The Simulation Index was derived from 15 error types on the WAIS-R and WMS occurring with a high degree of specificity in a criterion simulation group (n= 16) and a further five error types specific to a group of patients sustaining severe head injury (n= 16). The reliability of the scoring criteria was evaluated in terms of the rate of agreement in scoring for the presence or absence in each of the 20 error types achieved by two independent neuropsychologists. Percentage agreement rates for each of the 20 error types ranged between 83 and 100%. When identification of group membership was attempted on the basis of simple inspection of the test protocols, diagnostic hit rates were at chance level. When identification of group membership was based on Simulation Index cut-off scores, each of the raters achieved a diagnostic hit rate of 94% for each of the simulation and head-injury group patients. The results support the reliability and clinical utility of the Simulation Index.