Evaluation of the Response Bias Scale and Improbable Failure Scale in assessing feigned cognitive impairment

Abstract

The present study evaluated the Response Bias scale (RBS), a symptom validity test embedded within the Minnesota Multiphasic Personality Inventory (MMPI)-2 Restructured Form (MMPI-2-RF) that assesses for feigned neurocognitive complaints, in a sample of pretrial incompetent to stand trial (IST) criminal defendants. Additionally, we examined the Improbable Failure (IF) scale, a performance validity test embedded within the Structured Interview of Reported Symptoms, Second Edition (SIRS-2), which similarly assesses for feigned cognitive impairment (FCI). Results indicated that both the RBS (area under the curve [AUC] = .76) and IF scale (AUC = .72) achieved moderate classification accuracy using the Test of Memory Malingering (TOMM) as the criterion. Further, the RBS and IF scale appeared to be most useful for screening out those defendants who presented as genuine (specificity = 99% and 88%, respectively), and less effective at classifying those defendants suspected of feigning according to the TOMM (sensitivity = 29% and 46%, respectively). In order to identify a significant proportion of IST defendants who may be feigning impairment, considerably lower cutoff scores than those recommended in each measure's manual were evaluated. An RBS T score of 63 (sensitivity = 86%; specificity = 37%), and IF scale raw score of 2 (sensitivity = 80%; specificity = 43%), was required to achieve ≥80% sensitivity; these alternate cutoff scores may therefore be useful when screening inpatient forensic psychiatric IST defendants. Further, the 2 scales effectively predicted TOMM classification in combination, although only the RBS significantly contributed to the model. Implications for the assessment of FCI in forensic psychiatric settings are discussed. (PsycINFO Database Record