← Back to list
Term

Confirmation Bias - Selection Logic

Seeking or interpreting information to confirm existing beliefs.

Aliases: Confirmation bias

Definition

Confirmation Bias: Seeking or interpreting information to confirm existing beliefs.

1. Mechanism (why it happens)

Confirmation bias is driven by selective search, biased interpretation, and memory advantages for belief-consistent information. Under identity and commitment pressures, it reduces exploration of disconfirming evidence and increases overconfidence.[^2]

2. Classic experiments / evidence

2.1 Rule discovery and falsification failure (Wason, 1960)

  • Design: Participants attempted to discover a number rule (e.g., 2-4-6 by proposing sequences and receiving feedback.[^1]
  • Manipulation: Task structure makes it easy to test confirming cases but requires falsification attempts to identify the true rule.[^1]
  • Key finding: Participants predominantly tested confirming sequences, delaying falsification and often failing to find the rule.[^1]
  • Notes/limitations: A foundational demonstration of confirmation-seeking in hypothesis testing.

2.2 Review synthesis (Nickerson, 1998)

  • Design: Comprehensive review across domains (scientific reasoning, everyday judgment).[^2]
  • Manipulation: Not a single experiment; synthesizes many tasks showing confirmation bias mechanisms.[^2]
  • Key finding: Confirmation bias is ubiquitous and appears in multiple guises.[^2]
  • Notes/limitations: Useful for establishing generality and scope.

3. Consumer decision patterns

  • Reading only positive reviews after forming a preference.
  • Searching “X is good–instead of “X problems / alternatives.
  • Discounting negative evidence as “outliers.

4. How marketing leverages it

Marketing funnels exploit confirmation bias via curated testimonials, selective review highlights, and algorithmic personalization that reinforces prior clicks.[^4]

5. Mitigation (Selection Logic)

  1. Force “disconfirming evidence first” (read critical reviews before positives).
  2. Use a consistent shortlist rubric (M4): Comparative analysis.
  3. Use systematic sourcing (M3) and require at least one independent source: Systematic evaluation.
  4. Validate outcomes (M5) and record regrets to improve future calibration: Decision validation.

References

  1. Wason, P. C. (1960). On the failure to eliminate hypotheses in a conceptual task. Quarterly Journal of Experimental Psychology, 12(3), 129–40.[source]
  2. Nickerson, R. S. (1998). Confirmation bias: A ubiquitous phenomenon in many guises. Review of General Psychology, 2(2), 175–20.[source]
  3. Kahneman, D. (2011). Thinking, Fast and Slow. Farrar, Straus and Giroux.[source]

Further Reading