We look for information that confirms what we already believed.
After buying something expensive, it's common to read only the reviews that say it was worth it. Confirmation bias happens when we search for, interpret, and remember information in a way that confirms what we already believed, and ignore or dismiss whatever contradicts it.
One of the most cited examples is investors who only follow news and analysts that back the stock they already bought, and dismiss any sign they were wrong as "noise." The same pattern shows up in medical, political, and hiring decisions, where a first impression guides which later evidence gets sought out and which gets ignored.
Holding on to a belief you've already formed takes less mental effort than reviewing it from scratch. Seeking out information that confirms it, and avoiding what calls it into question, reduces the discomfort of being wrong.
Source: Nickerson, R. S. (1998). Confirmation Bias: A Ubiquitous Phenomenon in Many Guises. Review of General Psychology, 2(2), 175-220.
In user research, confirmation bias shows up when a team already has a favorite hypothesis and designs interview questions in a way that validates it. That's why experimental validation needs to be designed to be able to disprove a hypothesis, not just confirm it.
Related concepts
