Anchoring is the tendency for an initial piece of information — the anchor — to influence subsequent estimates by biasing them toward the anchor value. Tversky and Kahneman (1974) demonstrated this by spinning a wheel of fortune (producing a random number) and then asking whether the percentage of African nations in the UN was higher or lower than that number. The random anchor dramatically influenced estimates: those who saw a high number gave higher estimates than those who saw a low number.
Mechanisms
Two mechanisms have been proposed. Insufficient adjustment: people start from the anchor and adjust their estimate, but adjustment is typically insufficient (they stop adjusting too soon). Selective accessibility: the anchor activates anchor-consistent information in memory, biasing the judgment toward anchor-consistent values. Evidence supports both mechanisms, operating in different situations: adjustment from self-generated anchors, and selective accessibility for externally provided anchors.
Pervasiveness
Anchoring effects are remarkably robust. They persist even when people are warned about anchoring, when they are offered incentives for accuracy, and when the anchor is obviously arbitrary. They affect expert judgments as well as novice ones: real estate agents are anchored by listing prices, judges are anchored by sentencing requests, and physicians are anchored by initial diagnostic hypotheses. The pervasiveness and resistance to debiasing make anchoring one of the most practically important cognitive biases.
Anchoring has powerful effects in negotiation. The first offer serves as an anchor that influences the final agreement. Research consistently shows that higher first offers lead to higher settlement prices, even when the initial offer is extreme. Understanding anchoring has become a standard component of negotiation training, with the recommendation to either make the first offer (to set a favorable anchor) or to explicitly counter-anchor if the other party goes first.