Detection of very high correlation in the alpha band between temporal regions of the human brain using MEG

doi:10.1016/j.neuroimage.2004.04.016

NeuroImage

Volume 22, Issue 4, August 2004, Pages 1432-1437

https://doi.org/10.1016/j.neuroimage.2004.04.016 Get rights and content

Abstract

It is generally believed that alpha band (8–12 Hz) electric and magnetic activity in the area of the left and right temporal regions in the human brain are at best poorly correlated. There are no previous reports of very high alpha band correlation between left and right temporal regions by magnetoencephalography (MEG) or electroencephalography (EEG). We present whole head magnetoencephalography (MEG) results that demonstrate that, for temporal channels in the majority of healthy subjects tested, the alpha band signals are highly to very highly correlated and are antiparallel in direction. A correlation as high as −0.97 was found for a limited time in one subject. We suggest that the correlation found may be the consequence of strong direct or indirect coupling between homologue areas in left and right temporal regions rather than a common source. The correlation may provide a valuable index of loss of connectivity in the brain due to disease as well providing valuable insight to brain function and deserves further investigation.

Introduction

Highly correlated anterior–posterior electroencephalography (EEG) signals in the alpha band (8–12 Hz) have been reported for some time (e.g. Leocani et al., 2000, Manshanden et al., 2002, Nunez et al., 1997). In some cases, correlations as high as 0.9 have been measured. But there have been no reports of high or very high correlation between left and right temporal signals in the alpha band. Indeed, although there are many publications on synchronization measures between various regions of the brain, there are few publications by any sort of synchronization measures—be it coherence, phase synchronization, correlation or any one of a host of other synchronization measures—between the temporal regions.

A few publications do discuss results from inter-temporal region synchronization measures using EEG. Duffy et al (1996) studied the age effects of interhemispheric EEG coherence including between temporal regions in the alpha band for both eyes-open and eyes-closed brain states. They found the alpha band inter-temporal region coherence to be weak. Winterer et al (2001) studied EEG coherence with eyes closed as part of a genetic risk for schizophrenia study. They reported very high coherence in the delta band between posterior temporal lobe regions but did not report similarly high coherence for the alpha band. Moorehouse et al. (2002) found low inter-temporal coherence during sleep in adolescent girls at high risk for depression in the EEG theta and delta bands. Knyazeva et al (1999) found that a single grating that extended across the vertical meridian significantly increased the EEG interhemispheric coherence in normal adult subjects in the gamma band.

As is well known, there are several sources of alpha band signals in the human brain. The most familiar is probably the alpha rhythm, which emanates from the visual cortex. However, the mu and tau rhythms are also in the alpha band. Detection of reactive magnetic rhythm near 10 Hz in the human auditory cortex has also been reported but signal correlation between different magnetoencephalography (MEG) channels was not considered and the frequency range (6.5–9.5 Hz) is different than the one used herein (Lehtela et al., 1997).

Using MEG, we have detected epochs of high to very high correlation between at least a few left and right temporal MEG channels in most of 10 healthy controls during an eyes-closed no-task condition. In two of the subjects (204 and 210), the correlation was very high for extended periods and extended superior to the temporal regions. With further study, the detected correlation between temporal regions may be useful in understanding the communication between temporal regions in healthy individuals. Also, loss of correlation can be indicative of loss of connectivity in the brain. Loss of connectivity can be an indication of disease and other sources of neurological deficits. Thus, the detected high correlation may be of help in studying and monitoring when loss of connectivity is a problem including diseases such as multiple sclerosis.

Section snippets

Methods

Ten healthy volunteers (age 23–48 years, four females) were scanned for 5 min with eyes closed and then for 3 min with eyes open. They were seated in a 151-channel MEG scanner (CTF Systems, Vancouver, Canada; Vrba, 1996). The no-task eyes-closed data were acquired after 15 min of visually evoked field data using the standard reversing checkerboard pattern (Niedermeyer and Lopes da Silva, 1993). Electrooculography (EOG) data were acquired simultaneously on all subjects.

The subjects were scanned

Results

Fig. 1 shows the correlation maps for each of the subjects for the 4-s epochs extracted from near the start of each subject's scan. An expected feature common to all the subjects is a high positive correlation along the midline of each map. This high positive correlation is at least in part due to physical adjacent coils picking up the same signal (similar to the volume conduction effect).

Moving away from the midline, the correlations change from positive values to negative ones. The map for

Discussion

When left and right temporal channels of the MEG scanner produce signals which are nearly mirror images of each other, such as in Fig. 2, it is natural to ask if the correlation could be due to some sort of artifact.

However, many characteristics of the data demonstrate the correlation is physiological in nature:

(1)
the alpha band power and correlation drops substantially when the eyes are opened as shown in Fig. 2, Fig. 3, Fig. 4;
(2)
weekly calibrations of our MEG scanner show negligible cross talk

Conclusions

Most of the 10 healthy volunteers undergoing a MEG scan with eyes closed clearly demonstrated high correlation between left and right temporal channels. It is suggested the temporal correlation may be due to anterior-pointing current dipoles in each superior temporal gyrus and adjacent regions that are tightly correlated by direct or indirect cortical–cortical interconnections. Monitoring the loss of correlation between the temporal lobes may provides a valuable way of monitoring disease and

Acknowledgements

We thank H. Vrenken, J.J.G. Geurts, B. van Oosten, B. Jelles and C.H. Polman for their input. We would also like to thank Peter Jan Ris, Ilonka Zuiderwijk, Geert deVos and Jeroen Verbunt. K.S.C. was funded by Drs. D.W. Paty and D.K.B. Li of the MS/MRI group at the University of British Columbia and VSM (CTF) Systems Inc.

References (15)

F.H. Duffy et al.
Effects of age upon interhemispheric EEG coherence in normal adults
Neurobiol. Aging
(1996)
L. Lehtela et al.
Evidence for reactive magnetic 10Hz rhythm in the human auditory cortex
Neurosci. Lett.
(1997)
I. Manshanden et al.
Source localization of MEG sleep spindles and the relation to sources of alpha band rhythms
Clin. Neurol.
(2002)
P.L. Nunez et al.
EEG coherency. 1. Statistics, reference electrode, volume conduction, Laplacians, cortical imaging, and interpretation at multiple scales
Electroencephalogr. Clin. Neurophysiol.
(1997)
G. Winterer et al.
An association between reduced interhemispheric EEG coherence in the temporal lobe genetic risk for schizophrenia
Schizophr. Res.
(2001)
D.G. Altman
Practical Statistics for Medical Research
(1991)
M.G. Knyazeva et al.
Visual stimulus-dependent changes in interhemispheric EEG coherence in humans
J. Neurophysiol.
(1999)

There are more references available in the full text version of this article.

Cited by (20)

Alpha and gamma EEG coherence during on-task and mind wandering states in schizophrenia
2023, Clinical Neurophysiology
Citation Excerpt :
In our study, the differences in EEG coherence between groups reside in different cortical regions. It has been suggested that alpha activity reflects long-range cortical interactions (Nunez et al., 2001; Von Stein et al., 2000) and alpha activity is typically found to be highly correlated between temporal areas (Cover et al., 2004). In alpha band coherence analysis, we found an increase in coherence during the attentional processing of the task for schizophrenic located in the fronto-central and fronto-temporal regions, whereas the control group in this condition was characterized by a more distributed increase in coherence throughout the cortex encompassing the central-posterior regions.
Electroencephalographic (EEG) coherence is one of the most relevant physiological measures used to detect abnormalities in patients with schizophrenia. The present study applies a task-related EEG coherence approach to understand cognitive processing in patients with schizophrenia and healthy controls.
EEG coherence for alpha and gamma frequency bands was analyzed in a group of patients with schizophrenia and a group of healthy controls during the performance of an ecological task of sustained attention. We compared EEG coherence when participants presented externally directed cognitive states (On-Task) and when they presented cognitive distraction episodes (Mind-Wandering).
Results reflect cortical differences between groups (higher coherence for schizophrenia in the frontocentral and fronto-temporal regions, and higher coherence for healthy-controls in the postero-central regions), especially in the On-Task condition for the alpha band, compared to Mind-Wandering episodes. Few individual differences in gamma coherence were found.
The current study provides evidence of neurophysiological differences underlying different cognitive states in schizophrenia and healthy controls.
Differences between groups may reflect inhibitory processes necessary for the successful processing of information, especially in the alpha band, given its role in cortical inhibition processes. Patients may activate compensatory inhibitory mechanisms when performing the task, reflected in increased coherence in fronto-temporal regions.
Reconstruction of EEG from limited channel acquisition using estimated signal correlation
2016, Biomedical Signal Processing and Control
Citation Excerpt :
Interhemispheric coherence in the gamma band has been studied in normal adults in [6]. Existence of very high correlation, between temporal regions of the human brain, in the alpha band has been reported in [7]. The primary interest in understanding inter-channel correlation in multi-channel EEG is to identify the scope for information redundancy in a measurement involving the full set of electrodes.
Nearby scalp channels in multi-channel EEG data exhibit high correlation. A question that naturally arises is whether it is required to record signals from all the electrodes in a group of closely spaced electrodes in a typical measurement setup. One could save on the number of channels that are recorded, if it were possible to reconstruct the omitted channels to the accuracy needed for identifying the relevant information (say, spectral content in the signal), required to carry out a preliminary diagnosis. We address this problem from a compressed sensing perspective and propose a measurement and reconstruction scheme. Working with publicly available EEG database, we have demonstrated that up to 12 channels in the 10-10 system of electrode placement can be estimated within an average error of 2% from recordings of the remaining channels. As a limiting case, all the channels of the 10-10 system can be estimated using recordings on the sparser 10-20 system within an error of less than 20% in each of the significant bands: delta, theta, beta and alpha.
Interhemispheric EEG coherence is reduced in auditory cortical regions in schizophrenia patients with auditory hallucinations
2013, International Journal of Psychophysiology
Citation Excerpt :
It is also reduced in many people by performing cognitive tasks (Nunez and Srinivasan, 2006). It has been hypothesised that alpha-activity reflects long-range cortical interactions (Nunez et al., 2001; von Stein and Sarnthein, 2000) and alpha activity is typically observed to be highly correlated between homologous temporal areas (Cover et al., 2004). Cortical neurons discharging in synchronous rhythms within the high beta and gamma frequency ranges occur in response to sensory stimuli over a range of modalities.
Central auditory processing has been reported to be impaired in schizophrenia patients who experience auditory hallucinations, and interhemispheric transfer in auditory circuits may be compromised. In this study, we used EEG spectral coherence to examine interhemispheric connectivity between cortical areas known to be important in the processing of auditory information. Coherence was compared across three subject groups: schizophrenia patients with a recent history of auditory hallucinations (AH), schizophrenia patients who did not experience auditory hallucinations (nonAH), and healthy controls (HC). Subjects listened to pure tone and word stimuli while EEG was recorded continuously. Upper alpha and upper beta band coherence was calculated from six pairs of electrodes located over homologous auditory areas in the left and right cerebral hemispheres. Significant between-group differences were found on four electrode pairs (C3–C4, C5–C6, Ft7–Ft8 and Cp5–Cp6) in the upper alpha band. Relative to both the HC and nonAH groups, coherence was lower in the AH patients, consistent with the hypothesis that interhemispheric connectivity is reduced in these patients.
EEG alpha phase shifts during transition from wakefulness to drowsiness
2012, International Journal of Psychophysiology
Citation Excerpt :
Classical spectral analysis (Värri et al., 1992; Jung et al., 1997; Morikawa et al., 1997; Cantero et al., 1999; Cantero and Atienza, 2000; Bódizs et al., 2008) and nonlinear techniques (Accardo et al., 1997; Pereda et al., 1998, 1999; Kobayashi et al., 1999; Acharya et al., 2005; Šušmáková and Krakovská, 2008; Bojić et al., 2010) revealed that various brain regions differ electrophysiologically during the wake-to-sleep transition. Three groups of different but complementary results could be summarized based on the most frequently used methods: a) those using Fourier amplitude or power topography (Ota et al., 1996; Jung et al., 1997; Cantero and Atienza, 2000) where changes in the activity of separate cortical regions were measured; b) analyses which reveal strength of intracortical connections between pairs of EEG channels, such as correlation and coherence, (Tanaka et al., 1996; Cantero et al., 2002; Cover et al., 2004; Koenig et al., 2005); c) phase, synchronicity and time delay measurements which offer an insight into time-related electrophysiological events between different cortical regions (Govindan et al., 2005, 2006; Jann et al., 2009). While the first two groups of methods are more frequent, not many studies cover the third issue, particularly in experimental paradigms related to different spontaneous brain states, such as wake and drowsy.
Phases of alpha oscillations recorded by EEG were typically studied in the context of event or task related experiments, rarely during spontaneous alpha activity and in different brain states. During wake-to-drowsy transition they change unevenly, depending on the brain region. To explore their dynamics, we recorded ten adult healthy individuals in these two states. Alpha waves were treated as stable frequency and variable amplitude signals with one carrier frequency (CF). A method for calculating their CF phase shifts (CFPS) and CF phase potentials (CFPP) was developed and verified on surrogate signals as more accurate than phase shifts of Fourier components. Probability density estimate (PDE) of CFPS, CFPP and CF phase locking showed that frontal and fronto-temporal areas of the cortex underwent more extensive changes than posterior regions. The greatest differences were found between pairs of channels involving F7, F8, F3 and F4 (PDE of CFPS); F7, F8, T3 and T4 (CFPP); F7, F8, F3, F4, C3, C4 and T3 (decrease in CF phase locking). A topographic distribution of channels with above the average phase locking in the wake state revealed two separate regions occupying anterior and posterior brain areas (with intra regional and inter hemispheric connections). These regions merged and became mutually phase locked longitudinally in the drowsy state. Changes occurring primarily in the frontal and fronto-temporal regions correlated with an early decrease of alertness. Areas of increased phase locking might be correlated with topography of synchronous neuronal assemblies conceptualized within neural correlates of consciousness.
Neuromagnetic oscillations and hemodynamic correlates of P50 suppression in schizophrenia
2011, Psychiatry Research - Neuroimaging
Citation Excerpt :
Traditionally, alpha oscillations are attributed to the central-visual system since in the ‘idle state’ at closed eyes EEG and MEG alpha oscillations emerge over occipital scalp areas. More recently, however, a very high synchronization between left and right temporal alpha activity has been reported (Cover et al., 2004). Besides occipital structures, these findings suggest a contribution of temporal generators, localized, e.g. within the auditory cortices, to the generation of alpha activity.
Behavioral and electrophysiological data indicate compromised stimulus suppression in schizophrenia. The physiological basis of this effect and its contributions to the etiology of the disease are poorly understood. We examined neural and metabolic measures of P50 suppression in 12 patients with schizophrenia and controls. First, whole-head magnetoencephalography (MEG) assessed amplitudes of left- and right-hemispheric evoked responses and induced oscillations. Secondly, functional magnetic resonance imaging (fMRI) measured the hemodynamic responses to pairs of beeps with a short interval (500 ms) as compared with those with a long interval (1500 ms). The suppression of alpha power (8–13 Hz) time-locked to the stimuli was negatively correlated with the suppression of evoked components and the hemodynamic measures. Remarkably, the suppression of alpha power was reduced in the patients already prior to stimulus onset. Conceivably, alpha oscillations play a central role in stimulus adaptation of neuronal networks and reflect an active mechanism for sensory suppression. The reduced stimulus suppression in schizophrenia seems to be in part due to impaired generation of alpha oscillations in the auditory cortex, resulting in higher metabolic demand as detected by fMRI. Delayed recovery of alpha rhythm may reflect an impaired gating function and contribute to sensory and cognitive deficits in schizophrenia.
Disturbed functional connectivity in brain tumour patients: Evaluation by graph analysis of synchronization matrices
2006, Clinical Neurophysiology
Cerebral functions are based on the functional interactions between multiple distinct specialized regions of the brain. Functional interactions require anatomical connections as well as the synchronization of brain oscillations. The present work aims at evaluating the impact of brain tumours on spatial patterns of functional connectivity of the brain measured at rest by MEG.
We analyzed the statistical dependency (by computing the synchronization likelihood (SL, a measure of generalized synchronization)) between MEG signals at rest, in 17 patients with a brain tumour and in 15 healthy controls. Following an approach that derives from graph theory, we also analyzed the architectural properties of the networks by computing two parameters from the SL matrix, the cluster coefficient C and the characteristic path length L.
Alterations in synchronization levels were found in the patients and were not focal but involved intra-hemispheric connectivity. Effects were different considering the frequencies sub-bands, predominating in a decrease in high frequencies bands for long-distance connections and an increase in slower bands for local connectivity. In addition, graph analysis reveals changes in the normal “small-world” network architecture in addition to changes in synchronization levels with some differences according to the studied frequency sub-bands.
Brain tumours alter the functional connectivity and the “network” architecture of the brain. These alterations are not focal and effects are different considering the frequencies sub-bands.
These neurophysiological changes may contribute to the cognitive alterations observed in patients with brain tumours.

View all citing articles on Scopus

View full text

Detection of very high correlation in the alpha band between temporal regions of the human brain using MEG

Abstract

Introduction

Section snippets

Methods

Results

Discussion

Conclusions

Acknowledgements

Neurobiol. Aging

Neurosci. Lett.

Clin. Neurol.

Electroencephalogr. Clin. Neurophysiol.

Schizophr. Res.

Practical Statistics for Medical Research

Visual stimulus-dependent changes in interhemispheric EEG coherence in humans

J. Neurophysiol.