Motivated reasoning
We reason toward the conclusion we want, not the one the evidence supports.
What it means
Reasoning in which the desire to reach a particular conclusion biases the cognitive processes of evidence search, evaluation, and memory, while preserving a feeling of objectivity. People apply asymmetric scrutiny — eagerly accepting congenial evidence while subjecting unwelcome evidence to harsh examination — and recruit memory and rules selectively to build a case for the favored view. Critically, the conclusion must still seem justifiable: people are constrained by the need to construct a plausible rationale, so they cannot believe just anything. The mechanism underlies self-serving judgments, partisan perception, and resistance to threatening health or scientific information, and it can grow stronger, not weaker, with greater knowledge and reasoning ability. It is a unifying account of why intelligence and effort often fail to debias belief.
Examples
Partisans shown identical mixed evidence on a policy each conclude the data clearly support their side, scrutinizing only the opposing studies.
A smoker reads the study linking cigarettes to heart disease with forensic care, picking apart the sample, then accepts the headline about coffee being healthy without a second thought.
The manager who chose the vendor combs the complaints for reasons they are unrepresentative, while treating one glowing review as proof the decision was right all along.
First described in Ziva Kunda (1990).