11. Psychology, Neuro- 131 may be associated with different underlying neuropsychological processes. Further, these results may help to explain some of the inconsistency associated with previous findings of neuropsycholog- ical impairment and the positive syndrome. PERINATAL PREDICTORS OF NEUROPSYCHOLOGICAL OUTCOME IN YOUNG ADULTS BORN PRETERM M. C. Cuddy,* L. Rifkin, A. Stewart, J. Townsend, R. Mun'ay Psychological Medicine, Institute of Psychiatry, London, England, United Kingdom Recent research suggests a higher than expected rate of schizophre- nia in adults born preterml. Furthermore, individuals who become schizophrenic are more likely than healthy controls to have experi- enced obstetric complications at birth2. Schizophrenia is associated with a characteristic pattern of neuropsychogical deficits in episod- ic memory, attention, and executive functioning. These deficits appear to be present before the onset of illness3 and are thought to arise from abnormalities in brain development. Young adults born preterm show some deficits in neuropsychological functioning. These are also thought to reflect disordered neurodevelopment. Exploring the impact of perinatal status on the development of this population may help to understand their importance for individuals who later develop schizophrenia. This project is part of a longitudi- nal follow-up study investigating the outcome of very preterm birth. A cohort of children born at less than 32 weeks gestation have been followed up at the ages of 1, 4, 8, 14, and 18. Detailed perinatal infor- mation has also been collected. The results presented here concern 79 members of this cohort who underwent a detailed neuropsycho- logical assessment at the age of 17-18. This included measures of IQ, memory, attention and executive function. The relationship between perinatal variables (e.g. birthweight, gestational age, Apgar score) and neuropsychological outcome was explored through regression analysis. Perinatal status was a significant predictor of full-scale IQ in young adulthood, but was not related to any other area of neuropsychological functioning. It would seem that in young adults born preterm, perinatal variables are predictive of indices of general cognitive functioning, but not of specific neuropsychologi- cal skills. 1. Smith, G.N., MacEwan, G.W., Dunn, H.D., et al. (1992). Psychiatric sequelae of birth problems: the Vancouver Study. In: Schizophrenia 1992: Poised for Change. Programs and Abstracts from the Vancouver Conference. p. 150 2. Rifkin, L., Lewis, S., Jones, R, Toone, B., & Murray, R. (1994). Low birthweight and Schizo- phrenia. British Journal of Psychiatry, 165, 357-362. 3. Elvevag, B. & Weinberger, D. (2001). The neuropsychology of schizophrenia and its relationship to the neurodevelopmental model. In C. Nelson & M. Luciana (Eds): Handbook of Developmental Cognitive Neu- roscience. GENERAL AND SPECIFIC COGNITIVE DEFICITS IN SCHIZOPHRENIA D. Dickinson,* V. N. Iannone, C, Wilk, J. M. Gold MIRECC, VA Capitol Health Care Network, Baltimore, MD, USA It remains controversial whether deficits in specific neuropsycho- logical performance domains in schizophrenia are relatively inde- pendent of one another or are reflections of a common, generalized deficit. To address this question, a specialized structural equation modeling procedure, single common factor analysis (SCFA; Salt- house, 1994), was carried out on data from a sample including patients (n = 98) and controls (n = 95). The data were index scores from the Wechsler Adult Intelligence Scale and Wechsler Memory Scale for domains of verbal comprehension, perceptual organization, working memory, processing speed, auditory episodic memory, and visual episodic memory. Each index is derived from scores on a dis- tinct set of Wechsler scale subtests and all showed large mean dif- ferences in our sample (i.e., between schizophrenia patients and con- trois). The SCFA procedure allows these group differences to be decomposed into two parts on an index-by-index basis: one part that is shared in common with other indexes and another part that is unique to each index. One major finding of this analysis was that a large proportion of the variance associated with group membership was shared in common among the different Wechsler indexes. All the indexes showed strong relationships to a common factor (mean standardized coefficient value = 0.76; mean t-value = 11.94). By con- trast, the independent relationships between the indexes and group status (after controlling for the common factor) were of much small- er magnitude (mean standardized coefficient value = 0.11; mean t- value = 1.66). Each of the coefficients for associations between the indexes and the common factor easily met a statistical significance criterion (coefficient > 2 x SE; Salthouse, 1994). In contrast, only indexes for processing speed, verbal comprehension, and visual episodic memory showed significant independent relationships to group status. Notably, despite the cormnon belief that verbal mem- ory and working memory are differentially impaired in schizophre- nia, indexes for these domains showed no significant relationship to group status apart from what they shared with the other indexes. The common factor in this analysis is, in essence, a statistical approxi- mation of a generalized cognitive effect, and its strength relative to independent effects in our sample suggests a central role for a com- mon, generalized deficit in distinguishing cognitive functioning in people with and without schizophrenia. EFFECTS OF OLANZAPINE VERSUS CLOZAPINE ON EXECUTIVE FUNCTIONS IN SCHIZOPHRENIA A. Dittmann-Balcm,* S. Bender, U. Schall, A. Klimke, N. Mueller, U. Vorbach, K. U. Kuehn, R. W. Dittmann, D. Naber Clinic for Psychiatry and Psychotherapy, University of Essen, Essen, Germany Introduction: Compared to conventional neuroleptics, so-called atyp- ical neuroleptics are supposed to improve cognitive functions in patients with schizophrenia. However, main drawbacks of previous studies are small sample sizes, an open-label approach and the com- parison with conventional instead of other atypical agents. It is still unclear if the improvements in cognitive functions are primary or rather secondary (e.g. reduction of side-effects). Methods: In a mul- ti-center study patients with schizophrenia (DSM-IV) were assigned to randomized double-blind 26 weeks of treatment with olanzapine (n=57) versus clozapine (n=57). Executive functions (Tower of Lon- don [TOLl, Wisconsin Card Sorting [WCST], Stroop [ST]), psy- chopathology (Positive and Negative Syndrome Scale), and extrapyramidal side-effects (EPS; Simpson-Angus-Scale) were assessed in a subpopulation (olanzapine -- 30 vs. clozapine = 24) at baseline and after 4 (T1) and 26 weeks (T2) of treatment. Results: Executive functions improved significantly at T l and T2 without dif- ferences between the agents. Psychopathology and EPS improved International Congress on Schizophrenia Research 2003