The Test of Memory Malingering (TOMM) has not been adequately validated in a forensic psychiatric setting. Dissimulation of cognitive impairment, as assessed by the TOMM, was evaluated in a group of 25 forensic inpatients admitted for evaluation of Competency to Stand Trial (CST/MSO group), and hypothesized to be at higher risk for feigning cognitive impairment. A comparison group of 36 patients, who were either civilly committed or adjudicated Not Guilty by Reason of Insanity (CIVIL/NGRI group), were hypothesized to be less likely to feign cognitive impairment. Groups were comparable in age, education, premorbid intelligence, and psychiatric symptom severity. Significantly more CST/MSO patients (36%) scored below a recommended TOMM cutoff score relative to CIVIL/NGRI patients (6%). Findings indicate excellent specificity and modest sensitivity, and generally support the validity of the TOMM in a forensic psychiatric population. The utility of different cutoff scores and need for multiple indicators of effort are discussed.