measured by AQ in a clinical group of patients with schizophrenia and compare these with a group of individuals with a diagnosis of Asperger syndrome and healthy controls. Methods: The AQ is a self-administered 50-item questionnaire, assessing personal preferences and habits. Participants rate to what extent they agree or disagree with the statements on a 4- point Likert scale. All the item scores are summed: a high AQ score indicates a high autistic load. The AQ was completed by 17 patients with schizophrenia (9 male, 8 female, mean age 28), 44 patients with Asperger syndrome (23 male, 21 female, mean age 27) and 33 healthy controls (17 male, 16 female, mean age 25). Results: The two clinical groups scored significantly higher than the control group. The mean AQ score for the schizophrenia group was 24.6 (SD 5.7), for Asperger syndrome 26.9 (SD 8.8) and for healthy controls 13.4 (5.6). Discussion: Patients with schizophrenia, as well as patients with Asperger syndrome, report high scores on the AQ compared to healthy controls. The questionnaire cannot be used to differentiate between the two conditions. doi:10.1016/j.schres.2010.02.496 Poster 2 WORKING MEMORY IMPAIRMENTS IN SCHIZOPHRENIA AS VALID PREDICTORS OF ILNESS SEVERITY AND FUNCTIONAL OUTCOME Oded Meiron 1 , Abraham Weizman 2 1 Tel-Aviv University Tel-Aviv, Israel, Israel; 2 Geha Mental Health Center Petah Tikva, Israel, Israel Background: Clinical studies consistently report a close link between working memory impairments and schizophrenia (Braver et al., 1997, Keefe et al., 2004, Ritter et al., 2004, Tan et al., 2006, Lesson et al., 2007). Such working memory impairments (e.g., executive attention failure, Kane & Engle, 2002) are mainly related to dorsolateral prefrontal cortex hypofunction (Kane & Engle, 2002, Jansama et al., 2004, Tan et al., 2006). The present study attempts to bridge between neurocognitive research and the clinic by proposing a novel neuro-diagnostic tool to evaluate working memory impairments in medicated schizophrenia patients, and to support the use of executive attention differences as predictors of disease psychopathology and functional-outcome (e.g., level of recovery scale, Schrank & Slade, 2007). Methods: The present study utilizes standard and revised versions of the n-Back task (Smith et al., 1996, Krieger et al., 2005 Tan et al., 2006,) for the measurement of executive attention differences in medicated schizophrenia patients (MSZ) and in healthy individuals (HG). Prior to performing the n-Back task, MSZ (N=32) were clinically evaluated using the PANSS (Kay et al., 2004) symptom-severity rating scales by two trained clinicians. HG (N=30) provided their medical and psychiatric history prior to executive attention testing. In both study groups, the computerized task consisted of two consecutive 10-minute blocks, where the first block (e.g., control block) was always the standard version of the task and the next block (e.g., experimental blocks) was one out of three different revised versions of the n-Back task. The dependent variable (e.g., response accuracy) on the working memory blocks was number of correct responses. Results: The control block was internally consistent in both HG and MSZ. Coefficient alphas were .90, and .917, respectively. The HG had significantly greater number of correct responses than the MSZ (t = -6.03, p < .0001). Most importantly, accuracy scores on the control block predicted psychopathological dimension scores in MSZ. Speci- fically, patients who scored equal to or below the average score (i.e., lower performing group) of the control block scored significantly higher on the negative symptoms scale (t=-2.26, p<.05) than pa- tients who scored above the average (e.g., higher performing group). Moreover, there were significant correlations of accuracy scores with delusions (r = -.843, p = .017), total PANSS score (r = -.465, p = .007), and functional-outcome (r=.526, p=.002); control block accuracy scores seem to provide a valid cognitive deficit index for assessing the patient's current clinical condition. Multiple regression analysis predicting functional outcome (R=.569, SE=.76, F=4.45. p=.011) revealed that both delusion and total PANSS scores were significantly correlated only with control block accuracy scores. However, in contrast to symptom-severity scores, accuracy scores on the control block were found to be an exclusive significant predictor (b=.481, p=.02) of functional outcome. Discussion: Our results suggest that the standard version of the n-Back task is a reliable and valid measurement of working memory impairments in medicated schizophrenia patients. The present findings could support the construction of a reliable neurocognitive diagnostic tool, that could offer a clinically valid working memory impairment measurement that "specifically assesses important cog- nitive deficits" (Keefe et al., 2004) in schizophrenia. doi:10.1016/j.schres.2010.02.497 Poster 3 EXAMINATION REGARDING THE EFFECT OF INTERVENTION TO AT RISK MENTAL STATE CASES AND THE CASES OF FALSE FALSE-POSITIVE Keiko Morita 1 , Kiyoaki Takeshi 1 , Naohisa Tsujino 1 , Naoyuki Katagiri 1,2 , Hiroyuki Kobayashi 2,3 , Takahiro Nemoto 1 , Masafumi Mizuno 1 1 Department of Neuropsychiatry, Toho University School of Medicine, Tokyo, Japan; 2 Tokyo Musashino Hospital, Affiliated with a Juridical Foundation The Institute of Clinical Psychiatry, Tokyo, Japan; 3 Department of Neuropsychiatry, Keio University School of Medicine, Tokyo, Japan Background: The ratio of transition from At Risk Mental State (ARMS) to psychosis decreased from about initial 40% to more or less 10%. However, the cases which do not transit to psychosis include false false-positive cases not deteriorated to the psychosis level through intervention. As for the false positive cases, redundant intervention should not be executed, but it's extremely difficult to foreseeingly distinguish the false positive cases which will not intrinsically shift to psychosis and the false false-positive cases. At the Toho University Medical Center, Tokyo, as an integrated approach to the early intervention to psychosis, the out-patient clinic (Youth Clinic) specialized for ARMS and the day-care facility ("Il Bosco") only for youth were set up. And based on the Optimal Treatment Project (Falloon, 2004), the support by integrated approach for ARMS including managed care, CBT, psychoeducation, and pharmacotherapy with minimum dosage of atypical neurolep- tics, are being carried out by the multidisciplinary-team of specialists. Based on the hypothesis that the false false-positive cases needed more pharmacotherapy than false positive cases, the characteristics at baseline are to be examined looking back and the therapeutic process of ARMS cases which didn't shift to the psychosis is to be examined retrospectively. Methods: We observed the therapeutic process of ARMS naturally, and we examined the characteristics of ARMS cases at the baseline which didn't shift to psychosis, and the period which was required for the neuroleptic treatment. The subjects are the out-patients who were 15 to 39 years old and diagnosed as ARMS at their first visit to the Medical Center from June 1st 2007 to March 31st 2009. As for the diagnosis Abstracts 300