Faking depression: what type of coaching reduces the chance of detection?

Document Type


Publication details

Yoxall, JS 2015, 'Faking depression: what type of coaching reduces the chance of detection?', paper presented to the 2015 APS College of Forensic Psychologists Conference, Sydney, Australia, 16-18 April.

Abstract available on open access


In the civil forensic arena, and particularly in workers’ compensation matters, some individuals may be more motivated to feign depression rather than the less common, and more overt symptoms of schizophrenia or similar disorders. Psychologist who are routinely involved in pre-liability assessments frequently administer multi-scale measure of psychopathology which include validity scales. In Australia, the Personality Assessment Inventory (PAI: Morey, 1991) is one of the most popular of such measures (Yoxall, Bahr & Barling, 2010). Whilst there is now a substantial literature on the validity of the three negative distortion validity indices of the PAI (the Negative Impression Management Scale, The Malingering Scale, and Rogers Discriminant Function); and the impact of various types of coaching on such scales, most previous studies have focused on the detection of feigning of major mental illness such as schizophrenia and other psychotic disorders. Less is known about the validity of such scales in detecting feigning of more subtle disorders such as depression. The current paper reports on a study that compared the impact of two common coaching strategies (symptom focused vs detection strategy focused) on the detection of feigned depression on the PAI. Sixty one university students were randomly allocated to one of three experimental groups; Symptom Coached, Validity Scale Coached, or Honest/Control. Coached participants were instructed to feign Depression, and were given a financial incentive to believably do so. Using a between groups simulation design, PAI group differences were assessed. Results revealed that both feigning groups were successful in feigning depression. Analysis of PAI malingering indices (Negative Impression Management (NIM), Malingering Index (MAL), and Roger’s Discriminant Function (RDF)), revealed that validity scale coached participants were less detectable than symptom coached feigners. On average, symptom coached feigners elevated on the NIM, MAL and RDF, however, validity scale coached feigners elevated only on RDF. RDF had the highest classification accuracy, showing high sensitivity.