128 Abstracts / Schizophrenia Research 102/1–3, Supplement 2 (2008) 1–279 statements in schizophrenics and depressive patients. Evidence for common and differential cognitive patterns. Psychopathology, 39, 277-285 212 – EARLY COGNITIVE RESPONSE TO TREATMENT IN FIRST EPISODE PSYCHOSIS Renata Schoeman , Bonga Chiliza, Robin Emsley Department of Psychiatry, University of Stellenbosch, Cape Town, South Africa bibitica@mweb.co.za Introduction: Cognitive impairment is well documented in schizo- phrenia, and improves to some extent with treatment. Early cognitive changes in response to antipsychotic treatment are not well doc- umented. We assessed early (12wk) cognitive changes and their relationships to psychopathology in 20 patients in an interim analysis of an ongoing study. Methods: Patients with a first episode of schizophrenia underwent MATRICS Consensus Cognitive Battery assessment at baseline, 4 and 12wks. PANSS ratings were completed. Treatment with low doses of oral and depot flupenthixol was commenced. Results: The sample consisted of 4 females and 16 males, with an average age of 21.22 years. The mean modal dose of flupenthixol was 10mg 2x weekly IMI. Two patients were withdrawn due to side-effects. The average premorbid IQ of the sample was on the 91st percentile as estimated by the Vocabulary subtest of the WAIS. Early changes in attention and concentration from baseline were present. Other cognitive domains (speed of processing, working memory, and visual learning) also improved and was correlated with changes in PANSS scores, with 62% of the variance predicted by negative symptoms at baseline. Conclusions: Low doses of a long-acting conventional antipsychotic are effective in improving some aspects of cognitive function after a first episode of schizophrenia. The response was rapid, and co- incided with improvement in positive and negative symptoms. The failure of previous studies to demonstrate cognitive benefits of conven- tional antipsychotics may have been related to excessive doses being prescribed. 213 – CHALLENGING THE WORKING MEMORY MODEL OF THOUGHT ABSTRACTION Andor Simon 1 , Katja Cattapan-Ludewig 1 , Daniel Umbricht 2 1 University of Bern, Bruderholz; 2 Translational Medicine Neuroscience, Novartis Pharma, Basel, Switzerland andor.simon@tiscalinet.ch Introduction: Disorders of thought and language are considered a core symptom of schizophrenia. Proverbs are among the most common instances of everyday non-literal communications. We hypothesised that if the WM model of thought abstraction was correct, then WM would correlate with proverb interpretation abilities in patients with first-episode psychosis, in at-risk patients and in control subjects alike. Methods: The present study investigated the relationship between WM and the ability to interpret proverbs in 42 patients with a first psychotic episode (FE), 64 patients meeting Ultra-High Risk criteria for psychosis (UHR), 25 patients meeting Basic Symptom at-risk criteria for psychosis (BS), and in 42 help-seeking patient controls (Pco). WM was assessed with the Letter Number Span test (LNS) and the 2-Back test. Results: Thought abstraction was increasingly impaired across the groups in the order Pco<BS<UHR than on the LNS. We found significant correlations between SIPS N5 scores and all WM measures in all groups with the exception of the LNS in the FE group. Conclusions: Our results support the WM model in thought abstrac- tion. However, this model does not seem to apply in FE patients for high-demanding WM tasks with elevated complexity such as the LNS. This finding suggests that the role of WM may be outweighed by other cognitive processes that come into play. 214 – A FAILURE OF CONFLICT TO MODULATE DUAL-STREAM PROCESSING MAY UNDERLIE DELUSIONAL IDEATION IN SCHIZOPHRENIA William Speechley , E. Ngan University of British Columbia, Vancouver, British Columbia, Canada willspeechley@gmail.com Introduction: Dual-stream decision-making models propose two in- teracting processes; a fast, intuitive system (Stream 1) and a slower, more logical process (Stream 2). When the two streams diverge, healthy individuals may experience conflict, which we suggest biases decision-making towards Stream 2, increasing the likelihood of a judgment in keeping with the available evidence. In schizophrenia, a failure in this conflict-modulation system may allow erroneous intuitive interpretations to endure unchallenged. Methods: Subjects (patients with schizophrenia and healthy controls) completed a simple deductive reasoning task requiring judgment of the logical validity of two-part conditional statements while undergo- ing functional magnetic resonance imaging (fMRI) in a 3T scanner. By varying the agreement between logical validity (Stream 2) and the believability of the concluding sentence (an automatic, Stream 1 judgment) we constructed conditionals that generated either consensus or conflict between the streams. Results: In piloting, patients (n = 14) performed significantly worse than controls (n = 14) when determining the validity of conditional statements for both conflict and non-conflict conditions. This dif- ference was significantly larger for the conflict condition. Controls showed no significant difference in performance between conditions. fMRI data collection is ongoing. Conclusions: The results are consistent with our hypothesis that delusional schizophrenia patients fail to use conflict to bias towards Stream 2 when the two streams of reasoning arrive at incompatible judgments. We anticipate that this will be reflected in the fMRI data through greater activation of areas related to conflict detection and logical reasoning in healthy subjects than in those with schizophrenia. References [1] Kahneman D. A perspective on judgment and choice: Mapping bounded rationality. Am Psychol 2003; 58: 697-720. [2] Epstein S. Integration of the cognitive and the psychodynamic unconscious. Am Psychol 1994; 49: 709-724. [3] Goel V, Dolan RJ. Explaining modulation of reasoning by belief. Cognition 2003; 87: B11-B22. [4] Goel V, Bartolo A, St Clair D, Venneri A. Logical reasoning deficits in schizophrenia. Schizophr Res 2004; 66: 87-8. 215 – COGNITIVE BIAS MODIFICATION IN SCHIZOPHRENIA Craig Steel 1 , Til Wykes 2 , Anna Ruddle 3 , Gina Smith 4 , Dhruvi M. Shah 5 , Emily A. Holmes 5 1 Department of Psychology, University of Reading, UK; 2 Department of Psychology, Institute of Psychiatry, London, UK; 3 Department of Psychology, University College London, UK; 4 Department of Psychology, University of Surrey, UK; 5 Department of Psychiatry, University of Oxford, Oxford, UK c.steel@reading.ac.uk Introduction: There has been recent success in developing comput- erised interventions which modify negative interpretation biases, a core psychological process underlying the development and mainte- nance of anxiety. The cognitive bias modification (CBM) paradigm involves individuals being repeatedly exposed to ambiguous informa- tion which ultimately resolves positively. The current study is the first to report data from a modified form of CBM for individuals diagnosed with schizophrenia and co-morbid anxiety.