Why Does the First Number Pull Later Judgements Towards It?

When looking at a used car, a buyer may keep thinking around the seller's high opening price even after negotiating it down substantially. This is the anchoring effect: an initial number becomes a reference point and pulls later estimates towards it. In Tversky and Kahneman's classic 1974 experiment, participants first encountered a number produced by a roulette wheel and then estimated the proportion of African countries in the United Nations. Different starting numbers systematically shifted the answers.

The early account emphasised anchoring and adjustment: people begin from a reference point and revise, but usually not far enough. Later research suggests that this need not be the only mechanism. An anchor can also make anchor-consistent information more accessible. A high asking price, for example, may make features supporting a high valuation easier to retrieve. Relevant anchors often exert more influence than entirely irrelevant numbers because they provide genuine information as well as bias.

This does not mean that any number spoken first can manipulate anyone. Expertise, additional evidence, anchor credibility and task design all change the effect, and an average difference in an experiment cannot predict a particular real-world transaction. A practical response is not merely to command yourself to ignore the anchor. Form an independent range before seeing the other party's number, then deliberately seek evidence that points away from it. This cannot guarantee immunity, but it reduces the chance that a starting point will replace the evidence.

https://doi.org/10.1126/science.185.4157.1124
https://doi.org/10.1177/00332941211016750