K-letter position experiment

The K-letter position experiment is a classic demonstration of the availability heuristic, in which the mind estimates frequency by the ease with which instances can be retrieved. Asked whether K is more likely to appear as the first letter of a word or as the third, people are drawn toward the answer that feels easiest to support from memory. The experiment exposes how a convenient mental shortcut can produce a systematic error even in a familiar verbal task.

The Question and Its Intuitive Pull

The task asks respondents to compare two positions for the letter K in a typical English text: the first position in a word and the third position. The question appears simple, but the two alternatives do not offer equally accessible searches. It is straightforward to generate words beginning with K, whereas finding words with K in the third position requires a less natural and more laborious search through memory.

That asymmetry makes the first-position answer feel more plausible. The judgment is not based on a deliberate count of words in a representative sample. Instead, the ease of producing examples becomes an implicit signal of frequency: what comes readily to mind is treated as though it occurs more often in the world.

Result and Cognitive Mechanism

The experiment predicted that respondents would exaggerate the frequency of letters appearing in the first position, including letters such as K, L, N, R, and V that in fact occur more frequently in the third position. The error therefore does not arise because participants lack all relevant knowledge of English words. It arises because the retrieval process is easier for one category than for the other, and that subjective ease is mistaken for statistical evidence.

The mechanism was named the availability heuristic. When people estimate the size or frequency of a category, they often search memory for relevant instances and use the fluency of that search as a proxy for prevalence. The shortcut is efficient when ease of recall reliably tracks frequency, but it becomes biased when the ease of retrieval is affected by the form of the question, the vividness of examples, or the structure of memory itself.

From Letter Counts to Social Beliefs

The same pattern extends beyond words. Daniel Kahneman and Amos Tversky noticed that a question about the divorce rate among professors at their university triggered a search for divorced professors known personally or by reputation. The number of such examples that came to mind then shaped the judgment of the category as a whole. The K task supplied a clean laboratory version of that broader process because the actual frequency of letters could be checked independently of the participants’ impressions.

A related example involved the impression that adultery was more common among politicians than among physicians or lawyers. That impression seemed to invite causal explanations, including the temptations associated with power and life away from home. It later became doubtful because political transgressions are more likely to be reported by journalists. The greater availability of news stories about politicians could therefore create the impression of a higher underlying rate, even without evidence that the conduct itself was more frequent.

Statistical Lesson and Wider Significance

The experiment belongs to a larger program that studied intuitive judgments about probabilities, future events, hypotheses, and frequencies. Its central lesson is that intuitive thinking can produce predictable biases rather than merely occasional mistakes. The mind’s shortcut is useful because it avoids exhaustive enumeration, but its answer can be systematically distorted when accessibility and actual frequency diverge.

That lesson connects the letter task to Bayesian reasoning and reference-class prediction. Evidence about a particular case should be combined with relevant base rates, rather than allowed to replace them; in the same way, the ease of recalling examples should not be treated as a direct count of how common the underlying category is. The experiment’s enduring significance lies in making an abstract error visible: a person can feel that a judgment is well supported because examples arrive effortlessly, while the true statistical pattern points in the opposite direction.