Forms it takes
The bias can occur at several points in reasoning. It need not be deliberate, and it appears in everyday judgment, professional diagnosis, and science.
- Selective search: choosing sources or questions likely to agree with an expectation
- Biased interpretation: scrutinizing unwelcome evidence harder than welcome evidence
- Selective recall: remembering examples that fit an existing view
Classic research
Peter Wason's 2-4-6 task of 1960 asked people to discover a number rule by proposing sequences. Many proposed only examples that fit their hunch and so missed the broader rule. Klayman and Ha later argued that this behavior reflects a positive test strategy, which is often sensible, and that it only misleads under particular conditions. Testing a hypothesis by its positive cases is therefore not the same as trying to protect it.
How strong and how fixable
The term is sometimes stretched to cover all belief persistence, including cases in which holding on to a well-supported belief in the face of weak contrary evidence is reasonable. Motivated reasoning and purely cognitive shortcuts are also often blended together. Debiasing is possible in part: asking people to consider the opposite can help in some settings, and structured procedures such as blinding and preregistration guard against the bias in research. Benefits of brief training may not last or transfer widely.
Further reading
- Wason, P. C. (1960). On the failure to eliminate hypotheses in a conceptual task. Quarterly Journal of Experimental Psychology, 12(3).
- Klayman, J., & Ha, Y.-W. (1987). Confirmation, disconfirmation, and information in hypothesis testing. Psychological Review, 94(2).
- Nickerson, R. S. (1998). Confirmation bias: A ubiquitous phenomenon in many guises. Review of General Psychology, 2(2).