Electrophysiological correlates of indirect priming in normal volunteers and schizophrenic patients

Electrophysiological correlates of indirect priming in normal volunteers and schizophrenic patients

Abstracts / PsychiatryResearch: Neuroimaging68 (1997) 155-184 mean distance between corresponding centers-of-mass of SM activation was 6.4 mm (range 2...

102KB Sizes 0 Downloads 70 Views

Abstracts / PsychiatryResearch: Neuroimaging68 (1997) 155-184 mean distance between corresponding centers-of-mass of SM activation was 6.4 mm (range 2.4-10.8). We have recently completed a similar study of isotropic whole brain EPI. Using a T statistical analysis (the characteristics of EPI data do not permit a z statistic analysis), activation of primary sensorimotot cortex, SMA, and cerebellum were found in every individual, and putamenal activation was also observed in most subjects. The magnitude and extent of activation was greater than with PRESTO. Comparisons with PET showed results comparable to those of PRESTO. These studies indicate that sensorimotor activation maps based on BOLD-sensitive 3D fMRI correspond very well to changes in regional cerebral perfusion. Left lateralized P300 amplitude deficit in schizophrenic patients depends on pitch disparity M. Weisbrod, S. Winkler, S. Maier, H. Hill, Ch. Thomas, M. Spitzer

Psychiatric Hospital, University of Heidelberg, Germany Structural abnormalities of the left temporal lobe have been implicated in the genesis of schizophrenia, and left lateralized reduced P300 amplitudes in schizophrenic patients have been interpreted as their electrophysiological correlate. We examined the effect of stimulus discriminability on P300 amplitude distribution in 20 schizophrenic patients and in 20 control subjects, using two different auditory oddball paradigms. The detection of the target tone was either simple or difficult, due to manipulation of pitch disparity. The P300 was recorded at 20 electrode sites. The controls showed a symmetrical P300 amplitude distribution in both paradigms. In contrast, the P300 distribution in schizophrenic patients was asymmetrical, with smaller amplitudes over the left hemisphere, but only whcn the simple paradigm was used. Our results support the findings of a left temporal structural deficit in schizophrenic patients. The corresponding deficit in the P300 scalp voltage distribution seems to depend on thc level of difficulty in the stimulus condition. Electrophysiological correlates of indirect priming in normal volunteers and schizophrenic patients M. Wcisbrod, S. Winkler, S. Maier, M. Spitzer

183

differently in both groups. Results are in line with E. Bleuler's suggestion of a special role of indirect associations in schizophrenic psychopathology and with recent spreading activation models of schizophrenic language dysfunction Acute effects of repeated electroconvulsive therapy (ECT) on cerebral blood flow assessed by Tc-99m-HMPAO SPECT: a preliminary study M.H. Wiegand b, D. Duncan ~, A. Alavi ~', M. Stecker a, A. Loessner ~, F. Payer ~, W. BalP, L. Gyulai ~, D. Miller ~, P. Veloso~, A. Lang ~

aDepartment of Psychiatry and Nuclear Medicine, University of Pennsylvania, Philadelphia, USA. bPsychiatrischeKlinik, Technical University of Munich, German), Electroconvulsive therapy (ECT) is an effective antidepressant treatment whose mechanism of action, however, has remained unclear as yet. In the present study, the immediate responses of cerebral perfusion to electrical stimulation were investigated using 99mTc-HMPAO SPECT. Eight patients with major depression underwent a series of ECT treatments with unilateral or bilateral electrode placement. Unilateral and bilateral stimulation led to different activation patterns, always involving frontal cortical areas. In repeated ECT sessions, a decreased brain perfusion response over time to electrical stimulation was observed. This may correspond to an increasing seizure threshold during a course of ECT treatments, and may be a phenomenon-linked to the antidepressant action. The observation of time-related changes in brain function, as demonstrated in the present study, may thus contribute to better understanding the mechanisms underlying the beneficial clinical effects of ECT and other antidepressant treatments Quantitative EEG (qEEG) predicts relapse in patients with primary alcoholism - - a comparison between multivariate statistics and artificial neural network approaches G. Winterer ~', B. Kl6ppel b, A. Heinz J, M. Ziller a, L.G. Schmidt a, W.M. Herrmann a

~Psychiatric Clinic Free University Berlin, Germany. blnstitute of Mathematics and Computer Science, University, of Kassel, Germany

P~ychiatric Ho.spital, University of Heidelbe~, Germany The N400 component of event related potentials (ERPs) was investigated in schizophrenic patients and normal control subjects in a [exical decision task in which semantic distance bctwccn primc and target was varied. Prime-target relations included directly related words (hen-egg), indirectly related words (Icmon-sweet), and non-related words (sofa-wing). ERPs wclc recorded from 20 scalp electrodes. An N400 component was obtained in the non-related condition in both groups, controls ( n - 17) and schizophrenic patients (n = 12), while eithcr no or only a small component was detected in the related condition. Most notably, peak- and latency-data for the indirectly related words suggest that they are processed

In clarifying the potential use of qEEG for the prediction of relapse, we investigated 16 abstainers and 49 relapsers. One 30 s artifact-free EEG-segment was analyzed for each patient. For quantitative analysis (pre-processing) we chose spectral as well as Hjorth's parameters and the correlation dimension of Grassberger-Proccacia. Analyzing our data further, we created a discriminant function (approximated F-test for Wilk's lambda) out of five pre-selected variables, comparing six relapsers and 21 abstainers. By applying the resulting discriminant function to the remaining 28 relapsers and 10 abstainers, 72% of all cases were predicted successfully. On a trial basis, we finally compared both groups with several artificial neural network approaches (feed-forward multi-layer perceptrons,