Christoph Mulert

Integration of fMRI and simultaneous EEG: towards a comprehensive understanding of localization and time-course of brain activity in target detection.

fMRI and EEG are complimentary methods for the analysis of brain activity since each method has its strength where the other one has limits: The spatial resolution is thus in the range of millimeters with fMRI and the time resolution is in the range of milliseconds with EEG. For a comprehensive understanding of brain activity in target detection, nine healthy subjects (age 24.2 +/- 2.9) were investigated with simultaneous EEG (27 electrodes) and fMRI using an auditory oddball paradigm. As a first step, event-related potentials, measured inside the scanner, have been compared with the potentials recorded in a directly preceding session in front of the scanner. Attenuated amplitudes were found inside the scanner for the earlier N1/P2 component but not for the late P300 component. Second, an independent analysis of the localizations of the fMRI activations and the current source density as revealed by low resolution electromagnetic tomography (LORETA) has been done. Concordant activations were found in most regions, including the temporoparietal junction (TPJ), the supplementary motor area (SMA)/anterior cingulate cortex (ACC), the insula, and the middle frontal gyrus, with a mean Euclidean distance of 16.0 +/- 6.6 mm between the BOLD centers of gravity and the LORETA-maxima. Finally, a time-course analysis based on the current source density maxima was done. It revealed different time-course patterns in the left and right hemisphere with earlier activations in frontal and parietal regions in the right hemisphere. The results suggest that the combination of EEG and fMRI permits an improved understanding of the spatiotemporal dynamics of brain activity.

Prefrontal Transcranial Direct Current Stimulation Changes Connectivity of Resting-State Networks during fMRI

Transcranial direct current stimulation (tDCS) has been proposed for experimental and therapeutic modulation of regional brain function. Specifically, anodal tDCS of the dorsolateral prefrontal cortex (DLPFC) together with cathodal tDCS of the supraorbital region have been associated with improvement of cognition and mood, and have been suggested for the treatment of several neurological and psychiatric disorders. Although modeled mathematically, the distribution, direction, and extent of tDCS-mediated effects on brain physiology are not well understood. The current study investigates whether tDCS of the human prefrontal cortex modulates resting-state network (RSN) connectivity measured by functional magnetic resonance imaging (fMRI). Thirteen healthy subjects underwent real and sham tDCS in random order on separate days. tDCS was applied for 20 min at 2 mA with the anode positioned over the left DLPFC and the cathode over the right supraorbital region. Patterns of resting-state brain connectivity were assessed before and after tDCS with 3 T fMRI, and changes were analyzed for relevant networks related to the stimulation–electrode localizations. At baseline, four RSNs were detected, corresponding to the default mode network (DMN), the left and right frontal-parietal networks (FPNs) and the self-referential network. After real tDCS and compared with sham tDCS, significant changes of regional brain connectivity were found for the DMN and the FPNs both close to the primary stimulation site and in connected brain regions. These findings show that prefrontal tDCS modulates resting-state functional connectivity in distinct functional networks of the human brain.

EEG-vigilance and BOLD effect during simultaneous EEG/fMRI measurement.

Different EEG-vigilance stages from full alertness to sleep onset can be separated during rest. Also fMRI research recently focused on the resting condition and identified several resting state networks. In order to deepen the understanding of different levels of global brain function from relaxed wakefulness to sleep onset the association between EEG-vigilance stages and BOLD signals was analysed. EEG-vigilance stages were attributed to consecutive 3-sec-EEG-segments by an algorithm using topographic and spectral information. Results of the classification were validated by analysing the heart rates during the different brain states. Vigilance stages served as regressors for the analysis of the simultaneously acquired fMRI data. Additionally resting state networks were derived from the fMRI data using independent component analysis (ICA). Also vigilance associated brain activity revealed by EEG-based standardized low resolution tomography (sLORETA) was compared to the results of the fMRI analysis. Results showed increased BOLD signal in the occipital cortex, the anterior cingulate cortex, the frontal cortex, the parietal cortices and the temporal cortices and decreasing BOLD signals in the thalamus and the frontal cortex for declining vigilance stages (A2, A3, B1, B2/B3) in comparison to the high vigilance stage A1. Resting state networks revealed a spatial overlap with the vigilance stage associated BOLD maps in conjunction analyses. sLORETA showed increased neuroelectric alpha activity at the occipital cortex comparable to occipital BOLD signal decreases when comparing stage A with stage B. Different EEG-vigilance stages during rest are associated with pronounced differences of BOLD signals in several brain areas which partly correspond to the resting state networks. For cognitive fMRI-research it therefore seems important to pay attention to vigilance switches in order to separate vigilance associated BOLD signal changes from those specifically related to cognition.

Prefrontal direct current stimulation modulates resting EEG and event-related potentials in healthy subjects: a standardized low resolution tomography (sLORETA) study.

Prefrontal transcranial direct current stimulation (tDCS) with the anode placed on the left dorsolateral prefrontal cortex (DLPFC) has been reported to enhance working memory in healthy subjects and to improve mood in major depression. However, its putative antidepressant, cognitive and behavior action is not well understood. Here, we evaluated the distribution of neuronal electrical activity changes after anodal tDCS of the left DLPFC and cathodal tDCS of the right supraorbital region using spectral power analysis and standardized low resolution tomography (sLORETA). Ten healthy subjects underwent real and sham tDCS on separate days in a double-blind, placebo-controlled cross-over trial. Anodal tDCS was applied for 20 min at 2 mA intensity over the left DLPFC, while the cathode was positioned over the contralateral supraorbital region. After tDCS, EEG was recorded during an eyes-closed resting state followed by a working memory (n-back) task. Statistical non-parametric mapping showed reduced left frontal delta activity in the real tDCS condition. Specifically, a significant reduction of mean current densities (sLORETA) for the delta band was detected in the left subgenual PFC, the anterior cingulate and in the left medial frontal gyrus. Moreover, the effect was strongest for the first 5 min (p<0.01). The following n-back task revealed a positive impact of prefrontal tDCS on error rate, accuracy and reaction time. This was accompanied by increased P2- and P3- event-related potentials (ERP) component-amplitudes for the 2-back condition at the electrode Fz. A source localization using sLORETA for the time window 250-450 ms showed enhanced activity in the left parahippocampal gyrus for the 2-back condition. These results suggest that anodal tDCS of the left DLPFC and/or cathodal tDCS of the contralateral supraorbital region may modulate regional electrical activity in the prefrontal and anterior cingulate cortex in addition to improving working memory performance.

Schizophrenia: reduced signal-to-noise ratio and impaired phase-locking during information processing.

OBJECTIVE This study was performed in order to clarify the mechanisms which underlie the reduced signal-to-noise of event-related potentials in schizophrenic patients. Specifically, we wanted to find out, whether it is reduced activation and/or synchronization (phase-locking) in specific frequency bands of the ongoing EEG which is related to the decreased signal amplitude and signal-to-noise ratio in schizophrenics. METHODS We investigated 41 unmedicated schizophrenics (10 of them drug-naïve) and compared them with healthy control subjects (n = 233) as well as unmedicated subjects with schizotypal personality (n = 21), who were considered to be high-risk subjects for schizophrenia, and unmedicated depressive patients (n = 71). We measured event-related activity during an acoustical choice reaction paradigm and calculated the signal-to-noise ratio, signal power and noise for a time interval of 50-200 ms after stimulus presentation. Signal-to-noise ratio was calculated from the power of the averaged trials (signal power) divided by the mean power of the single trials minus the power of the average (noise power). Also, we performed a frequency analysis of the pre- and poststimulus EEG based on a factor analytical approach. Group comparisons were performed with ANCOVA. RESULTS As expected, a decreased signal-to-noise ratio of evoked activity was found in the schizophrenic and a non-significant trend in the schizotypal subjects and the depressive patients. We were able to show that the observed decrease is due to a reduced signal power and an increase of absolute noise power. Frequency analysis of the evoked activity revealed that normals, schizophrenics schizotypal subjects and depressive patients increased theta/delta activity between pre- and poststimulus interval to a similar extend. However, this theta/delta-augmentation does not correlate with signal power in schizophrenics. Also, normals and depressive subjects augment coherence between both temporal lobes during information processing, which is not found in schizophrenics and schizotypal subjects. In contrast, these two groups augment frontal lobe coherence, which goes along with an increase of noise. CONCLUSIONS Reduced stimulus-induced phase-locking and bitemporal coherence of cortically evoked activity but not a failure to activate the cortex may be responsible for the observed low signal-to-noise ratio during information processing in schizophrenics. Accordingly, schizophrenics increase noise after stimulus presentation instead of building up a signal. This is discussed in the framework of the theory of stochastic resonance.

Sound level dependence of the primary auditory cortex: Simultaneous measurement with 61-channel EEG and fMRI

Sound level dependence has been investigated for years with event-related potentials (ERP). A serotonergic modulation of the sound level dependence only of the primary auditory cortex but not of the auditory association cortex has been suggested by a number of clinical and preclinical studies. Therefore, a precise covering of the activity of the primary auditory cortex seems necessary if sound level dependence is used as an indicator of the central serotonergic system. Recent fMRI studies described a pronounced sound level dependence only in the Heschl gyrus/primary auditory cortex but not in auditory association areas. In the present simultaneous 61-channel EEG and fMRI study investigating fourteen healthy subjects, we found a high correlation between the loudness-dependent change of the extent of fMRI activation (number of activated voxels) and the corresponding changes of the mean current source density within the same region of interest covering the primary auditory cortex (r = 0.84, P < 0.001). Our findings suggest a close relationship between the fMRI signal and event-related potential activity. In addition, the correspondence of the ERP-based data and the fMRI results further supports the validity of the ERP localization approach.

Frontal and temporal dysfunction of auditory stimulus processing in schizophrenia.

Attention deficits have been consistently described in schizophrenia. Functional neuroimaging and electrophysiological studies have focused on anterior cingulate cortex (ACC) dysfunction as a possible mediator. However, recent basic research has suggested that the effect of attention is also observed as a relative amplification of activity in modality-associated cortical areas. In the present study, the question was addressed whether an amplification deficit is seen in the auditory cortex of schizophrenic patients during an attention-requiring choice reaction task. Twenty-one drug-free schizophrenic patients and 21 age- and sex-matched healthy controls were studied (32-channel EEG). The underlying generators of the event-related N1 component were separated in neuroanatomic space using a minimum-norm (LORETA) and a multiple dipole (BESA) approach. Both methods revealed activation in the primary auditory cortex (peak latency approximately 100 ms) and in the area of the ACC (peak latency approximately 130 ms). In addition, the adapted multiple dipole model also showed a temporal-radial source activation in nonprimary auditory areas (peak latency approximately 140 ms). In schizophrenic patients, significant activation deficits were found in the ACC as well as in the left nonprimary auditory areas that differentially correlated with negative and positive symptoms. The results suggest that (1) the source in the nonprimary auditory cortex is detected only with a multiple dipole approach and (2) that the N1 generators in the ACC and in the nonprimary auditory cortex are dysfunctional in schizophrenia. This would be in line with the notion that attention deficits in schizophrenia involve an extended cortical network.

Single-trial coupling of EEG and fMRI reveals the involvement of early anterior cingulate cortex activation in effortful decision making.

While the precise role of the anterior cingulate cortex (ACC) is still being discussed, it has been suggested that ACC activity might reflect the amount of mental effort associated with cognitive processing. So far, not much is known about the temporal dynamics of ACC activity in effort-related decision making or auditory attention, because fMRI is limited concerning its temporal resolution and electroencephalography (EEG) is limited concerning its spatial resolution. Single-trial coupling of EEG and fMRI can be used to predict the BOLD signal specifically related to amplitude variations of electrophysiological components. The striking feature of single-trial coupling is its ability to separate different aspects of the BOLD signal according to their specific relationship to a distinct neural process. In the present study we investigated 10 healthy subjects with a forced choice reaction task under both low and high effort conditions and a control condition (passive listening) using simultaneous EEG and fMRI. We detected a significant effect of mental effort only for the N1 potential, but not for the P300 potential. In the fMRI analysis, ACC activation was present only in the high effort condition. We used single-trial coupling of EEG and fMRI in order to separate information specific to N1-amplitude variations from the unrelated BOLD response. Under high effort conditions we were able to detect circumscribed BOLD activations specific to the N1 potential in the ACC (t=4.7) and the auditory cortex (t=6.1). Comparing the N1-specific BOLD activity of the high effort condition versus the control condition we found only activation of the ACC (random effects analysis, corrected for multiple comparisons, t=4.4). These findings suggest a role of early ACC activation in effort-related decision making and provide a direct link between the N1 component and its corresponding BOLD signal.

Long-range synchrony of gamma oscillations and auditory hallucination symptoms in schizophrenia

Phase locking in the gamma-band range has been shown to be diminished in patients with schizophrenia. Moreover, there have been reports of positive correlations between phase locking in the gamma-band range and positive symptoms, especially hallucinations. The aim of the present study was to use a new methodological approach in order to investigate gamma-band phase synchronization between the left and right auditory cortex in patients with schizophrenia and its relationship to auditory hallucinations. Subjects were 18 patients with chronic schizophrenia (SZ) and 16 healthy control (HC) subjects. Auditory hallucination symptom scores were obtained using the Scale for the Assessment of Positive Symptoms. Stimuli were 40-Hz binaural click trains. The generators of the 40Hz-ASSR were localized using eLORETA and based on the computed intracranial signals lagged interhemispheric phase locking between primary and secondary auditory cortices was analyzed. Current source density of the 40 ASSR response was significantly diminished in SZ in comparison to HC in the right superior and middle temporal gyrus (p<0.05). Interhemispheric phase locking was reduced in SZ in comparison to HC for the primary auditory cortices (p<0.05) but not in the secondary auditory cortices. A significant positive correlation was found between auditory hallucination symptom scores and phase synchronization between the primary auditory cortices (p<0.05, corrected for multiple testing) but not for the secondary auditory cortices. These results suggest that long-range synchrony of gamma oscillations is disturbed in schizophrenia and that this deficit is related to clinical symptoms such as auditory hallucinations.

P300 and LORETA : comparison of normal subjects and schizophrenic patients

It was the aim of the present study 1) to investigate how many cortical activity maxima of scalp-recorded P300 are detected by Low Resolution Electromagentic Tomography (LORETA) when analyses are performed with high time-resolution, 2) to see if the resulting LORETA-solution is in accordance with intracortical recordings as reported by others and 3) to compare the given pattern of cortical activation maxima in the P300-timeframe between schizophrenic patients and normal controls. Current density analysis was performed in 3-D Talairach space with high time resolution i.e. in 6 ms steps. This was done during an auditory choice reaction paradigm separately for normal subjects and schizophrenic patients with subsequent group comparisons. In normal subjects, a sequence of at least seven cortical activation maxima was found between 240-420ms poststimulus: the prefrontal cortex, anterior or medial cingulum, posterior cingulum, parietal cortex, temporal lobe, prefrontal cortex, medial or anterior cingulum. Within the given limits of spatial resolution, this sequential maxima distribution largely met the expectations from reports on intracranial recordings and functional neuroimaging studies. However, localization accuracy was higher near the central midline than at lateral aspects of the brain. Schizophrenic patients less activated their cortex in a widespread area mainly in the left hemisphere including the prefrontal cortex, posterior cingulum and the temporal lobe. From these analyses and comparsions with intracranial recordings as reported by others, it is concluded that LORETA correctly localizes P300-related cortical activity maxima on the basis of 19 electrodes except for lateral cortical aspects which is most likely an edge-phenomenon. The data further suggest that the P300-deficit in schizophrenics involves an extended cortical network of the left hemisphere at several steps in time during the information processing stream.