Detecting deception: adversarial problem solving in a low base-rate world

The work presented here investigates the process by which one group of indi viduals solves the problem of detecting deceptions created by other agents. A field experiment was conducted in which twenty-four auditors (partners i n international public accounting firms) were asked to review four cases de scribing real companies that, unknown to the auditors, had perpetrated fina ncial frauds. While many of the auditors failed to detect the manipulations in the cases, a small number of auditors were consistently successful. Sin ce the detection of frauds occurs infrequently in the work of a given audit or, we explain success by the application of powerful heuristics gained fro m experience with deceptions in everyday life. These heuristics implement a variation of Dennett's intentional stance strategy, which is based on inte rpreting detected inconsistencies in the light of the Deceiver's (i.e., man agement's) goals and possible actions. We explain failure to detect decepti on by means of perturbations (bugs) in the domain knowledge of accounting n eeded to apply these heuristics to the specific context of financial statem ent fraud. We test our theory by showing that a computational model of frau d detection that employs the proposed heuristics successfully detects fraud s in the cases given to the auditors. We then modify the model by introduci ng perturbations based on the errors made by each of the auditors in the fo ur cases. The resulting models account for 84 of the 96 observations (i.e., 24 auditors x four cases) in our data. (C) 2001 Cognitive Science Society, inc. All rights reserved.