Anchoring bias, also termed focalism, describes a systematic pattern of deviation from rationality in judgement where individuals rely excessively on the first piece of information encountered when making decisions. Cognitive Bias.
This initial value, known as the anchor, serves as a psychological foundation for all subsequent estimations, even if that value is arbitrary or irrelevant to the task. The phenomenon demonstrates that human cognition often grounds numerical values on a prior point of departure rather than objective reality.
As a form of stability bias, anchoring involves the tying of actions to an initial value and a subsequent failure to adjust sufficiently to take into account new information.
Theoretical Foundations and Mechanisms
The adjustment and anchoring heuristic involves individuals making estimates by starting from an initial value that is then adjusted to yield a final answer. These adjustments are typically insufficient, leaving final estimates biased toward the initial anchor.
Susceptibility to anchoring does not correlate with intelligence or other measures of cognitive ability. While many individuals described themselves as excellent decision-makers free from bias, formal testing batteries frequently reveal poor performance regarding these cognitive traps.
Economic and Financial Implications
In the asset management sector, anchoring acts as a stability bias that undermines optimal capital reallocation. Analysis of publicly traded companies between 1990 and 2010 indicates that top performers are those that least anchor current capital decisions to those of the previous year.
Dynamic reallocators achieved median returns approximately four percentage points higher than dormant reallocators. Furthermore, professional consensus forecasts of macroeconomic releases are systematically biased toward previous months' data releases.
Recent past values of a data release act as an anchor on expert forecasts, causing them to weight the typical forecast too heavily towards its recent past by approximately 30 per cent.
Bond yields often react to the residual or unpredictable component of a surprise rather than the expected piece of the forecast error induced by anchoring, indicating that some market participants anticipate this bias in expert forecasts.
Legal and Public Sector Decision-Making
Numeric decisions in law, such as the determination of damages or prison terms, are susceptible to the influence of salient numbers present in the decision context.
Jurors and judges often anchor on the initial demand of a party, which creates a troubling effect regarding the fairness of proceedings. In criminal contexts, irrelevant anchors such as those determined by playing dice can influence experts' judicial decision-making. Within public management, previous years' performance ratings influence new ratings irrespective of actual performance.
Anchoring bias is robust across different politico-administrative regimes, though effect sizes vary; for instance, goal-setting practices in the United Kingdom showed a very large effect compared to a medium effect in Italian studies.
Healthcare and Project Management
Anchoring is a prevalent heuristic in clinical medicine. It leads to diagnostic inaccuracies, inappropriate use of resources, and patient harm. Research suggests that up to 75 per cent of errors in internal medicine are cognitive in origin, occurring during steps such as information gathering, context formulation, and processing.
Clinicians often fixate on a previously assigned label, such as a suspected drug overdose, while ignoring contradictory evidence that might later reveal a brain tumour. In project management, anchoring is regarded as a silent killer of success, driving scope creep, cost overruns, and missed deadlines.
Project leaders frequently rely too heavily on the first estimate or number presented, effectively hijacking leadership decisions and draining budgets.
Commercial Pricing and Retail Strategy
Price anchoring is a psychological pricing strategy where a seller sets a high initial price to impact the percieved value of a subsequent discounted price. The anchor price serves as a reference point for consumers, shaping their understanding of what constitutes a reasonable cost.
This strategy leverages consumer psychology, as people evaluate value based on comparison rather than absolute cost. By introducing a high-price product before a low-price alternative, companies make the latter appear more affordable, enhancing willingness to buy and driving revenue. However, if shoppers view the anchor as arbitrary or unfair, it may undermine trust and negatively affect long-term pricing perception.
Artificial Intelligence and Machine Learning
Large Language Models reflect human-like biases by learning language patterns from human-generated text. These models exhibit bias-consistent behaviour in 17.8 per cent to 57.3 per cent of instances across judgment and decision-making contexts. Susceptibility decreases as model size increases up to a point, specifically around 32 billion parameters.
Models trained with explicit reasoning objectives demonstrate improved resistance to anchoring compared to similarly sized peers. Prompt detail also influences bias expression; higher prompt specificity, such as providing basic situational context and clear high-level directives, generally reduces bias by up to 14.9 per cent.
Mitigation and Debiasing Strategies
Debiasing strategies aim to transition the decision-maker into a slower, deliberate, and analytic mode of thought. Metacognition, or the awareness of and insight into one’s own thought processes, is a powerful tool for improvement. Forcing clinicians to ask what else a condition could be helps to mitigate the anchoring effect.
The consider-the-opposite strategy is a robust technique for association-based biases; it involves asking individuals to list two reasons why the anchor value is inappropriate.
Specific structural interventions include clean-sheet redesign, where decision-makers treat a strategy as if it were a new investment rather than adjusting from a prior baseline. Organisational strategies such as the red team–blue team approach involve independent groups representing opposing positions to ensure that each side challenges the other’s arguments.
Checklists have also proven efficient in slowing down decision-making, particularly in medicine and law. Additionally, dialectical bootstrapping requires individuals to average an initial quantitative estimate with a second estimate reached after assuming the first was incorrect, resulting in demonstrably higher accuracy. Successful transformation depends on awareness building and the reinforcement of reasoned decision-making through formal procedures.