Back to docs

Trait AnchoringBias

Category: Tier 3 - Decision-Making Traits Scale: 0.0 (low susceptibility) to 1.0 (high susceptibility)

Definition

Anchoring Bias is the tendency to rely on the first piece of information seen (the "anchor") when making later judgments. This holds even when the anchor is arbitrary or irrelevant.

In web contexts, this trait affects price perception, quantity estimates, quality evaluation, and number processing. High-anchoring users' judgments drift toward initial values. Low-anchoring users adjust more fully toward rational estimates.

Research Foundation

Primary Citation

"In many situations, people make estimates by starting from an initial value that is adjusted to yield the final answer... adjustments are typically insufficient. That is, different starting points yield different estimates, which are biased toward the initial values." — Tversky & Kahneman, 1974, p. 1128

Full Citation (APA 7): Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124-1131.

DOI: https://doi.org/10.1126/science.185.4157.1124

The Wheel Experiment

The landmark demonstration of anchoring:

"Subjects were asked to estimate various quantities, stated in percentages (for example, the percentage of African countries in the United Nations). A wheel of fortune with numbers 1-100 was spun in subjects' presence. Subjects were first asked whether the quantity was higher or lower than the number on the wheel, and then asked for their estimate. The arbitrary number had a marked effect on estimates." — Tversky & Kahneman, 1974, p. 1128

Key Finding:

  • When the wheel stopped at 10: Median estimate of African UN countries = 25%
  • When the wheel stopped at 65: Median estimate of African UN countries = 45%
  • The anchor shifted estimates by 20 percentage points despite being completely random

Key Numerical Values

Metric Value Source
Low anchor (10) -> estimate 25% Tversky & Kahneman (1974)
High anchor (65) -> estimate 45% Tversky & Kahneman (1974)
Anchor effect size 20 percentage points Tversky & Kahneman (1974)
Real estate listing anchor effect $11,000-14,000 Northcraft & Neale (1987)
Price anchor persistence 48+ hours Ariely et al. (2003)
Anchor effect on WTP (willingness to pay) 60-120% Ariely et al. (2003)
Expert susceptibility (real estate agents) Nearly equal to amateurs Northcraft & Neale (1987)

Behavioral Levels

Value Label Behaviors
0.0-0.2 Anchor Resistant Largely ignores suggested values; makes independent estimates; skeptical of "was/now" pricing; compares across sources before forming judgments; resets expectations when context changes
0.2-0.4 Low Susceptibility Acknowledges anchors but adjusts significantly; cross-references prices and ratings; somewhat influenced by defaults but overrides when motivated; moderate adjustment from starting points
0.4-0.6 Moderate Susceptibility Noticeable anchor influence; accepts default form values frequently; price perception shaped by strikethrough prices; rating expectations set by first reviews; partial adjustment from anchors
0.6-0.8 High Susceptibility Strong anchor influence on judgments; "was $99, now $49" highly persuasive; first review strongly shapes opinion; default values rarely changed; limited adjustment from starting points
0.8-1.0 Extreme Susceptibility Anchors dominate judgment; original prices define value perception; first information encountered becomes truth; almost never changes default values; minimal adjustment regardless of evidence

Web Behavior Patterns

Price Perception

Anchor-Resistant (0.0-0.3):

  • Ignores "was/now" strikethrough pricing
  • Compares prices across multiple sites
  • Uses price history tools
  • Skeptical of "limited time" claims
  • Values absolute price over relative discount

Highly Anchored (0.7-1.0):

  • "Was $200, now $99" feels like genuine 50% savings
  • First price seen sets value expectation
  • MSRP anchors all discount evaluations
  • Higher anchor makes actual price seem reasonable
  • "Compare at $150" influences perception

Form Default Values

Anchor-Resistant:

  • Reviews and changes default selections
  • Calculates appropriate values independently
  • Questions why defaults are set as they are
  • Changes tip percentages from suggested amounts

Highly Anchored:

  • Accepts pre-filled values as appropriate
  • Uses suggested donation amounts
  • Leaves tip percentage at first option
  • Rarely modifies quantity defaults (qty: 1)

Rating and Review Perception

Anchor-Resistant:

  • Reads multiple reviews before forming opinion
  • Weights recent reviews appropriately
  • Discounts extreme first impressions
  • Considers review distribution not just average

Highly Anchored:

  • First review shapes product perception
  • Initial star rating becomes expected quality
  • Early negative review creates lasting negative impression
  • "Featured review" disproportionately influential

Numerical Estimation

Anchor-Resistant:

  • Makes independent estimates before seeing suggestions
  • Recognizes irrelevant numbers as manipulation
  • Adjusts fully when given new information

Highly Anchored:

  • "Enter amount: $100" influences donation amount
  • Suggested search refinements affect query
  • Countdown timers affect urgency perception
  • "X people are viewing this" shapes demand perception

Estimated Trait Correlations

Correlation estimates are derived from related research findings and theoretical models. Empirical calibration is planned (GitHub #95).

Related Trait Correlation Mechanism
Trait-Comprehension r = -0.22 Understanding enables anchor recognition
Trait-RiskTolerance r = 0.18 Risk-takers may use anchors as shortcuts
Trait-Satisficing r = 0.35 Satisficers accept anchored "good enough" values
Trait-SelfEfficacy r = -0.24 Confidence enables independent judgment
Trait-TrustCalibration r = -0.31 Skeptics question anchor validity
Trait-AuthoritySensitivity r = 0.38 Authority-sensitive users accept suggested values

Persona Values

Persona Anchoring Bias Value Rationale
Elderly Novice 0.80 Trusts displayed values as authoritative
Distracted Teen 0.70 Quick processing relies on anchors
First-Time User 0.65 Lacks context for independent judgment
Overwhelmed Parent 0.60 Cognitive load increases heuristic use
Anxious User 0.55 Uncertainty increases anchor reliance
Careful Senior 0.45 Methodical but still susceptible
Rushed Professional 0.50 Time pressure increases anchoring
Power User 0.30 Experience provides comparison context
Tech Enthusiast 0.25 Research habits reduce anchor influence

Design Implications

Ethical Anchoring

  1. Reasonable defaults - Pre-fill values that genuinely help users
  2. Accurate original prices - Show real previous prices, not inflated MSRPs
  3. Balanced review display - Don't always show extreme reviews first
  4. Transparent suggestions - Explain why values are suggested

Dark Pattern Awareness

Sites exploit anchoring through:

  • Inflated "original" prices
  • Extreme high-anchor subscription tiers ("Enterprise: $999/mo")
  • Pre-selected quantities or options
  • Artificially high "compare at" prices
  • Suggested tip amounts that anchor high

Testing Considerations

CBrowser tests should verify:

  • Users aren't manipulated by arbitrary anchors
  • Default values are genuinely helpful
  • Price presentations are honest
  • Review ordering is fair

Measurement in CBrowser

// Anchoring affects value perception and defaults
function perceiveValue(
  displayedPrice: number,
  originalPrice: number | null,
  traits: Traits
): PerceivedValue {
  if (originalPrice === null) {
    return { value: displayedPrice, confidence: 'neutral' };
  }

  const discount = (originalPrice - displayedPrice) / originalPrice;
  const anchorInfluence = discount * traits.anchoringBias;

  // Highly anchored users perceive more value from discount framing
  const perceivedValue = displayedPrice * (1 - anchorInfluence * 0.5);

  return {
    value: perceivedValue,
    confidence: anchorInfluence > 0.3 ? 'good-deal' : 'neutral',
    likelyToPurchase: anchorInfluence > 0.4
  };
}

// Default value acceptance
function modifyDefault(defaultValue: number, optimalValue: number, traits: Traits): number {
  // High anchoring = accept default; low = adjust to optimal
  const adjustment = (optimalValue - defaultValue) * (1 - traits.anchoringBias);
  return defaultValue + adjustment;
}

See Also

Bibliography

Ariely, D., Loewenstein, G., & Prelec, D. (2003). "Coherent arbitrariness": Stable demand curves without stable preferences. The Quarterly Journal of Economics, 118(1), 73-106. https://doi.org/10.1162/00335530360535153

Furnham, A., & Boo, H. C. (2011). A literature review of the anchoring effect. The Journal of Socio-Economics, 40(1), 35-42. https://doi.org/10.1016/j.socec.2010.10.008

Kahneman, D. (2011). Thinking, fast and slow. Farrar, Straus and Giroux.

Northcraft, G. B., & Neale, M. A. (1987). Experts, amateurs, and real estate: An anchoring-and-adjustment perspective on property pricing decisions. Organizational Behavior and Human Decision Processes, 39(1), 84-97. https://doi.org/10.1016/0749-5978(87)90046-X

Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124-1131. https://doi.org/10.1126/science.185.4157.1124

Copyright: (c) 2026 Alexa Eden.

License: MIT License

Contact: [email protected]

From the Blog