In 1998, a Tufts psychologist named Raymond Nickerson published a long review article pulling together decades of scattered experiments under one heading. The title called confirmation bias a ubiquitous phenomenon in many guises.
What he described was a habit most of us would recognise if we caught ourselves doing it, which is exactly the problem: we usually don’t.
A note before going further. We are writers who read the research, not psychologists or clinicians. The studies below describe patterns across groups of people, not a diagnosis of how any one reader thinks, and the science here is livelier and less settled than a tidy summary can suggest. Treat what follows as reflection on the literature, not advice.
The paper that named the pattern
Nickerson gave the term a working definition that has stuck. As the phrase is typically used in psychology, he wrote, it connotes “the seeking or interpreting of evidence in ways that are partial to existing beliefs, expectations, or a hypothesis in hand”.
The trail runs back to Peter Wason’s rule-testing experiments around 1960, in which people offered evidence that would confirm their guess rather than the kind that could prove it wrong. Jonathan Evans later called confirmation bias one of the best known and most widely accepted notions of inferential error in the whole literature on human reasoning.
Why the mind might work this way
If the bias is so common, the obvious question is why. One influential answer flips the usual assumption. The usual assumption is that reasoning exists to help us privately work out what is true. The cognitive scientists Hugo Mercier and Dan Sperber proposed, in a much-discussed 2011 paper, that it evolved for something else: to produce and evaluate arguments in social settings, to persuade and to check the persuasion of others.
On that account, a bias toward your own side isn’t a flaw in the machinery; it’s what the machinery is for. Mercier and Sperber write that “skilled arguers, however, are not after the truth but after arguments supporting their views. This explains the notorious confirmation bias.” They contend that “reasoning can lead to poor outcomes, not because humans are bad at it, but because they systematically strive for arguments that justify their beliefs or their actions.”
This is the claim of one theory, contested rather than settled, and worth taking seriously as an account of why the pattern might exist rather than as the final word on it. What it offers is a reframe: the mind that waves away inconvenient facts may be doing precisely the job it was built to do.
Where it shows up
Perhaps the best example is an older one. A 1979 study by Charles Lord, Lee Ross and Mark Lepper took 48 undergraduates, half in favour of the death penalty and half against, and gave both groups the same mixed bundle of made-up studies, some supporting the deterrent case and some undermining it. Each side judged the study that agreed with them to be the better-designed one. More striking, both groups came away more convinced than before. The same evidence pushed the two camps further apart, a result the authors called biased assimilation.
You do not need a laboratory to see the shape of it, though. The person who half-remembers a symptom, types it into a search engine, and reads past the reassuring explanations to land on the frightening one has done a version of this. So has the investor who reads glowing coverage of a stock they already own and skims the warnings. The domains differ; the mechanism is the same.
Why it is so hard to catch in yourself
The bias is slippery because it does not feel like bias from the inside. It feels like being reasonable. Work on motivated reasoning describes the asymmetry as a kind of uneven scrutiny: research suggests people are “uneven skeptics” who hold evidence they dislike to a much higher standard than evidence they welcome. Congenial claims get a nod and a pass. Uncongenial ones get cross-examined until a flaw turns up, and there is almost always a flaw if you look hard enough.
Both moves feel like careful thinking. Scrutinising a weak study is what a good reasoner should do. The trouble is that the scrutiny is applied unevenly, and the unevenness is invisible to the person doing it. The extra scepticism registers as rigour, not as a thumb on the scale.
What, if anything, can be done
Simply knowing the bias exists does not reliably switch it off. A preregistered study tested three debiasing techniques, including the much-recommended “consider the opposite” and a psychoeducation approach. One method, drawing on social norms, reduced how selectively people sought out information, but none significantly reduced biased assimilation compared with a control group.
The picture is not entirely bleak. A separate experiment with national risk analysts found that experienced analysts showed less confirmation bias than students, and that a single debiasing session reduced the bias in both groups, hinting that cheap interventions can help even skilled judgement.
Read together, the two results suggest something narrow: certain structured practices, applied in the right setting, can nudge the needle, while general awareness on its own tends not to.
If wrestling with your own certainty on a subject that matters starts to feel heavier than an interesting puzzle, a good counsellor or therapist is worth talking to.
