| name | bias-detection |
| description | Library of cognitive biases for identifying and correcting systematic errors in reasoning during decision-making |
Bias Detection Reference
Purpose: A comprehensive reference for identifying and correcting cognitive biases that undermine decision quality.
When to Use: Consult during any decision gate, especially during Calibration and Contrarian Analysis.
Severity Levels
| Level | Description | Response |
|---|
| Minor | Bias present but unlikely to significantly affect outcome | Note and monitor; proceed with awareness |
| Moderate | Bias could meaningfully skew decision | Pause to address; apply specific correction technique |
| Major | Bias likely to cause serious decision error | Stop and remediate; consider restarting analysis from earlier gate |
Common Biases
1. Confirmation Bias
What it is: The tendency to search for, interpret, and recall information that confirms pre-existing beliefs while ignoring contradictory evidence.
Signs:
- Evidence search focuses only on sources likely to support preferred conclusion
- Dismissing contradictory evidence as "outliers" or "unreliable" without rigorous evaluation
- Interpreting ambiguous information as supporting the favored option
Correction:
- Actively seek out the strongest arguments against your preferred position
- Assign someone the explicit role of devil's advocate with genuine authority to challenge
- Before concluding research, ask: "What evidence would change my mind?" and search specifically for it
Severity: Major - This bias is pervasive and can undermine the entire decision process
2. Anchoring
What it is: Over-relying on the first piece of information encountered (the "anchor") when making decisions, even when that information is arbitrary or irrelevant.
Signs:
- Initial estimates or numbers disproportionately influence final conclusions
- Adjustments from starting point feel insufficient even when evidence suggests larger changes
- First option presented becomes the implicit benchmark against which all others are judged
Correction:
- Generate estimates independently before seeing any reference numbers
- Consider multiple starting points and work from each separately, then compare conclusions
- Explicitly ask: "If we hadn't seen that first number/option, what would we think?"
Severity: Moderate - Common in negotiations and financial decisions; addressable with awareness
3. Overconfidence
What it is: Excessive certainty in one's own answers, judgments, and predictions; believing you know more than you actually do.
Signs:
- Confidence intervals are too narrow (e.g., 90% confidence ranges that are correct only 50% of the time)
- Dismissing uncertainty or risk factors as unlikely without rigorous probability assessment
- Treating assumptions as facts and projections as certainties
Correction:
- Explicitly state confidence levels and track calibration over time
- Require evidence proportional to certainty claims - extraordinary confidence requires extraordinary evidence
- Use pre-mortem technique: "Assume this failed - why did it fail?"
Severity: Major - Leads to inadequate risk management and surprise failures
4. Sunk Cost Fallacy
What it is: Continuing to invest in a decision because of previously invested resources (time, money, effort) rather than future value.
Signs:
- Arguments for continuing reference past investment rather than future returns
- Reluctance to abandon projects despite poor prospects because of "how far we've come"
- Framing quitting as "waste" rather than rational resource reallocation
Correction:
- Evaluate decisions based solely on future costs and benefits, ignoring past investment
- Ask: "If we were starting fresh today with no history, would we choose this path?"
- Set explicit kill criteria at the start of projects and honor them when triggered
Severity: Moderate - Common but correctable with discipline; watch for escalation of commitment
5. Availability Bias
What it is: Judging likelihood or frequency based on how easily examples come to mind, rather than actual probability.
Signs:
- Recent or vivid events disproportionately influence risk assessment
- Overweighting dramatic risks (plane crashes) while underweighting mundane ones (car accidents)
- Personal experience trumping statistical evidence
Correction:
- Seek base rate data before forming probability estimates
- Ask: "Is this example memorable because it's common or because it's unusual?"
- Deliberately search for examples that contradict easily recalled ones
Severity: Moderate - Particularly dangerous in risk assessment and resource allocation
6. Groupthink
What it is: Prioritizing group harmony and conformity over realistic evaluation of alternatives; suppressing dissent to maintain consensus.
Signs:
- Dissenting opinions quickly abandoned or not voiced at all
- Pressure (subtle or explicit) on members who question the group's direction
- Illusion of unanimity - assuming silence means agreement
Correction:
- Assign a formal devil's advocate role that rotates among members
- Collect opinions independently (written) before group discussion
- Leader speaks last; explicitly invite and reward constructive disagreement
Severity: Major - Can lead to catastrophic collective failures; requires structural safeguards
7. Framing Effects
What it is: Being influenced by how information is presented rather than the information itself; different descriptions of identical outcomes lead to different decisions.
Signs:
- Strong preference shifts depending on whether options are described as gains vs. losses
- Risk tolerance changes based on problem description rather than underlying probabilities
- Emotional reactions to presentation style override rational evaluation
Correction:
- Reframe the same decision multiple ways: as gain/loss, risk/opportunity, short-term/long-term
- Convert qualitative descriptions to quantitative terms where possible
- Ask: "Would my decision change if this were described differently?"
Severity: Moderate - Addressable through deliberate reframing exercises
8. Planning Fallacy
What it is: Systematically underestimating time, costs, and risks of future actions while overestimating benefits; believing your project is different from similar past projects that failed.
Signs:
- Estimates based on best-case scenarios rather than realistic or base-rate outcomes
- Ignoring or discounting data from comparable past projects
- Explanations for why "this time is different" without rigorous justification
Correction:
- Use reference class forecasting: find similar past projects and use their actual outcomes as baseline
- Apply systematic multipliers to initial estimates based on historical accuracy
- Conduct pre-mortem analysis before committing to timeline
Severity: Major - Causes chronic under-delivery and budget overruns; requires systematic correction
Bias Audit Checklist
Use this checklist during Calibration and before finalizing decisions:
Information Gathering
Evaluation Process
Group Dynamics
Overall Assessment
