Using the WMS-III to detect malingering: empirical validation of the rarely missed index (RMI)

To detect malingering during memory assessment, we evaluated item response biases to the Logical Memory Delayed Recognition (LMDR) subtest of the WMS-III. In a sample of 50 healthy volunteers who were completely naïve to the content of the Logical Memory stories, 6 LMDR items were correctly endorsed above chance probabilities. These 6 rarely missed items significantly discriminated 51 patients with neurological impairment from 36 volunteers who attempted to feign head injury and poor cognitive performance. A weighted combination of the 6 items was summed to form a single Rarely Missed Index (RMI). The RMI accurately classified over 98% of participants and demonstrated high sensitivity (97%) and specificity (100%) in discriminating between analog malingerers and patients. Because the RMI is calculated directly from the LMDR items, it has the advantage of requiring no additional administration time or materials, and thus may serve as a quick screen for dissimulation that can be obtained without additional testing.