Doi:10.1016/s0920-9964(02)00213-x

Schizophrenia Research 63 (2003) 63 – 71 Source distribution of neuromagnetic slow-wave activity in schizophrenic patients—effects of activation Thorsten Fehr, Johanna Kissler, Christian Wienbruch, Stephan Moratti, Thomas Elbert, Hans Watzl, Brigitte Rockstroh * Department of Psychology, University of Konstanz, P.O. Box D23, D-78457 Konstanz, Germany Center for Psychiatry Reichenau, D-78467 Reichenau, Germany When slow waves in the EEG delta and theta frequency range appear in the waking state, they may indicate pathological conditions including psychopathology. The generators of focal slow waves can be mapped using magnetic source imaging. Theresulting brain maps may possibly characterize dysfunctional brain areas. The present study examined the stability of thedensity and distribution of MEG slow waves during three conditions—rest, mental arithmetic and imagery—in 30schizophrenic patients and 17 healthy controls. Schizophrenic patients displayed a higher density of delta and theta generatorsprimarily in temporal and parietal areas. The group difference was not affected by the particular conditions. The focalconcentration of delta and theta slow waves did not differ between patients with and without neuroleptic medication, whereasthe prominence of theta dipoles in the temporal area correlated with neuroleptic dosage. The relative amount of temporal slowwaves was correlated with the negative symptoms score (PANSS-N) suggesting that temporal dysfunction may be related tonegative symptomatology. Results suggest that the distribution of slow-wave activity, measured in a standardized setting, mightadd diagnostic information about brain abnormalities in schizophrenia.
D 2002 Elsevier Science B.V. All rights reserved.
Keywords: Schizophrenia; Magnetoencephalography; Delta; Theta; Dipole density; Negative symptoms Lopes da Silva, 1987; Lewine and Orrison, 1995;Matsuoka, 1990; Vieth et al., 2000; De Jongh et al., If prominent during the waking state, slow waves 2001). Since enhanced activity in lower EEG-fre- generated in a circumscribed brain region typically quency bands has also been reported in a variety of characterize pathological or dysfunctional neural tis- psychopathological conditions (for schizophrenia, see, sue. Focal slow waves appear in the vicinity of struc- tural lesions like cerebral infarcts, contusions, local 1991; Rockstroh et al., 1997), we have previously infections, tumors, developmental defects, degenera- suggested that dysfunctional brain areas in psychiatric tive defects or subdural hematomas patients might be indicated by the concentration of focalmagnetic slow waves Inschizophrenic patients, we found slow-wave generators Corresponding author. Department of Psychology, University in the magnetoencephalogram (MEG) to be more of Konstanz, P.O. Box D23, D-78457 Konstanz, Germany. Tel.:+49-7531-882085; fax: +49-7531-882891.
frequent in association cortices, mostly in temporal 0920-9964/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0920-9964(02)00213-X T. Fehr et al. / Schizophrenia Research 63 (2003) 63–71 and parietal regions Given that focal The data reported here result from a larger long- slow-wave concentration is related to neuro- term project that attempts to examine the diagnostic psychological dysfunctions in schizophrenic patients, usefulness of focal slow-wave mapping. Further out- brain mapping of focal slow-wave generators may add come will be reported elsewhere. For the resting state information about possible neurophysiological corre- only, a fraction of the presently reported data from a lates of psychopathology. However, the dependency of subsample of 16 patients was included in this phenomenon on the particular context, state of (2001). As there were no differences between the activation or mental task activity of a patient and its original 16 and the presently added 14 patients in interaction with psychoactive medication needs to be the outcome measures, they are pooled here in one thoroughly explored before further consideration from the perspective of diagnostic usefulness.
Previous reports of enhanced electric slow waves in psychiatric patients over posterior, frontal-midline 2. Methods and materials and frontal-temporal regions Fernandez et al., 1995; Iramina et al., 1996; Naka- shima and Sato, 1992) did not distinguish focalgenerators from more widespread slow activity, i.e., Thirty patients (12 females) aged 31.6 F 8.9 years they did not differentiate likely indices of patholog- with the DSM-IV diagnosis of a schizophrenic dis- ical brain activity from those that appear in the intact order were compared to 18 healthy subjects (2 brain under various conditions of activation and females). After one control subject had to be excluded mental load. A first indication that the slow waves because of artifact-contaminated data, the mean age of in schizophrenic patients may be of a particular the group of 17 controls was 32.4 F 11.2 years.
nature came from a study by All patients, inpatients of the university research which showed not only enhanced delta power in ward at the local Center of Psychiatry, were asked by schizophrenic patients, but also a different form of the psychiatrist or psychologist in charge whether they these slow waves and the embedding EEG time would be willing to participate in the study. Diagnoses course between groups.
were given by the psychiatrist/psychologist in charge On this background, the present study explored the on the bases of ICD-10 diagnostic criteria. As sum- stability of focal magnetic slow waves across three marized in the majority of patients met the different conditions of mental activation, designed to diagnosis of paranoid – hallucinatory subtype (N = 22; enhance either left (mental arithmetic) or right parieto- three patients were diagnosed as undifferentiated, two temporal (spatial imagery) brain hemispheric activa- as disorganised; two patients met the criteria of a tion. We wanted to know to what extent the topo- schizophreniform disorder and one of a schizoaffec- graphical pattern of slow-wave activity may serve as a tive disorder. Additional diagnoses—mostly of drug marker of cortical dysfunction during the waking state dependence—were given to six patients.
irrespectively of the type of the particular mental The psychopathological status of each patient was activity during which the measurement is obtained.
assessed on the day of the experiment by the psychol- Effects of task or activation would make it difficult to ogist/psychiatrist in charge by means of the Positive establish a diagnostic tool based on focal slow-wave and Negative Syndrome Scale (PANSS) mapping. Another possible complication may come 1987) (average scores PANSS-P: 15.8 F 4.9, range 8 – from the impact of medication. General slowing of 26; PANSS-N: 21.2 F 6.7, range 9 – 33; PANSS-G: EEG frequencies has been reported as a consequence 37.8 F 12.0, range 25 – 87) (see for individual of neuroleptic medication symptom scores). Twenty-one patients were under Malow et al., 1994), but also normalizing' effects neuroleptic medication at the time of the assessment, 12 receiving typical neuroleptics only, 3 atypical fore, the possible relationship between the focal neuroleptics only and 5 a combination of typical and clustering of slow waves and neuroleptic medication atypical neuroleptics (see also The patient was also addressed.
with schizoaffective disorder received a combination T. Fehr et al. / Schizophrenia Research 63 (2003) 63–71 Table 1ICD-10 diagnoses, type of medication (INN) and PANSS scores for the 30 schizophrenic patients clozapin, flupentixol clozapin, melperon lithium, chlorprothixen haloperidol, clozapin chlorprothixen, atypical fluphenazin, perphenazin F20.0, F10.1, F19.1 haloperidol perphenazin Each PANSS scale comprises seven items with scores ranging from 1 (not apparent) to 7 (maximal strength of symptom). Thus, sum scores canvary between 7 (asymptomatic) to 49.
of lithium and chlorprothixene. Two patients received did not report any history of psychiatric illness for additional anticholinergics. The average daily dosage themselves or first degree relatives, if they did not was 144.99 F 163.6 mg CPZ eq (CPZ eq were deter- report head injury or other neurological disorders mined after the atypical affecting the brain, and if they did not report to be neuroleptics used, clozapine and risperidone, were under current psychoactive medication or regular adjusted for the CPZ eq according to the same table drug use or abuse. In all subjects, handedness was with 0.9 and 8, respectively). Nine patients were assessed by a modified version of the Edinburgh unmedicated at the time of the measurement, because Handedness Questionnaire asking they had refused neuroleptic treatment until their subjects to demonstrate hand use on various actions admission. For all patients, it was ascertained that (like using a broom, brushing teeth, writing, etc.).
patients did not use any psychoactive substances other Five patients proved to be left-handed, while all than nicotine during their inpatient treatment. Dura- controls were determined as right-handed. Prior to tion of illness varied between 1 and 107 months the experiment, subjects were familiarized with the (mean 22.04 F 29.37 months).
recording environment, informed about the procedure Control subjects, recruited by announcements in and gave written consent to participate in the experi- the hospital and the university, were interviewed by a ment, for which they received a financial bonus of trained psychologist and were only accepted if they about US$10.

Fig. 1. Determination of dipole density. (a) Selection of MEG traces. Large-amplitude slow waves can be observed in a subset of channels. (b) Contour plot. The algorithm detects aregional dipolar source which can be fitted with an equivalent current dipole model (GOF > 0.90 required). (c) Localization of the dipolar activity and determination of number ofdipoles located within predefined brain regions per unit time. Note that a high goodness for the dipole fit is required only for a subset of 37 channels (e.g., within the circle indicated in(b)) and that activity may or may not appear simultaneously at other sensors.

T. Fehr et al. / Schizophrenia Research 63 (2003) 63–71 2.2. Data collection segments were determined by visual inspection. Sin-gle-equivalent current dipoles in a homogeneous Using a 148-channel whole-head neuromagnetom- sphere were fitted for each time point in the selected eter (MAGNESk 2500 WH, 4D Neuroimaging, San epochs. Only dipole fit solutions at time points with a Diego, USA), the MEG was measured during three root mean square 100 fT < (RMS=( periods of 5 min each. In the resting condition, < 300 fT and a goodness of fit (GOF) greater than 0.90 subjects were asked to relax but to stay awake and were accepted for further analysis. These restrictions not to engage in any specific mental activity. In the should ensure that neither artifacts nor small ampli- mental arithmetic condition, subjects were asked to tude biological noise would affect the results, and that translate the words of a common German folksong only dipolar fields that were generated by focal letter by letter into numbers (a corresponding to 1, b to sources were analyzed. Since artifact-free epochs 2, c to 3, etc.) and total them up. In the mental varied in length, the percentage of data time points imagery condition, subjects were asked to imagine per second that could be fitted by the dipole model in as vividly as possible walking a well-known and a particular area was submitted to the statistical recently strolled footpath, e.g., through the hospital analyses (see for an illustration of steps in data area. Breaks separated the three recording periods and allowed subjects to move. MEG recordings were The distribution of dipole density was assessed by obtained in a supine position, and subjects were asked dividing the total brain volume into 10 regions, five in to fixate upon a colored fixation mark on the ceiling each hemisphere: prefrontal, frontal, temporal, parietal of the magnetically shielded room throughout the and occipital. Effects of conditions and diagnosis on recording in order to avoid eye- and head-movement.
the pattern of dipole densities in the delta and theta A video camera installed inside the chamber allowedmonitoring the subject's behavior and compliance atany time throughout the experiment. For the mentalarithmetic condition, compliance was assessed bycomparing the result of totaling up the word sumsand the position in the song that was reached by thesubject at the end of the 5-min period. For the mentalimagery condition compliance was assessed by askingthe subject to describe the imagined tour in detail.
The MEG was recorded with a 678.17-Hz sam- pling rate, using a band-pass filter of 0.1 – 200 Hz. Forartifact control, eye movements (EOG) were recordedfrom four electrodes attached to the left and rightouter canthus and above and below the right eye. Theelectrocardiogram (EKG) was monitored via electro-des attached to the right collarbone and the lowest leftrib. A Synamps amplifier (NEUROSCAN) served forthe recording of EOG and EKG.
2.3. Data reduction and analysis For each of the three 5-min recording epochs the data were band-pass filtered by a second order filter in Fig. 2. Boxplot for the percentage of delta dipoles per second the delta (1.5 – 4.0 Hz) and theta (4.0 – 8.0 Hz) band.
(ordinate) for the group of controls and the groups of schizophrenicswith and without medication. The three different conditions The number of sample points was reduced by factor (abscissa: resting, arithmetic, imagery) produce similar differences 16 prior to further analysis, each sample point repre- between groups in temporal regions, with many schizophrenic senting an epoch of about 20.8 ms. Artifact-free time subjects exhibiting values outside the range of controls.

T. Fehr et al. / Schizophrenia Research 63 (2003) 63–71 range were evaluated by analyses of variance with thebetween-subjects factor GROUP and the within-sub-jects factors CONDITION, AREA (comparing pre-frontal, frontal, temporal, parietal, and occipital dipoledensities) and HEMISPHERE (comparing the left-and right-hemispheric areas). For interactions withdegrees of freedom larger than 1, the degrees offreedom were corrected using the Greenhouse –Geisser procedure to account for possible violationsof the sphericity assumption.
Schizophrenic patients exhibited more focal delta activity (i.e., a higher percentage of delta dipoles persecond) than controls, group differences being most Fig. 3. Correlation between the relative density of temporal delta pronounced in temporal and parietal areas [GROUP dipoles (percentage temporal dipoles per second divided by the average percentage of dipoles per second across all brain regions) and the amount of negative symptoms as measured by the PANSS- p < 0.05; GROUP, F(1,45) = 6.41, p < 0.05; for tem- N scale. The correlation is significant even if one subject with a high poral areas, GROUP, F(1,45) = 10.75, p < 0.01; for dipole density in all areas (indicated by the dark dot) is included.
parietal areas, GROUP, F(1,45) = 5.47, p < 0.05]. This result was equally prominent for all conditions, i.e., there was no interaction of group with task (and Mean ( F S.D. in brackets) percentage of delta dipoles per second of Temporal delta was equally prominent in artifact-free epochs in the two groups (p = schizophrenic patients, medicated (N = 21) and nonmedicated (N = 9) patients c = controls) for the three conditions (resting, mental arithmetic and The temporal percentage of delta dipoles per sec- Mental arithmetic Mental imagery ond relative to the total amount (averaged across all areas) correlated positively with the negative symp- toms (r = 0.42, P < 0.05; excluding one subject with a (1.00) (1.25) (1.36) relative large amount of delta dipoles in all other Right prefrontal 1.27 regions and, therefore, a lower fraction of temporal (0.98) (0.59) (1.55) activity would increase the correlation coefficient to (0.61) (0.48) (0.75) In both groups, mental arithmetic produced a (1.00) (0.45) (1.22) somewhat higher amount of focal slow waves than rest and imagery in temporal [CONDITION, (0.92) (0.77) (0.98) F(2,90) = 4.51, p < 0.05] and prefrontal [CONDI- (0.68) (0.55) (0.73) TION, F(2,90) = 4.83, p < 0.05] areas.
Patients also produced a significantly greater per- (1.32) (0.52) (0.90) centage of theta dipoles per second than controls, group differences being again most pronounced in temporal (1.20) (0.43) (0.85) and parietal areas compared to prefrontal, frontal and (0.86) (0.40) (0.73) occipital areas [GROUP AREA  HEMISPHERE, F(4,180) = 5.11, p < 0.01; GROUP, F(1,45) = 5.73, (0.71) (0.62) (0.44) p < 0.05; for temporal areas, GROUP, F(1,45) = 7.97, T. Fehr et al. / Schizophrenia Research 63 (2003) 63–71 p < 0.01; for parietal areas, GROUP, F(1,45) = 4.27, 2001) in addition to the equally often reported frontal p < 0.05]. While the prominence of temporal theta dysfunction. Indications of temporal dysfunctions activity did not differ significantly between medicated have been obtained from imaging and nonmedicated patients ( F < 1), the patients with 1995; Hirayasu et al., 1999) and event-related poten- higher daily medication displayed more pronounced temporal theta prominence (r = 0.51, p < 0.01).
If we consider volume reduction in the temporal lobe Effects of mental activation on theta dipole density were similar in both groups (interactions n.s.) with indication of dysfunction comparable to a lesion, higher density of theta dipoles during spatial imagery enhanced slow-wave activity arising from these than during the other two conditions [CONDITION, regions is a conceivable finding.
F(2,90) = 3.41, p < 0.05]. Imagery reduced and mental Patients with and without neuroleptic medication arithmetic increased parietal theta activity, while both did not differ in the temporal enhancement of slow- activation conditions reduced frontal theta activity rel- wave (delta and theta) activity, whereas temporal theta ative to rest [CONDITION  AREA, F(8.360) = prominence correlated with daily dosage of neuro- 4.92, p < 0.01].
leptics. This does not seem to support findings of ageneral slowing of EEG frequencies as a consequenceof neuroleptic medication et al., 1994), but might be considered in line with thefinding of increased EEG theta activity concomitant The present findings underscore the previous with an increase in haloperidol plasma levels in patients report of more frequent generators who responded to the treatment of focal slow waves (in the delta and theta bands) in 1991). However, pharmaco-EEG studies have mostly schizophrenic patients. This activity prevails in tem- reported evenly distributed activity poral and parietal areas. A generator within the Morikawa et al., 1997) or a ‘‘flat table distribution'' of temporal lobe may be oriented such that its volume theta activity sometimes currents project to frontal scalp regions (like, e.g., the with anterior predominance of theta bursts generators of the auditory evoked N100). In this case and Herrmann, 1996), but no relationship between EEG recordings would pick up activity over frontal temporal enhancement of slow-wave generators and regions. The group-specific distribution of focal slow medication. Therefore, we assume a rather weak impact waves appeared under all conditions of mental acti- of neuroleptic medication on the diagnosis-specific vation. Mental activation changed the pattern of focal pattern of slow-wave generators. Moreover, an indirect slow-wave activity relatively little, only in prefrontal relationship between focal slow waves and severity of areas, and to a similar degree, whereas the prominent illness is indicated by the negative symptom score, difference between groups in temporal brain regions while the impact of neuroleptic medication was not was not affected at all. Thus, a disease-specific pattern supported by a significant correlation between the two of focal slow-wave activity and the effect of activation latter measures.
on slow waves seem to add together. It might be Group differences were also statistically mean- possible that characteristics in the course of brain ingful for the theta band. However, the magnitude of activity over time might allow to tease apart these effects was less pronounced, possibly, because focal two different types of slow-wave activity, given that theta activity was affected by medication and the nonlinear measures add to the differentiation between type of task. Increased power in the EEG theta frequency band has been found with different mental The accentuation of focal slow wave in the tem- tasks including, for instance, the encoding of mate- poral regions correlated with the negative symptoms rial that was later successfully retrieved score. If these focal waves do indicate dysfunctional al., 1996, 1997, 2001; Doppelmayr et al., 1998; brain tissue, then the present results add to structural Yamamoto and Matsuoka, 1990). Although the type and functional evidence of temporal abnormalities in of task differs between those studies and the present one, a common element in memory encoding and T. Fehr et al. / Schizophrenia Research 63 (2003) 63–71 retrieval and the imagination of a well-known path R.D., Harsh, J.R. (Eds.), Sleep Onset. APA Press, Washington, from memory might be associated with enhanced pp. 201 – 218.
Canive, J.M., Lewine, J.D., Edgar, J.C., Davis, J.T., Torres, F., theta activity. Thus, the present analysis confirms Roberts, B., Graeber, D., Orrison, W.W., Tuason, V.B., 1996.
that mental imagery activates theta sources and Magnetoencephalographic assessment of spontaneous brain ac- indicates that the sources of this activity are con- tivity in schizophrenia. Psychopharmacol. Bull. 32, 741 – 750.
fined to prefrontal brain areas.
Canive, J.M., Lewine, J.D., Edgar, J.C., Davis, J.T., Miller, G.A., Tasks had been selected to specifically activate Torres, F., Tuason, V.B., 1998. Spontaneous brain magneticactivity in schizophrenic patients treated with aripiprazole. Psy- either the left or the right hemisphere. Contrary to chopharmacol. Bull. 34, 101 – 105.
our expectation though, the distribution of delta and Czobor, P., Volavka, J., 1991. Level of haloperidol in plasma is theta dipole densities was not asymmetrical and did related to electroencephalo-graphic findings in patients who im- not vary with the task in a hemisphere-specific man- prove. Psychiatry Res. 42, 129 – 144.
ner. This suggests that delta and theta dipole densities Deicken, R.F., Calabrese, G., Merrin, E.L., Vinogradov, S., Fein, G., Weiner, W., 1995. Asymmetry of temporal lobe phosphorous are not task-related—like activity in the gamma band.
metabolism in schizophrenia: a phosphorous magnetic reso- Thus, we have to conclude the tasks selected for nance spectroscopy study. Biol. Psychiatry 38, 279 – 286.
hemisphere-specific mental activation and the meas- De Jongh, A., de Munck, J.C., Baayen, J.C., Jonkman, E.J., Hee- ure selected as indication of (dys)function did not fit thaar, R.M., van Dijk, B.W., 2001. The localization of sponta- most favorably.
neous brain activity: first results in patients with cerebraltumors. Clin. Neurophysiol. 112, 378 – 385.
In sum, the present results support previous EEG Doppelmayr, M., Klimesch, W., Schwaiger, J., Auinger, P., Winkler, and MEG studies regarding the evidence of an acti- T., 1998. Theta synchronization in the human EEG and episodic vation-invariant pattern of prominent temporal slow- retrieval. Neurosci. Lett. 20, 41 – 44.
wave activity in schizophrenics. The distribution of Elbert, T., Lutzenberger, W., Rockstroh, B., Berg, P., Cohen, R., focal slow-wave clusters might add diagnostic infor- 1992. Physical aspects of the EEG in schizophrenics. Biol. Psy-chiatry 32, 595 – 606.
mation, the more so as recent results from our labo- Faux, S.F., Nestor, P.G., McCarley, R.W., Shenton, M.E., Horvath, ratory indicate a diagnostic specificity of the presently T., Davis, K., 1990. P300 asymmetries in unmedicated schizo- reported pattern for schizophrenia when contrasted phrenics. In: Brunia, C.H.M., Gaillard, A.W.K. (Eds.), Psycho- physiological Brain Research. Tilburg Univ. Press, Tilburg, pp.
publication). An abnormal activity in the temporal 209 – 212.
Fehr, T., Kissler, J., Elbert, T., Wienbruch, C., Moratti, S., Rock- association cortices seems to vary with negative stroh, B., 2001. Source distribution of neuromagnetic focal slow symptoms, while it seems possible that abnormal waves and MEG-delta activity in schizophrenic patients. Biol.
activity in other areas could represent other aspects Psychiatry 50, 108 – 116.
of the schizophrenic symptomatology. It might be Fernandez, T., Harmony, T., Rodroguez, M., Bernal, J., 1995. EEG worthwhile to examine the clinical usefulness of this activation patterns during the performance of tasks involvingdifferent components of mental calculation. Electroencephalogr.
Clin. Neurophysiol. 94, 175 – 182.
Hasan, J., Broughton, R., 1994. Quantitative topographic EEG map- ping during drowsiness and sleep onset. In: Ogilvie, R.D., Harsh, J.R. (Eds.), Sleep Onset. APA Press, Washington, pp.
219 – 236.
This research was supported by the Deutsche Hirayasu, Y., Shenton, M.F., Salisbury, D.F., Frumin, M., Fischer, L.A., Farrell, D., Yurgelu-Todd, D.A., Zarate, C., McCarley, Forschungsgemeinschaft and the Volkswagen-Stif- R.W., 1999. Progressive change in posterior superior temporal tung. We thank Drs. K. Pro¨pster, B. Schuller and P.
gyrus in schizophrenia. Biol. Psychiatry 45S, 117S, abstract.
Ro¨ssner for accomplishing the diagnostics and clinical Iramina, K., Ueno, S., Matsuoka, S., 1996. MEG and EEG topog- status of the patients.
raphy of the frontal midline theta rhythm and source localiza-tion. Brain Topogr. 8, 329 – 331.
Jahn, T., Mussgay, L., 1989. Die statistische Kontrolle mo¨glicher Medikamenteneinflu¨sse in experimentalpsychologischen Schiz- ophreniestudien. Z. Klin. Psychol. 18, 257 – 267.
Johnstone, E.C., Owens, D.G., Crow, T.J., Frith, C.D., Alexandrop- Badia, P., Wright, K., Wauquier, A., 1994. Fluctuation in single- olis, K., Bydder, G., Colter, N., 1989. Temporal lobe structure as Hertz EEG activity during the transition to sleep. In: Ogilvie, determined by nuclear magnetic resonance in schizophrenia and T. Fehr et al. / Schizophrenia Research 63 (2003) 63–71 bipolar affective disorder. J. Neurol., Neurosurg. Psychiatry 52, on appearance of frontal midline theta activity in EEG. J. Hu- 736 – 741.
man Ergol. 21, 201 – 206.
Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and neg- Niedermeyer, E., Lopes da Silva, F., 1987. Electroencephalography: ative syndrome scale (PANSS) for schizophrenia. Schizophr.
Basic Principles, Clinical Applications and Related Fields. Ur- Bull. 13, 261 – 276.
ban & Schwarzenberg, Baltimore.
Klimesch, W., Doppelmayr, M., Russegger, H., Pachinger, T., 1996.
Oldfield, R., 1971. The assessment and analysis of handedness: the Theta band power in the human scalp EEG and the encoding of Edinburgh Questionnaire. Neuropsychologia 9, 97 – 113.
new information. NeuroReport 7, 1235 – 1240.
Rockstroh, B., Elbert, T., Berg, P., 1997. Die Untersuchung von Klimesch, W., Doppelmayr, M., Schimke, H., Ripper, B., 1997.
Strukturen und Funktionen des Gehirns in der experimentellen Theta synchronization and alpha desynchronization in a memo- Psychopathologie am Beispiel der Schizophrenien. In: Rock- ry task. Psychophysiology 34, 169 – 176.
stroh, B., Elbert, T., Watzl, H. (Eds.), Impulse fu¨r die Klinische Klimesch, W., Doppelmayr, M., Stadler, W., Pollhuber, D., Suu- Psychologie. Hogrefe, Go¨ttingen, pp. 1 – 29.
seng, P., Rohm, D., 2001. Episodic retrieval is reflected by a Rockstroh, B., Wienbruch, C., Moratti, S., Kissler, J., Fehr, T., process specific increase in human electroencephalographic the- Elbert, T., 2000. Magnetic source imaging of slow wave activity ta activity. Neurosci. Lett. 13, 49 – 52.
in psychiatric samples. Proceedings of the 12th Int. Conference Koshino, Y., Murata, I., Morata, T., Omori, M., Hamada, T., Miya- on Biomagnetism, Helsinki, pp. 395 – 397.
goshi, M., Isaki, K., 1993. Frontal intermittent delta activity in Saletu, B., Kuefferle, B., Grunberger, J., Foldes, P., Topitz, A., schizophrenic patients receiving antipsychotic drugs. Clin. En- Anderer, P., 1994. Clinical EEG mapping and psychometric cephalogr. 24, 13 – 18.
studies in negative schizophrenia: comparative trials with ami- Kubicki, S., Herrmann, W.M., 1996. The future of computer-assis- sulpiride and fluphenazine. Neuropsychobiology 29, 125 – 135.
ted investigation of the polysomnogram: sleep microstructure.
Schober, F., Schellenberg, R., Dimpfel, W., 1996. Reflection of Clin. Neurophysiol. 13, 285 – 294.
mental exercise in the dynamic quantitative topographical Lewine, J.D., Orrison, W.W., 1995. Magnetoencephalography and EEG. Neuropsychobiology 31, 98 – 112.
magnetic source imaging. In: Orrison, W.W., Lewine, J.D., Shagass, C., 1991. EEG studies of schizophrenia. In: Steinhauer, Sanders, J.A., Hartshorne, M.F. (Eds.), Functional Brain Imag- S.R., Gruzelier, J.H., et al., (Eds.), Neuropsychology, Psycho- ing. Mosby-Year Book, St. Louis, pp. 404 – 413.
physiology, and Information Processing. Handbook of Schizo- Malow, B.A., Reese, K.B., Sato, S., Bogard, P.J., Malhotra, A.K., phrenia. Elsevier, Amsterdam, pp. 39 – 69.
Su, T.P., Pickar, D., 1994. Spectrum of EEG abnormalities dur- Shenton, M.F., Dickey, C.C., Frumin, M., McCarley, R.W., 2001. A ing clozapine treatment. Electroencephalogr. Clin. Neurophy- review of MRI findings in schizophrenia. Schizophr. Res. 15, siol. 91, 205 – 211.
Matsuoka, S., 1990. Theta rhythms: state of consciousness. Brain Vieth, J.B., Kober, H., Ganslandt, O., Mo¨ller, M., Kamada, K., Topogr. 3, 203 – 208.
2000. The clinical use of MEG activity associated with brain McCarley, R.W., Shenton, M.E., O'Donnell, B.F., Faux, F.S., Kiki- lesions. Proceedings of the 12th Int. Conference on Biomagnet- nis, R., Nestor, P.G., Jolesz, F.A., 1993. Auditory P300 abnor- ism, Helsinki, pp. 387 – 394.
malities and left posterior superior temporal gyris volume Wienbruch, C., Moratti, S, Vogel, U., Fehr, T., Kissler, J., Elbert, T., reduction in schizophrenia. Arch. Gen. Psychiatry 50, 190 – 197.
Schiller, A., Rockstroh, B. Source distribution of neuromagnetic McCarley, R.W., Wible, C.G., Frumin, M., Hirayasu, Y., Levitt, J.J., slow wave activity in schizophrenic and depressive patients, Fischer, I.A., Shenton, M.E., 1999. MRI anatomy of schizophre- submitted for publication.
nia. Biol. Psychiatry 45, 1099 – 1119.
Winterer, G., Herrmann, W.M., 1995. U ¨ ber das Elektroenzephalo- Morikawa, T., Hayashi, M., Hori, T., 1997. Auto power and coher- gramm in der Psychiatrie: Eine kritische Bewertung. Z. Elek- ence analysis of delta – theta band EEG during waking – sleeping troenzephalogr., Elektromyogr. Verw. Geb. 26, 19 – 37.
transition. Electroencephalogr. Clin. Neurophysiol. 103, 633 – Yamamoto, S., Matsuoka, S., 1990. Topographic EEG study of visual display terminal (VDT) performance with special refer- Nakashima, K., Sato, H., 1992. The effects of various mental tasks ence to frontal midline theta waves. Brain Topogr. 2, 257 – 267.

Source: http://www.exzellentforschen.de/psychologie/personen/ae02/KisslerPDFs/FehrKissler_et_al_SZres2003.pdf