16. Electrophysiology 253 its locus of generation. Method: Groups of patients with schizo- phrenia (n= 12) and matched controls (n= 12) were studied. An MMN auditory oddball design was generated using the image encoding gra- dients of the functional Magnetic Resonance Imaging (fMRI) scan- ner as stimuli, thus avoiding any interfering background noise. The stimuli comprised of amplitude and duration deviants in a random- ized sequence. The scanner noise was recorded and applied to the same subjects in a whole-head Magnetoencephalography (MEG) device. Neuromagnetic and haemodynamic responses to the identi- cal stimuli were compared between the groups. Results: As expect- ed, neuromagnetic mismatch fields were smaller in the patient group (p < 0.001). More specifically, a lateralisation of duration mismatch to the right was only found in controls (between groups p < 0.05). For the relative amplitude of the BOLD effect, differences were only found in the secondary auditory cortex. Duration deviants achieved a right hemispheric advantage only in the control group (p < 0.001, between group difference n.s.). In contrast, a significantly stronger lateralisation to the left (both groups p < 0.001) was found in the patients for the amplitude mismatch (between groups p < 0.01). Con- clusions: The MEG data supports the view of a altered hemispheric interaction in the formation of short term memory traces, which are necessary for the integration of auditory stimuli. This process is pre- dominantly mediated by the secondary auditory cortex. Altered hemispheric interaction of regions within the superior temporal gyrus might be in part responsible for language mediated cognitive and psychopathological symptomatology in schizophrenia. SENSORY GATING IN INFANTS M. A. Kisley,* S. D. Polk, S. K. Hunter, R. G. Ross, R M. Levisohn, R. Freedman Psychiatry, University of Colorado Health Sciences Centel; Denver, CO, USA Physiological measures of sensory gating hold promise for tracking neurodevelopment associated with schizophrenia, and for assessing subsequent risk of developing the illness. The present study was designed to determine whether the neural circuits underlying senso- ry gating are functional in the early postnatal period in the general population. Sensory gating was measured from vertex (Cz) with a paired-click paradigm (0.5 sec inter-click interval, 10 sec inter-pair interval) during REM sleep that occurred during a daytime nap. Twenty-one infants and their mothers were enrolled. Of these, 10 infants were excluded because they did not exhibit sufficient REM sleep (15 continuous minutes) to allow for evoked potential analysis. The remaining 11 infants ranged from 45 to 55 weeks conceptual age (approx. 1-4 months post-term), and were all healthy. Gating was quantified by the T/C ratio, a ratio of the magnitude of average evoked potential component P1 (i.e., P50) evoked by the second, or test, click divided by the magnitude of component P1 evoked by the first, or conditioning, click of the pair. Sleep staging was achieved by visual inspection of 20 sec epochs of ongoing EEG, EOG, EMG, and respiration signals. As a group, the infants exhibited significant sen- sory gating (mean T/C = 0.57, less than 1 by t-test, p<0.05). How- ever, across the age range studied, there was a correlation between advancing age and strength of response suppression (r=0.77, p<0.01). No variables associated with sleep, and no variables associated with prenatal or postnatal environment could explain this correlation. However, the sample size may have been too small to detect such effects. Also, spectral analysis of ongoing EEG confirmed that REM sleep physiology changes across the developmental period studied. In conclusion, the neural system responsible for P1 sensory gating appears to be functional very early in postnatal development, at least during paradoxical sleep. Applicability to the waking state is yet unproven, but previous studies have shown that sensory gating meas- ures for adults are similar between REM sleep and waking. There- fore, we believe that measurements of sensory gating during REM sleep in infants can be useful to assess the relative contributions of genes and environment to the early development of neural circuitry responsible for sensory inhibition, particularly as it relates to risk for development of schizophrenia. IMPAIRED PREPULSE INHIBITION OF THE STARTLE REFLEX IN PATIENTS WITH PRODROMAL SYMPTOMS OF SCHIZOPHRENIA I. Koev,* K. S. Cadenhead Psychiatry, UCSD, San Diego, CA, USA Prepulse inhibition (PPI) of the acoustic startle reflex is an opera- tional measure of sensorimotor gating. Patients diagnosed with schiz- ophrenia have typically shown reduced prepulse inhibition when compared to normal subjects. The aim of the present study was to assess whether patients who meet criteria for prodromal symptoms of schizophrenia, as assessed by the Structured Interview for Pro- dromal Symptoms (SIPS), and who may be at risk for developing the illness, exhibit reduced PPI relative to normal controls. The study examined PPI in 36 normal subjects and 30 patients showing pro- dromal symptoms of schizophrenia. We used EMG recordings of the obicularis oculi muscle of both eyes to measure the subjects' startle reflex. The PPI paradigm included three interstimulus intervals (30, 60, and 120 ms). Individuals manifesting prodromal symptoms had PPI deficits bilaterally when compared to normal subjects. The out- come of this study supports previous findings that PPI of the startle reflex is an important neurobiological marker for schizophrenia spec- trum disorders. Longitudinal assessment is needed to determine whether PPI deficits in subjects exhibiting early schizophrenia symp- toms are predictive of the development of the disorder. DIFFERENT PATTERN OF COGNITIVE DYSFUNCTION IN SCHIZOPHRENIA AND BIPOLAR AFFECTIVE DISORDER S. Krljes,* S. R. Hirsch, T. Ba|deweg Psychiatry, Imperial College, London, England, United Kingdom The present study used psychophysiological and neuropsychologi- cal tests to investigate the specificity of cognitive deficits in schizo- phrenia by comparing their cognitive profile with that of bipolar patients and healthy controls. Patients with schizophrenia and bipo- lar affective disorder share a number of clinical and biological fea- tures. Thus, a few studies have questioned whether these two disor- ders are distinct illnesses, or represent a part of a continuum of psychosis. The results remained inconclusive. The present study was set to investigate whether there are distinct patterns of cognitive dys- function across different domains in these two disorders. The event related potential mismatch negativity (MMN), a measure of preat- tentive information processing, was obtained in 45 schizophrenic patients, 19 Bipolar patients, and 45 age matched healthy controls. In addition, subjects underwent a battery of neuropsychological tests including NART, Quick test, Digit Span Backward and Forward, Rivermead Behavioural Memory Test, and Verbal Fluency. Overall, results show that both clinical groups exhibit cognitive impairment when compared to a healthy control sample. However, the findings International Congress on Schizophrenia Research 2003