the case in attempting to document PTSD in suspected survivors through neuropsychological assessment. Even in the case of assessing PTSD subjects for whom population-based norms are available, there are considerable difficulties to consider. PTSD is a psychiatric disorder and traditionally has not been the focus of neuropsychological assessment. Furthermore, PTSD does not conform to the classical paradigm of an analysis of identifiable brain lesions that can be confirmed by medical techniques. With an increased emphasis on and understanding of the biological mechanisms involved in psychiatric disorders generally, neuropsychological paradigms have been invoked more frequently than in the past. However, as has been pointed out, “… comparatively little has been written to date on PTSD from a neuropsychological perspective”.120 are among the possibilities to be considered. The use of neuropsychological evaluation procedures is usually indicated if there is a lack of gross neurological disturbance, reported symptoms are predominantly cognitive in nature or a differential diagnosis between brain impairment and PTSD has to be made. 303. The selection of neuropsychological tests and procedures is subject to the limitations specified above and, therefore, cannot follow a standard battery format, but rather must be case-specific and sensitive to individual characteristics. The flexibility required in the selection of tests and procedures demands considerable experience, knowledge and caution on the part of the examiner. As has been pointed out above, the range of instruments to be used will often be limited to non-verbal tasks, and the psychometric characteristics of any standardized tests will most likely suffer when populationbased norms do not apply to an individual subject. An absence of verbal measures represents a very serious limitation. Many areas of cognitive functioning are mediated through language, and systematic comparisons between various verbal and non-verbal measures are typically used in order to arrive at conclusions regarding the nature of deficits. 307. There is great variability among the samples used for the study of neuropsychological measures in post-traumatic stress. This may account for the variability of the cognitive problems reported from these studies. It was pointed out that “clinical observations suggest that PTSD symptoms show the most overlap with the neurocognitive domains of attention, memory and executive functioning”. This is consistent with complaints heard frequently from survivors of torture. Subjects complain of difficulties in concentrating and feeling unable to retain information and engage in planned, goal-directed activity. 304. What complicates matters further is evidence that significant inter-group differences in performances of non-verbal tasks have been found between relatively closely related cultures. For example, research compared the performance of randomly selected, community-based samples of 118 English-speaking and 118 Spanish-speaking elders on a brief neuropsychological test battery.118 The samples were randomly selected and demographically matched. Yet, while scores on verbal measures were similar, the Spanish-speaking subjects scored significantly lower on almost all non-verbal measures. These results suggest that caution is warranted when using nonverbal and verbal measures to assess non-English-speaking individuals, when tests are prepared for Englishspeaking subjects. 308. Neuropsychological assessment methods appear able to identify the presence of neurocognitive deficits in PTSD, even though the specificity of these deficits is more difficult to establish. Some studies have documented the presence of deficits in PTSD subjects when compared to normal controls but they have failed to discriminate these subjects from matched psychiatric controls.121, 122 In other words, it is likely that neurocognitive deficits on test performances will be evident in cases of PTSD, but insufficient for diagnosing it. As in many other types of assessment, interpretation of test results must be integrated into a larger context of interview information and possibly personality testing. In that sense, specific neuropsychological assessment methods can make a contribution to the documentation of PTSD in the same manner that they do for other psychiatric disorders associated with known neurocognitive deficits. 305. The choice of instruments and procedures in neuropsychological assessment of suspected torture victims must be left to the individual clinician, who will have to select them in accordance with the demands and possibilities of the situation. Neuropsychological tests cannot be used properly without extensive training and knowledge in brain-behaviour relations. Comprehensive lists of neuropsychological procedures and tests and their proper application can be found in standard references.119 309. Despite significant limitations, neuropsychological assessment may be useful in evaluating individ-uals suspected of having brain injury and in distinguishing brain injury from PTSD. Neuropsychological assessment may also be used to evaluate specific symptoms, such as problems with memory that occur in PTSD and related disorders. (c) Post-traumatic stress disorder 306. The considerations offered above should make it clear that great caution is needed when attempting neuropsychological assessment of brain impairment in suspected torture victims. This must be even more strongly 120 J. A. Knight, “Neuropsychological assessment in posttraumatic stress disorder”, Assessing Psychological Trauma and PTSD, J. P. Wilson and T. M. Keane, eds. (New York, Guilford Press, 1997). 121 J. E. Dalton, S. L. Pederson and J. J. Ryan, “Effects of posttraumatic stress disorder on neuropsychological test performance”, International Journal of Clinical Neuropsychology, vol. 11 (3) (1989), pp. 121-124. 118 D. M. Jacobs and others, “Cross-cultural neuropsychological assessment: a comparison of randomly selected, demographically matched cohorts of English and Spanish-speaking older adults”, Journal of Clinical and Experimental Neuropsychology, vol. 19 (No. 3) (1997), pp. 331-339. 119 O. Spreen and E. Strauss, A Compendium of Neuropsychological Tests, 2nd ed. (New York, Oxford University Press, 1998). 122 T. Gil and others, “Cognitive functioning in post-traumatic stress disorder”, Journal of Traumatic Stress, vol. 3, No. 1 (1990), pp. 29-45. 56