Defining & Identifying A Cognitive Bias
Definition (via Oxford)
A cognitive bias is any systematic deviation from a normative criterion that affects thinking, often leading to errors in judgment. Affect, in particular, may bias cognition, both by altering depth of cognitive processing and by impacting the content of cognitions. A useful example of emotion altering depth of processing appears in studies by Bodenhausen et al (1994). Happy individuals demonstrated less depth of processing than individuals in a neutral affective state, evidenced by their reliance on simple mental categories rather than on complex stimuli. A useful example of emotion altering the content of thought appears in studies on the affect-as-information model (see affect-as-information). Individuals in a positive mood judged their overall life satisfaction more positively than did individuals in a negative mood. That is, their temporary positive mood altered the content of their thoughts about satisfaction as a whole.
4 Demonstrations of Emotional Biases (via Oxford)
1. Judgment lacks correspondence with a criterion. The most direct way to measure a bias is to compare human judgment to a known normative criterion. For example, Bechara et al. (1997) examined the decision making of patients with lesions in their pre-frontal cortex – an area that integrates emotion with cognition. They
2. Judgment lacks correspondence with judgments of others. Another way to measure a bias is to compare the judgment of different groups on a task. If the groups’ judgments fail to cohere, then one can infer that at least one of the groups must be biased. When using this approach, it is especially useful to identify one group as expert, so it can serve as the standard for comparison.
3. Judgment relies on bad information. The existence of cognitive bias can also be inferred from individuals’ reliance on a bad judgmental cue.
4. Judgment fails to use good information. Finally, one can infer a bias if individuals fail to utilize a good judgmental cue. In Lerner et al’s (1998) study of legal decision making, participants were randomly assigned to an anger condition or a neutral condition.