The Psychology of Case-Based Reasoning: How Information Produced from Case-Matching Methods Facilitates the Use of Statistically Generated Information

Patrick Wheeler
University of Missouri-Columbia

Donald R. Jones
Texas Tech University

ABSTRACT:

Case-based reasoning (CBR) research has compared CBR output to other decision support methods such as statistics. This study differs by evaluating the effect of CBR as an input to judgment. We develop psychological theory concerning the use of CBR and its attenuation of biases that impair the use of statistical advice. Study 1 finds participants shown CBR cases made judgments closer to regression advice than do participants shown only regression advice. Study 2 finds that regression and CBR were used to corroborate each other. Participants used central tendency and dispersion information from CBR cases when this information was apparent in the cases. However, such use was not always normative. Results support including multiple decision aiding features in DSS.

Keywords: Decision support systems, judgment analysis, cased-based reasoning, information processing methods