Detecting faked psychopathology: a comparison of two tests to detect malingered psychopathology using a simulation design.

摘要

Malingered psychopathology has the potential to be a costly social problem and there is a need for studies that compare the malingering detection capabilities of tests of psychopathology. This study investigated the capacity of two measures to detect simulated psychopathology. Forty-one first-year psychology students were randomly allocated to experimental groups that included malingering and control conditions. Analogue malingerers were given a financial incentive to simulate believable psychological impairment. Controls received standardised test instructions and the prize incentive, contingent on good effort. In a between-group simulation design, group differences on the Personality Assessment Inventory (PAI) and the revised Symptom Checklist-90 (SCL-90-R) were assessed. Group comparisons revealed elevation of the majority of clinical index scores among malingerers and a consistent pattern of results across tests. Analysis of the test operating characteristics of the malingering indices for these measures revealed superior detection of simulated malingering using the PAI, particularly Rogers' Discriminant Function, although classification accuracy of all malingering indexes was improved when adjusted cut-offs were used. Overall, results from this study demonstrate the vulnerability of the PAI and (SCL-90-R) to simulated psychopathology, but also the capacity of these measures to detect such performance when specific indexes are used.