, MA: Schenkman.

[36] Baumeister, R., Bratslavsky, E., Finkenauer, C., &Vohs, K. (2001). “Bad is stronger than good.” Review of General Psychology, Vol 5(4), Dec 2001, 323-370.

[37] Walker, W., Vogl, R., & Thompson, C. (1997). “Autobiographical memory: unpleasantness fades faster than pleasantness over time.” Applied Cognitive Psychology, 11(5), 399–413.

[38] Raghubir, P. & Menon, G. (2005). “When and why is ease of retrieval informative?” Memory and Cognition 33(5), pp.821–832.

[39] Sunstein, C. R., & Zeckhauser, R. (2011). “Overreaction to fearsome risks.” Environmental and Resource Economics, vol. 48, 435–449.

[40] Dyson, F. (2011). “How to dispel your illusions.” New York Review of Books.

[41] Craik, F., Govoni, R., Naveh-Benjamin, M., & Anderson, N.D. (1996). “The effects of divided attention on encoding and retrieval processes in human memory.” Journal of Experimental Psychology 125(2), 159–180.

[42] Ofir, C. (2000). “Ease of Recall vs Recalled Evidence in Judgment: Experts vs Laymen.” Organizational Behavior and Human Decision Processes 81(1), 28–42.

[43] Schwarz, N., Bless, H., Wänke, M., & Winkielman, P (2003). Accessibility revisited. In Bodenhausen , G. and Lambert, A., (Eds.), Foundations of Social Cognition. Erlbaum.

[44] Weick, M. & Guinote, A. (2008), “When Subjective Experiences Matter: Power Increases Reliance on the Ease of Retrieval.” Journal of Personality and Social Psychology, Vol 94(6), June 2008, 956-970.

[45] Ruder, M. and Bless, H. (2003). “Mood and the reliance on the ease of retrieval heuristic.” Journal of Personality and Social Psychology 85(1), Jul 2003, 20-32.

[46] Pinker, S. (2018). Enlightenment Now: The Case for Reason, Science, Humanism, and Progress. New York, NY: Viking.

Advertisements