Confirmation bias has several forms
Confirmation bias is not only noticing agreeable facts. It can influence the questions we ask, sources we choose, meaning we assign to ambiguous evidence and memories we retrieve. A major review documented the pattern across hypothesis testing, beliefs and social judgment. [1]
Suppose you expect a colleague to be dismissive. You may watch closely for interruptions, interpret a short reply as contempt and forget a helpful message. The expectation does not invent every event, but it can organize which events become evidence.
Why contrary information is uncomfortable
A meta-analysis on selective exposure found that people generally preferred attitude-consistent information, especially when a belief was tied to values or confidence, while useful or high-quality contrary information could still attract attention. [2] Confirmation is a tendency with moderators, not a wall no evidence can cross.
Beliefs also reduce search costs. Reopening every assumption would make ordinary action impossible. The relevant question is when confidence or consequence makes a more deliberate test worthwhile.
Disconfirmation without performance
Trying to become perfectly neutral can itself become theatre. A better method is to state what observation would lower your confidence. If no possible observation counts, the belief is not currently being tested.
Search for the strongest plausible alternative, not a weak opposing example. Ask someone who disagrees to describe what you are missing. Where numbers matter, use the underlying rate rather than the most memorable story. Simple decision rules can also help when they fit the environment. [3]
A belief audit
For one belief affecting a choice, complete:
- I currently believe…
- The best evidence for it is…
- A credible alternative explanation is…
- I would lower my confidence if I observed…
The result need not be a reversed belief. Moving from certainty to calibrated confidence is already useful.
Do not weaponize the label
Calling another person biased does not answer their evidence. Confirmation bias also does not imply that deeply held beliefs are false. The concept describes how information can be processed, not the truth value of the final conclusion.
Ataraxia’s relevant move is a pause: name the interpretation, distinguish observation from expectation and decide what deserves another look.
Frequently asked questions
Is confirmation bias deliberate?
Often it is not. Search, attention and interpretation can favor a prior belief without conscious manipulation.
Does confirmation bias mean my belief is false?
No. A biased process can reach a true conclusion. The bias is a reason to check the evidence, not an automatic refutation.
What is the simplest check?
State one realistic observation that would reduce your confidence, then look for it fairly.
Sources and references
Prepared for Ataraxia using cited sources. No professional credential or scientific board is claimed.
- Scientific reviewNickerson, R. S. (1998). “Confirmation bias: A ubiquitous phenomenon in many guises.” Review of General Psychology, 2(2), 175–220. DOI: 10.1037/1089-2680.2.2.175. View source
- Meta-analysisHart, W.; Albarracín, D.; Eagly, A. H.; et al. (2009). “Feeling validated versus being correct: A meta-analysis of selective exposure to information.” Psychological Bulletin, 135(4), 555–588. DOI: 10.1037/a0015701. View source
- Scientific reviewGigerenzer, G.; Gaissmaier, W. (2011). “Heuristic decision making.” Annual Review of Psychology, 62, 451–482. DOI: 10.1146/annurev-psych-120709-145346. View source