People tend to rate the frequency of an occurrence or the probability of an event by the ease with which you can remember it. Availability is usually an effective heuristic tool because if you can easily remember an event, there is a chance that you just experienced it or were exposed to it. However, just as with similarity and probability, availability is affected by other factors as well, namely in the form of biases.
Biases Due to the Retrievability of Instances
When the size of a class is judged by the availability of instances, a person is displaying a bias due to the retrievability of instances. A class that is difficult to remember is likely to be judged as having fewer members. Salience and recency are largely biased by how easy them come to mind. In 1973, Galbraith and Underwood showed that because it is easier to think of abstract words used in various contexts compared to concrete words, participants judged the frequency of the word occurrence of abstract words and higher than concrete words. Even though the concrete words are more common, abstract words are not contextually constrained so they appear more salient.
Another example is evident from the Lichenstein et al. 1978 study. They asked participants to rate the likelihood of particular causes of death. He found that participants believed that accidents caused as many deaths as disease, and that murder was more common than suicide. In reality, diseases cause sixteen times as many deaths compared to accidents, and suicides are almost twice as common as murders. Availability heuristics suggest that murders and accidents are common because media coverage is far greater. As various types of media are far more accessible to us compared to disease or suicide statistics, information spread by the media becomes far more salient. Hence, the more salient information often translates into a false perception of its occurrence.
Biases of Imaginability
The imaginability of biases is a heuristic when one has to assess the likelihood of an event not using instances stored in your memory but according to a particular rule or task. For example, consider a group about to embark on an expedition together. As part of the preparation they must consider all possible difficulties they might encounter. Based on this “rule” the ease it takes them to recall possible difficulties in the wake of their trip seem more likely than they actually are. Again, discussing instances of difficulty heightens salience to a point of interference.
Chapman and Chapman (1967) described illusory correlation as the judgment of the frequency with which two events co-occur. The same year, Chapman and Chapman carried out a study investigating the strength of illusory correction. Participants were presented with several hypothetical mental patients. Each patient came with a clinical diagnosis and a drawing of a person made by that patient (Draw-a-person test). The results showed that participants overestimated the frequency of co-occurrence of “natural associates” such as suspiciousness and peculiar eyes; a finding remarkably similar to the overall clinical reports of the same task. Despite being presented with contradictory data, the illusory correlation remained resistant to the point of preventing participants from detecting relationships that were present.
As the name suggests, with recognition heuristics, a salience bias is established by how well we recognise an object, name, etc. Goldstein and Gigerenzer (2002) asked German and American students about four cities. Specifically, which city did they think was larger: San Antonio or San Diego? And Hamburg or Cologne? Contrary to expectations, the results showed that American students were more accurate for the German cities, and the German students were more accurate for the American cities. Unlike the other studies, heuristics actually benefited the students when they new little about the other country. San Diego is larger and also featured more often in films or the news; the same is true for Hamburg. Therefore, guessing that San Diego and Hamburg are the larger makes sense for foreign students because they have a very limited amount of information available to influence salience. For the natives, answering the question becomes more difficult because both cities are salient and factors such as their own familiarity or experiences are likely to interfere with their answer.
Biases Due to the Effectiveness of a Search Set
Lastly, Tverky and Kahenman (1974) describe the bias due to the effectiveness of a search set. The easiest way to describe this bias is through an example. For example, if participants are asked to list words beginning with and containing “T” it is easier to think of words starting with it than containing it; therefore, words beginning with “T” have greater salience. As has been established, greater salience means participants are likely to list far more words beginning with “T.”