This article is at an early stage of development.
Anchoring is our cognitive bias towards being unduly influenced by an initial piece of information when forming later judgements - that is, our later judgements tend to be somewhat anchored to that initial information. For example, if we see a car in a showroom with its price displayed, but then a salesperson informs us that there's 10% off the display price for that day only, then we’re biased towards concluding that this is a good deal, even though the offer price may actually be above the car's market value. That is, our judgement is somewhat anchored to the initial information - the displayed price.
Anchoring is actually a form of availability bias, and can therefore be explained by credulism, the certainty of belief, and the speed of the brain. In the example, as explained, the offer price may be above the car's market value. However, upon being informed of the offer by the salesperson, this information is thereby readily available to our thought processes, whereas the car’s market value isn’t, assuming that we don’t know that value. And 10% off the normal price is, for a relatively expensive purchase such as a car, obviously a significant saving. Therefore, the possibility that the offer is a good deal will likely enter our mind, via our reasoning or imagination, within a fraction of a second of learning of the offer, given the speed of the brain. And, given credulism, this possibility enters our thought processes as a certainty, and, as explained in Confirmation bias - in the section Belief is self-preserving - the certainty of belief then effectively provides a degree of protection to our belief, directly and via the confirmation bias. Specifically, while that certainty exists, we’re unlikely to question our interpretation, and are therefore biased against thinking of the more complex possibility that the offer price is above the car’s market value. Thus, we’re biased, due to availability bias, towards interpreting the offer as a good deal.
Another illustration of anchoring is a classic experiment conducted in 1974. US subjects witnessed a ‘wheel of fortune’ being used to apparently randomly select a number from 1 to 100, although the wheel was actually rigged to stop at either 10 or 65. The subjects were then asked whether they thought that the selected number was higher or lower than the percentage of African countries that were members of the United Nations. Upon giving their answer, they were then asked to estimate that percentage. When the wheel had stopped at 10, the average estimate was 25%, but when it had stopped at 65, the average estimate was 45%. Therefore, the subjects’ reasoning, in answering the second question, had tended to be somewhat anchored to the initial information - the selected number - even though they believed that that information was randomly generated. To understand why, consider the design of the experiment.
Given the subjects' likely lack of knowledge about the membership of African countries of the UN, their judgement regarding the first question will have likely merely been that the selected number was likely, rather than certainly, higher or lower than the true percentage. That is, their judgement will have likely involved there being at least some possibility that the selected number was the true percentage. And although the subjects would have likely concluded this for many other specific percentages if they'd considered them, the experiment didn’t require them to consider any other specific percentages. Therefore, when they were then asked to estimate the true percentage, it was likely not only that the specific percentage based on the selected number had, for them, at least some credibility, but also that they weren't thinking this of any other specific percentage. That is, when they were forming that estimate, the likely credibility of the specific percentage based on the selected number tended to be much more readily available to their thought processes than that of other specific percentages. And the likely credibility of that specific percentage also tended to be more readily available to their thought processes than the apparently random origin of that specific credibility, given that its formation, in the subjects’ minds, occurred more recently than the apparently random selection of the number on which the specific percentage was based. Therefore, given availability bias, the subjects’ estimates of the true percentage would have tended to be biased towards the specific percentage based on the apparently randomly selected number. That is, given availability bias, their estimates tended to be somewhat anchored to that number.
- Tversky, A., and Kahneman, D., 1974, 'Judgment under Uncertainty: Heuristics and Biases', Science, volume 185, number 4157, pages 1124–31.