Discriminating between simulated malingering and closed head injury on the Wechsler memory scale-revised

We tested the hypothesis that malingering is distinguished by poorer performance on recognition relative to recall tasks by evaluating the ability of discriminant functions to distinguish between 89 subjects simulating malingering and 44 subjects with a history of closed head injury (CHI) on the Wechsler Memory Scale - Revised (WMS-R). Functions with good accuracy in discriminating between simulated malingerers and controls in prior studies did not have adequate specificity when applied to the CHI group. A newly derived discriminant function achieved overall classification accuracy of 79% for the malingering versus CHI groups on cross-validation, with 79% sensitivity (true positives for malingering) and 80% specificity (true negatives for closed head injured). A complex performance pattern on seven WMS-R subtests distinguished malingering subjects from those with CHI, but did not support the recognition versus recall hypothesis. The malingering pattern involved: (a) power performance on two relatively (but inconspicuously) easy tasks dependent on immediate recall (Visual Reproduction I and Visual Memory Span); (b) better performance on two relatively (but inconspicuously) easy tasks dependent on immediate recall (Visual Paired Associates I and Digit Span); (c) poorer performance on two relatively difficult delayed tasks (Logical Memory II and Visual Paired Associates II); and (d) better performance on another difficult task involving delayed recall (Visual Reproduction II).