---
title: "Cognitive Bias"
source: "/reference/cognitive-bias"
description: "Judgment is shaped by predictable mental shortcuts, prior beliefs, framing, defaults, salience, and context."
---
# Cognitive Bias
Judgment is shaped by predictable mental shortcuts, prior beliefs, framing, defaults, salience, and context.
### Principle
Judgment is shaped by predictable mental shortcuts, prior beliefs, framing, defaults, salience, and context.
### Why it matters
Interface decisions are never interpreted in a vacuum. Defaults can feel recommended, wording can frame outcomes, order can anchor perception, and prominent values can influence comparison. This category is broad, so Layr should avoid diagnosing a named bias without evidence. Its practical use is to detect when presentation systematically tilts a decision or when the product relies on an unrealistic assumption of perfectly rational behavior.
### Deeper design interpretation
Interfaces shape decisions through framing, defaults, order, salience, reference values, and language. These mechanisms can help users interpret complex choices, yet they can also tilt decisions in ways that users may not notice. Because “cognitive bias” is a broad family of effects rather than one single mechanism, Layr should avoid vague psychological diagnosis.
The useful design question is concrete: what does the presentation make easier to notice, assume, or accept? A default can imply recommendation. A first value can become an anchor. A warning can exaggerate perceived loss. A preselected option can reduce active consideration. These are observable design choices even when the user's internal response cannot be known.
Layr should describe the mechanism first and use psychological terminology only when it improves precision.
### What the agent should reason about
Inspect default states, copy symmetry, visual emphasis, ordering, reference prices, opt-in/opt-out effort, and disclosure placement. Give higher severity when the decision affects money, privacy, consent, access, or long-term commitment.
Do not infer that a user “was manipulated” from UI alone. Report the asymmetry and its plausible decision consequence.
### Common failure patterns
- Paid or privacy-affecting defaults are preselected without nearby explanation.
- One option uses emotionally positive copy while the alternative is framed as loss.
- Strike-through prices dominate without explaining the reference period or basis.
- Cancellation language is intentionally harder to parse than purchase language.
- The first or recommended option is emphasized without explaining the recommendation criteria.
### Audit questions
- Are alternatives described with comparable clarity and tone?
- Does any default carry material consequence?
- Are reference values and recommendations explained?
- Would a user understand the decision the same way if the order or visual emphasis changed?
### Example agent finding
> **High — Renewal consent is preselected and framed asymmetrically.** “Keep my benefits” is selected by default while the alternative reads “Lose access,” and the renewal price appears below the fold. Make renewal state explicit, surface price and timing at the decision point, and use neutral parallel labels for both alternatives.
### Agent directive
> Inspect how framing, defaults, order, emphasis, and prior context shape decisions. Describe the observable mechanism before naming a cognitive bias.
### Apply this when
- Pricing, consent, subscription, upgrades, onboarding defaults, comparison tables, risk communication, dashboards, and confirmation flows.
### Observable signals
- A default carries financial or privacy consequences that are not clearly explained.
- One option is framed with emotionally loaded language while alternatives are neutral.
- Reference values or strike-through prices dominate the decision without context.
- Opt-out paths are visually or linguistically harder to understand than opt-in paths.
- Ordering makes an arbitrary first value the comparison anchor.
### Recommended actions
- Present material consequences clearly and near the decision.
- Use neutral, parallel wording for meaningful alternatives.
- Make defaults explicit when they carry cost, privacy, or commitment.
- Preserve user agency and avoid using psychological effects to obscure tradeoffs.
### Do not over-apply
Do not label every preference or conversion pattern as bias. Identify the presentation mechanism and user consequence. Avoid speculative claims about a user's internal mental state.
### Example
**Weak:** A renewal checkbox is preselected and described as 'Keep my benefits' while the alternative says 'Lose access,' with the price disclosed far away.
**Stronger:** Renewal state, price, timing, and alternatives are presented in parallel language at the decision point.
### Evaluation notes
- **Default severity:** High for financial, privacy, or consent decisions; medium elsewhere.
- **Typical confidence:** Medium; the UI mechanism may be observable while the psychological response is inferred.
- **Primary evidence:** Default states, copy, ordering, emphasis, disclosure location, decision consequences.
- **Related rules:** Choice Overload, Von Restorff Effect, Peak-End Rule, Mental Model
### References
- [Laws of UX — Cognitive Bias](https://lawsofux.com/cognitive-bias/)
## Related rules
- [Choice Overload](/__layr/pages/reference/choice-overload.md)
- [Von Restorff Effect](/__layr/pages/reference/von-restorff-effect.md)
- [Peak-End Rule](/__layr/pages/reference/peak-end-rule.md)
- [Mental Model](/__layr/pages/reference/mental-model.md)
## Agent guidance
Review criteria, the finding contract a report must use, and worked examples.
Established principle; Layr interpretation and agent guidance. Category: Memory & Cognition.
## Retrieve
- Canonical page: https://layr.design/reference/cognitive-bias
- Markdown for agents: https://layr.design/__layr/pages/reference/cognitive-bias.md
- Reference map: https://layr.design/llms.txt
- Route index: https://layr.design/__layr/routes.toon
## Rule object
```yaml
rule:
id: "cognitive-bias"
title: "Cognitive Bias"
category: "Memory & Cognition"
origin: established
canonical_url: "https://layr.design/reference/cognitive-bias"
principle: "Judgment is shaped by predictable mental shortcuts, prior beliefs, framing, defaults, salience, and context."
directive: "Inspect how framing, defaults, order, emphasis, and prior context shape decisions. Describe the observable mechanism before naming a cognitive bias."
applies_when:
- "Pricing, consent, subscription, upgrades, onboarding defaults, comparison tables, risk communication, dashboards, and confirmation flows."
observable_signals:
- "A default carries financial or privacy consequences that are not clearly explained."
- "One option is framed with emotionally loaded language while alternatives are neutral."
- "Reference values or strike-through prices dominate the decision without context."
- "Opt-out paths are visually or linguistically harder to understand than opt-in paths."
- "Ordering makes an arbitrary first value the comparison anchor."
audit_questions:
- "Are alternatives described with comparable clarity and tone?"
- "Does any default carry material consequence?"
- "Are reference values and recommendations explained?"
- "Would a user understand the decision the same way if the order or visual emphasis changed?"
recommended_actions:
- "Present material consequences clearly and near the decision."
- "Use neutral, parallel wording for meaningful alternatives."
- "Make defaults explicit when they carry cost, privacy, or commitment."
- "Preserve user agency and avoid using psychological effects to obscure tradeoffs."
avoid_when:
- "Do not label every preference or conversion pattern as bias. Identify the presentation mechanism and user consequence. Avoid speculative claims about a user's internal mental state."
common_failure_patterns:
- "Paid or privacy-affecting defaults are preselected without nearby explanation."
- "One option uses emotionally positive copy while the alternative is framed as loss."
- "Strike-through prices dominate without explaining the reference period or basis."
- "Cancellation language is intentionally harder to parse than purchase language."
- "The first or recommended option is emphasized without explaining the recommendation criteria."
evidence_types:
- "Default states"
- "copy"
- "ordering"
- "emphasis"
- "disclosure location"
- "decision consequences"
default_severity: "High for financial, privacy, or consent decisions; medium elsewhere."
typical_confidence: "Medium; the UI mechanism may be observable while the psychological response is inferred."
related_rules:
- { id: "choice-overload", title: "Choice Overload", url: "https://layr.design/reference/choice-overload" }
- { id: "von-restorff-effect", title: "Von Restorff Effect", url: "https://layr.design/reference/von-restorff-effect" }
- { id: "peak-end-rule", title: "Peak-End Rule", url: "https://layr.design/reference/peak-end-rule" }
- { id: "mental-model", title: "Mental Model", url: "https://layr.design/reference/mental-model" }
```
## What to reason about
Inspect default states, copy symmetry, visual emphasis, ordering, reference prices, opt-in/opt-out effort, and disclosure placement. Give higher severity when the decision affects money, privacy, consent, access, or long-term commitment. Do not infer that a user “was manipulated” from UI alone. Report the asymmetry and its plausible decision consequence.
## Routing
Route a review here when the user must remember, infer, translate, reconcile, or reconstruct information. These rules are especially relevant when the system already possesses context that the interface fails to expose.
## Review procedure
For every finding:
1. Identify the exact component, state, route, or interaction.
2. State the observable evidence.
3. Name the principle that applies.
4. Explain the user consequence in the current context.
5. Estimate severity from the consequence, frequency, and reversibility.
6. Recommend the smallest change that addresses the issue while preserving the product's existing design system.
7. Lower confidence when intent, frequency, audience, device, or product constraints must be inferred.
8. Do not produce generic advice such as "improve hierarchy" or "make this clearer" without specifying what is wrong and why.
## Finding contract — emit exactly this shape
```yaml
finding:
rule_id: string
location: string
evidence:
- observable fact
impact: string
severity: low | medium | high | critical
confidence: low | medium | high
recommendation:
change: string
rationale: string
preserve:
- existing design-system behavior
- valid product constraints
```
## Example finding for this rule
High — Renewal consent is preselected and framed asymmetrically. “Keep my benefits” is selected by default while the alternative reads “Lose access,” and the renewal price appears below the fold. Make renewal state explicit, surface price and timing at the decision point, and use neutral parallel labels for both alternatives.
## Code reference
The same principle as markup. Compare the structure, not the class names.
```html
BillingRenewal
Renew your plan
Keep my benefits
Lose access
$120$96
Price and renewal date shown on the next screen.Preselected · loss framing · reference price without a basis
```
```html
BillingRenewal
Renew your plan
Renew for another year$96 · charged 14 Oct
Do not renewAccess ends 14 Oct · data kept 30 days
$120$96Standard price until 30 Sep
Parallel language · nothing preselected · price at the decision