When Expert Decision Making Goes Wrong: Consensus, Bias, the Role of Experts, and Accuracy

Reply by the current authors to the comments made by William R. Oliver et al. (see record 2018-15854-021) on the original article (see record 2017-57700-015). Oliver presented a study in which contextual information impacted medical examiners’ decision making. Results showed that this information increased consensus among examiners by 18%, as well as confidence in their judgments. These findings are highly consistent with myriad research showing the impact of contextual information on decision making—including among experts within established forensic science domains such as fingerprinting and DNA. In his commentary on "Cognitive Bias and Blindness" interprets his own findings as proof that contextual information increased accuracy while diminishing error. Yet increased consensus and confidence cannot be used to infer accuracy. Rather, Oliver’s findings betray the fact that experts were more likely to disagree when they based their conclusions solely on the medical -relevant data. Adding non medical, irrelevant contextual information masks this problem by artificially increasing consensus and confidence, both of which can be misinterpreted as indicators of accuracy. (PsycInfo Database Record (c) 2023 APA, all rights reserved)