The authors examined the comparative predictive capacity of the Trauma Symptom Inventory (TSI) Atypical Response Scale (ATR) and the standard set of Minnesota Multiphasic Personality Inventory-2 (MMPI-2) fake-bad validity scales (i.e., F, F[subscript B[prime]], F[subscript p[prime]], FBS) to detect feigned posttraumatic stress disorder (PTSD). Remitted trauma victims (n = 60) completed the TSI and MMPI-2 under standard (honest) instructions and then were randomly assigned to 1 of 2 experimental conditions (noncoached/validity scale coached) in which they were administered these instruments again with instruction to fake PTSD. These test protocols were compared with TSI and MMPI-2 results from workplace injury claimants with PTSD (n = 84). The ATR and FBS were able to distinguish only the noncoached participants instructed to fake from the PTSD claimants; in contrast, the F, F[subscript B[prime]], and F[subscript p[prime]] scales were able to distinguish both the noncoached and the validity-scale-coached participants from the PTSD claimants. F, F[subscript B[prime]], and F[subscript p[prime]] always outperformed the ATR and FBS; neither the ATR nor the FBS was able to add incremental predictive variance to that of F, F[subscript B[prime]], or F[subscript p[prime]]. (Contains 5 tables and 8 footnotes.)