The availability heuristic is a cognitive bias in which you base a decision on information that is readily available to you (Tversky & Kahneman, 1973). This may not always be the best example to inform your decision.
In other words, easily recalled information is assumed to reflect more frequent or probable events, while harder-to-recall information is assumed to reflect rarer ones. The rule cuts both ways.
Consider, for example, a person trying to estimate the relative probability of owning a dog versus owning a ferret as a household pet. In all likelihood, it is easier to think of an example of a dog-owning household than it is to think of an example of a ferret-owning household.
Therefore, this person may reasonably conclude that dogs are the more common household pet. The example is simply easier to recall.
Frequent events are usually easier to recall than rare ones. This mental shortcut regularly leads to fast, accurate judgments in real-world scenarios (Markman & Medin, 2002).
However, the availability bias is also prone to predictable errors in certain situations and thus is not always a reliable shortcut for decision-making.
Historical Background
- Rational-actor model: In the early 1900s, economic research assumed people were entirely rational decision-makers who could accurately weigh every option and choose optimally. Under this model, systematic errors of judgment were both unexpected and unexplainable (Gilovich et al., 2002).
- Bounded rationality: Herbert Simon’s theory of bounded rationality proposed that people are rational only within the limits of their own cognitive capacity. Because attention, memory, and computation are finite, decision-makers cannot assess every option exhaustively and instead work with the most accessible and relevant information (Simon, 1955).
- A field is born: Bounded rationality made room for the idea that the shortcuts people actually use could themselves be studied. Psychologists Amos Tversky and Daniel Kahneman produced the field’s most influential early work, part of a wider framework of heuristics and cognitive biases within behavioral economics (American Psychological Association).
- Availability, then the trio: Tversky and Kahneman introduced availability in a dedicated 1973 paper, then presented it alongside the representativeness and anchoring/adjustment biases in their landmark 1974 Science review on judgment under uncertainty (Tversky & Kahneman, 1973; Tversky & Kahneman, 1974). That work remains widely cited by behavioral economists today.
How the Availability Heuristic Works
The human brain is eager to use whatever information it can to make good decisions. However, obtaining all relevant information in decision-making scenarios is not always easy, nor even possible.
And even in situations in which all relevant information is available, analyzing all potential options and outcomes is computationally expensive.
Therefore, the brain takes frequent and predictable shortcuts. The availability bias – in which the prevalence and likelihood of an event are estimated by the ease with which relevant examples can be recalled – is one such mental shortcut.
How the Mechanism Works
Two features of this mechanism matter. First, availability is a feeling, not a count: people do not need to actually retrieve any examples to sense availability.
Shown two jumbled letter sets and asked which could form more English words, most people answer almost immediately, without listing a single word (Kahneman, 2011). The mere promise of easy retrieval is itself the signal.
This is why the availability bias can be manipulated by anything that changes the felt ease of recall, quite apart from true frequency.
Second, the error is directional, not random. Whenever a class of events is easier to recall than an equally frequent class, its frequency gets over-estimated relative to the quieter class.
This happens whenever something is dramatic, recent, emotionally engaging, heavily reported, or personally experienced (Kahneman, 2011). The heuristic misleads exactly where recall and reality diverge, never at random.
This usually works in our favor. Thus, this heuristic allows people to make fast and accurate estimations in many real-world scenarios.
There are, however, certain predictable moments in which less frequent events are easier to recall than more frequent ones, and the availability bias errs (Markman & Medin, 2002). These are the examples explored below.
Markman and Medin (2002) compare this to another useful system of shortcuts that occasionally leads to faulty judgment: the human visual system.
The human perceptual system is refined and extremely useful. However, the same shortcuts that make it efficient also make it prone to error in the case of optical illusions.
Examples
Here are a few scenarios where this could play out in your day-to-day life.
Winning the Lottery
Hundreds of millions of people participate in lotteries every year, and yet by definition, very few are successful. So why do people continue to play?
The availability bias can help to explain why people have an unfortunate tendency to severely misjudge their personal probability of winning the lottery.
The probability of winning the Powerball jackpot lottery is approximately 1 in 300 million (Victor, 2016).
Lottery winners are easy to picture, winnings and all. Losers, and their lack of winnings, are not. This imbalance leads people to subconsciously believe that winning the lottery is a far more likely occurrence than it actually is (Griffiths & Wood, 2001; Kahneman, 2011).
Safety
It is common for people to overestimate the risk of dramatic events, such as plane crashes, shark attacks, and terrorist attacks. Car crashes and cancer, quieter risks, are underestimated by comparison.
For example, many people are warier of flying than driving, and some choose to drive rather than fly out of safety concerns.
In reality, driving the same distance is about 65 times riskier than flying (Sivak & Flannagan, 2003). Flying is safer.
Shark attacks are another common fear, despite being extremely rare.
The risk of death by shark attack is over 1 in 3.7 million. Death by lightning strike, also very rare, is about 47 times more likely (“Risk of death,” 2018).
The pattern goes beyond sharks. Lichtenstein et al. (1978) confirmed it formally, comparing judged frequency against actual mortality statistics for 41 causes of death. Dramatic causes such as accidents and tornadoes were overestimated, while diseases such as asthma and diabetes were underestimated despite killing far more people.
This overestimation often traces back to sensationalized, memorable media coverage.
Car crashes rarely make headlines. They stay less available in memory as a result, even though they are more common (Kahneman, 2011).
The availability bias even shapes public spending. Cancer kills far more Americans than terrorism does. Yet cancer research receives only a tiny fraction of the funding directed toward defense and the military (“Federal spending,” n.d.; “Risk of death,” 2018).
Insurance Rates
After natural disasters (i.e., floods), it has been observed that related insurance rates (i.e., the rate at which consumers purchase flood insurance) spike in affected communities.
It can be reasoned that the experience of disaster causes community members to reevaluate their perceived risk of danger and to protect themselves accordingly.
Insurance rates, however, decline steadily in the years that follow. This happens even though the underlying disaster risk in the community stays the same throughout (Gallagher, 2014).
It is not only the disaster’s risk that matters here. The ease with which the experience comes to mind also shapes a community member’s decision to buy protective insurance.
Recent disasters are easier to recall. So communities tend to overestimate the risk of a repeat event in the years right after it happens.
Older disasters fade from memory. As a result, communities underestimate the risk of a repeat event several years later.
This overestimation-then-underestimation pattern is the availability bias at work. It explains the spiking and declining insurance rates seen in disaster-struck communities (Gallagher, 2014; Kahneman, 2011).
Self-Evaluation
Schwarz et al. (1991) tested this with a clever study on self-perceptions of assertiveness.
Aim: To determine whether availability-based judgments rest on the amount of content people recall or on the subjective ease of recalling it.
Method: Participants were asked to recall either six or twelve examples of their own assertive behavior, then rate their own assertiveness on a scale from one to ten. Six are easy to produce; twelve are a genuine struggle.
Results: Participants who recalled six assertive behaviors rated themselves as significantly more assertive than those who recalled twelve, even though the twelve-example group had generated twice the supporting evidence.
Conclusion: Self-judgments moved with the felt difficulty of recall, not the amount of evidence recalled. The struggle itself was the signal. Twelve examples felt like proof that assertive behavior must be rare.
Evaluation: The study made the amount recalled and the ease of recalling it point in opposite directions. This showed which one the mind actually consults for self-judgment (Schwarz et al., 1991).
Course Evaluation
Fox (2006) replicated this logic in a real institutional setting: course evaluations.
Aim: To test whether asking for more criticism of a course could paradoxically raise its ratings, as ease-of-retrieval predicts.
Method: In a graduate business course at Duke University, students completing a mid-course evaluation were asked to list either two ways the course could improve, or ten. Two is easy; ten is hard.
Results: Students asked for ten improvements, a difficult task, rated the course more favorably than students asked for only two. The struggle was itself the evidence. Fewer complaints must mean less wrong with the course.
Conclusion: Ease of retrieval, not the number of criticisms generated, drove the ratings. This replicated Schwarz et al. (1991) in a real field setting with genuine stakes (Fox, 2006).
Word Frequency
In one of their earliest studies, Kahneman and Tversky asked people a simple question: do more English words begin with the letter K, or have K as their third letter? Think about it for a moment.
Aim: To test whether frequency judgments follow the ease of memory search, using a case where the easier search gives the objectively wrong answer.
Method: Participants judged whether K was more likely to appear first or third in a word. They also estimated the ratio.
Results: Words are mentally indexed by their first letter, so searching for words beginning with K is far easier than searching for words with K third. Most guessed wrong. About two-thirds of participants judged the first position more frequent, typically estimating it at roughly twice as common.
Conclusion: The true ratio runs the other way: there are about twice as many words with K as the third letter. The structure of memory search, not the structure of language, drove the judgment (Tversky & Kahneman, 1973; Tversky & Kahneman, 1974).
Implications
Though the availability bias often leads to accurate judgments in a range of real-world scenarios, it is still prone to error in certain predictable situations.
In these situations, the use of availability bias can lead to faulty judgment. These errors in judgment can have a significant and rapid impact on human behavior, sometimes with negative consequences.
Politics, Marketing, and Availability Cascades
Politicians and marketers exploit availability deliberately. By overemphasizing certain issues, threats, or an opposing candidate’s flaws, politicians can make people believe these things are more frequent and relevant than they actually are.
Marketing companies use the same logic. By overemphasizing the downsides of not buying a product, they convince customers that their need for it is greater than it really is.
These individual tactics can snowball into a society-wide pattern. Kuran and Sunstein (1999) describe the availability cascade: a media story about a minor risk catches public attention, the resulting concern is itself newsworthy, and further coverage follows.
Activists, politicians, and journalists can amplify the alarm further still.
The spiral of coverage and concern can end up dominating public policy far out of proportion to the actual hazard (Kuran & Sunstein, 1999).
Evaluation, Education, and Social Media
As seen in the assertiveness study by Schwarz et al. (1991), the availability bias can impact students’ evaluations of their own assertiveness (see ‘Examples – Self-Evaluation’). This effect likely extends to other character-trait evaluations too.
Memorable interactions with others also distort judgment. A person who is particularly rude, or particularly clumsy, on one occasion can make that trait seem more common in them than it really is.
As seen in the course evaluation study by Fox (2006), the availability bias can impact students’ evaluations of their own education (see ‘Examples – Course Evaluation’).
Because the effect showed up on a single mid-course survey, it demonstrates how quickly and efficiently the availability bias can work.
The pattern holds well beyond the classroom. Social media feeds tend to over-sample other people’s highlights, such as holidays and celebrations, more than their ordinary or sad moments. Viewers may overestimate others’ happiness, underestimating their own by comparison.
The same over-sampling effect applies to alarming content, feeding the availability cascades described above. Repeated exposure to a claim also breeds belief in it, one reason availability and confirmation bias compound each other.
Overcoming the Availability Bias
In many cases, the availability bias leads to accurate frequency and probability estimates. It is neither realistic nor possible to eliminate it entirely.
Its negative effects, however, can be softened. Remember to weigh all the relevant data when judging under uncertainty, not just what comes readily to mind.
Simply recognizing the predictable situations where the availability bias tends to err makes for a more thoughtful decision-maker. You are already one step ahead just by reading this.
Critical Evaluation
The availability bias faced critical evaluation by Schwarz et al. (1991) for being ambiguous in terms of its specific underlying process.
As covered under Self-Evaluation, Schwarz et al. (1991) showed that the availability bias operates on the ease of recall, not the amount of content recalled. Participants who found the task easy rated themselves as more assertive than those who found it hard, despite recalling fewer examples.
Contemporary Research
Half a century on, the availability heuristic has been stress-tested about as thoroughly as any idea in psychology. The core effect holds up. But it is smaller and more conditional than the textbook version suggests.
Aim: Weingarten and Hutchinson (2018) set out to establish how reliable, and how large, the classic ease-of-retrieval effect really is.
Method: The meta-analysis pooled 142 papers, 263 studies, and 582 effect sizes across the whole published and unpublished literature, testing whether the effect is really driven by felt ease. Publication bias and mediation were formally modelled.
Results: The standard manipulation produced a medium-sized effect on average, but publication bias inflated the naive estimate by up to a third. Felt ease explained only part of it. The rest came from other factors, like the content people actually recalled.
Conclusion: The effect is real and practically meaningful, but the classic demonstrations overstate its typical strength (Weingarten & Hutchinson, 2018).
But the story does not end there. Later work has also split availability into distinct, separately measurable mechanisms.
Hertwig et al. (2005) and Pachur et al. (2012) compared candidate explanations for real risk judgments.
Instances recalled from one’s own social network best predicted how frequent or fatal a risk seemed. Emotion mattered too. It contributed more to judgments of personal risk than to judgments of a risk’s overall frequency.
People are not always passive victims of the bias, either. Oppenheimer (2004) found that participants spontaneously judged famous surnames as less frequent than equally common non-famous ones. They correctly discounted fluency once its true source, fame rather than commonness, was obvious.
Related Cognitive Biases
The availability bias is one of several cognitive biases, or mental shortcuts, used in judgment-making scenarios. Two others stand out: representativeness and anchoring/adjustment.
These three biases were the primary focus of Tversky and Kahneman’s seminal work on judgment under uncertainty (Tversky & Kahneman, 1974). Each remains central to how we discuss decision-making today.
Each bias has a distinct definition and its own common examples of use and error. Even so, two or more biases can operate together in the same decision.
A decision is rarely shaped by just one bias; more often, several work together. That’s the norm, not the exception.
Individual differences and emotional response also shape the human decision-making process (Payne et al., 1993; Slovic et al., 2007).
A full discussion would need a much longer article.
Still, the availability, representativeness, and anchoring/adjustment biases each offer real insight into how the human mind works when judging under uncertainty.
Representativeness Bias
The representativeness bias, also called the representativeness heuristic, is a common cognitive shortcut for judging probability. It estimates the likelihood of an occurrence by how closely it resembles a typical example (Tversky & Kahneman, 1974).
The more an example resembles our preconceived idea of a typical case, the more likely it seems. The reverse is true too.
A classic example concerns randomness. Consider a coin-toss sequence, where H is heads and T is tails. The sequence H-T-T-H-T-H feels more likely than H-H-H-T-T-T, because it looks more like our idea of randomness.
In reality, every toss has the same 50% chance. So both six-toss sequences are exactly as likely as each other, about 1.5% each (Tversky & Kahneman, 1974).
Anchoring/Adjustment Bias
The anchoring/adjustment bias, also called the anchoring/adjustment heuristic, is a common cognitive shortcut for making evaluations and estimations. Assessments are made by adjusting from an initial reference point, or anchor.
This adjustment is often insufficient. It happens even when the anchor is unrelated to the estimate (Tversky & Kahneman, 1974).
In other words, people tend to overvalue initial information, regardless of relevance, when making evaluations and estimations. Consider a retail item that costs $100.
The $100 price feels more reasonable if it is on sale from an original price of $200 than if it recently rose from $50 to $100. It also feels more reasonable than if the price had simply stayed at $100 the whole time.
The final price is identical in each scenario. Yet the perceived reasonableness varies considerably, because the initial price serves as a mental anchor.
This happens even when the anchor is irrelevant. Kahneman and Tversky (1974) demonstrated this by asking subjects about the percentage of African countries in the United Nations.
One group was told the true figure might be higher or lower than 65%; another group was anchored to 10%. They then asked each subject for an exact estimate.
Subjects anchored to 65 gave significantly higher estimates for the percentage of African countries in the UN than subjects anchored to 10. Median estimates were 45% and 25%, respectively.
Key Takeaways
- Availability Heuristic: People judge how likely or frequent something is by how easily examples come to mind, not by actual counts (Tversky & Kahneman, 1973).
- Usually Accurate: Because common things are genuinely easier to recall, the shortcut is fast and often correct.
- Predictable Errors: It fails whenever something other than frequency, such as drama, recency, or personal experience, makes events easier or harder to recall.
- Ease, Not Count: Judgments track the felt ease of recall, not the number of examples actually retrieved (Schwarz et al., 1991).
- Real-World Reach: It shapes fears about flying and sharks, insurance-buying after disasters, self-ratings, and even public spending.
- Modern Evidence: A 2018 meta-analysis of over 250 studies confirms the effect is real, but smaller than the textbook version suggests (Weingarten & Hutchinson, 2018).
References
American Psychological Association. (n.d.). APA Dictionary of Psychology. https://dictionary.apa.org/behavioral-economics
Federal spending: Where does the money go. (n.d.). National Priorities Project. https://www.nationalpriorities.org/budget-basics/federal-budget-101/spending/
Fox, C. R. (2006). The availability heuristic in the classroom: How soliciting more criticism can boost your course ratings. Judgment and Decision Making, 1(1), 86-90.
Gallagher, J. (2014). Learning about an infrequent event: Evidence from flood insurance take-up in the United States. American Economic Journal: Applied Economics, 6(3), 206-233.
Gilovich, T., Griffin, D., & Kahneman, D. (Eds.). (2002). Heuristics and biases: The psychology of intuitive judgment. Cambridge university press.
Griffiths, M. D., & Wood, R. T. A. (2001). The psychology of lottery gambling. International Gambling Studies, 1(1), 27-45.
Hertwig, R., Pachur, T., & Kurzenhäuser, S. (2005). Judgments of risk frequencies: Tests of possible cognitive mechanisms. Journal of Experimental Psychology: Learning, Memory, and Cognition, 31(4), 621-642.
Kahneman, D. (2011). Thinking, fast and slow. Macmillan.
Kuran, T., & Sunstein, C. R. (1999). Availability cascades and risk regulation. Stanford Law Review, 51(4), 683-768.
Markman, A. B., & Medin, D. L. (2002). Decision making.
Oppenheimer, D. M. (2004). Spontaneous discounting of availability in frequency judgment tasks. Psychological Science, 15(2), 100-105.
Pachur, T., Hertwig, R., & Steinmann, F. (2012). How do people judge risks: Availability heuristic, affect heuristic, or both? Journal of Experimental Psychology: Applied, 18(3), 314-330.
Payne, J. W., Bettman, J. R., & Johnson, E. J. (1993). The adaptive decision maker. Cambridge university press.
Risk of death. Florida Museum. (2018). https://www.floridamuseum.ufl.edu/shark-attacks/odds/compare-risk/death/
Schwarz, N., Bless, H., Strack, F., Klumpp, G., Rittenauer-Schatka, H., & Simons, A. (1991). Ease of retrieval as information: Another look at the availability heuristic. Journal of Personality and Social Psychology, 61(2), 195–202.
Simon, H. A. (1955). A behavioral model of rational choice. The Quarterly Journal of Economics, 69(1), 99–118.
Sivak, M., & Flannagan, M. J. (2003). Macroscope: flying and driving after the september 11 attacks. American Scientist, 91(1), 6-8.
Slovic, P., Finucane, M. L., Peters, E., & MacGregor, D. G. (2007). The affect heuristic. European journal of operational research, 177(3), 1333-1352
Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive psychology, 5(2), 207-232.
Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124–1131.
Victor, D. (2016). You will not win the powerball jackpot. The New York Times. https://www.nytimes.com/2016/01/13/us/powerball-odds.htm
Weingarten, E., & Hutchinson, J. W. (2018). Does ease mediate the ease-of-retrieval effect? A meta-analysis. Psychological Bulletin, 144(3), 227-283.
