EEG Default Mode Network in the Human Brain

www.elsevier.com/locate/ynimg

EEG default mode network in the human brain:Spectral regional field powers

Andrew C.N. Chen,⁎ Weijia Feng, Huixuan Zhao, Yanling Yin, and Peipei Wang

Center for Higher Brain Functions, Capital Medical University, Beijing 100069, China

Received 17 September 2007; revised 12 December 2007; accepted 19 December 2007Available online 15 January 2008

Eyes-closed (EC) and eyes-open (EO) are essential behaviors in mam-malians, including man. At resting EC-EO state, brain activity in thedefault mode devoid of task-demand has recently been established infMRI. However, the corresponding comprehensive electrophysiologicalconditions are little known even though EEG has been recorded inhumans for nearly 80 years. In this study, we examined the spatialcharacteristics of spectral distribution in EEG field powers, i.e., sittingquietly with an EC and EO resting state of 3 min each, measured withhigh-density 128-ch EEG recording and FFT signal analyses in 15right-handed healthy college females. Region of interest was set at athreshold at 90% of the spectral effective value to delimit the dominantspatial field power of effective energy in brain activity. Low-frequencydelta (0.5–3.5 Hz) EEG field power was distributed at the prefrontalarea with great expansion of spatial field and enhancement of fieldpower (t=–2.72, pb0.02) from the EC to the EO state. Theta (4–7 Hz)EEG field power was distributed over the fronto-central area andleaned forward from EC to the EO state but with drastic reduction infield power (t=4.04, pb0.01). The middle-frequency alpha-1 (7.5–9.5 Hz) and alpha-2 (10–12 Hz) EEG powers exhibited bilateraldistribution over the posterior areas with an anterior field in loweralpha-1. Both showed significantly reduction of field powers (respec-tively, W=120, pb0.001 for alpha-1; t=4.12, pb0.001 for alpha-2)from EC to the EO state. Beta-1 (13–23 Hz) exhibited a similar spatialregion over the posterior area as in alpha-2 and showed reduction offield power (t=4.42, pb0.001) from EC to the EO state. In contrast,high-frequency beta-2 and gamma band exhibited similar, mainlyprefrontal distribution in field power, and exhibited no change from ECto the EO state. Corresponding correlation analyses indicatedsignificant group association between EC and EO only in the fieldpowers of delta (r=0.95, pb0.001) and theta (r=0.77, pb0.001) band.In addition, the great inter-individual variability (90 folds in alpha-1,62 folds in alpha-2) in regional field power was largely observed in theEC state (10 folds) than the EO state in subjects. To summarize, ourstudy depicts a network of spectral EEG activities simultaneouslyoperative at well defined regional fields in the EC state, varyingspecifically between EC and EO states. In contrast to transient EEGspectral rhythmic dynamics, current study of long-lasting (e.g. 3 min)

⁎ Corresponding author.E-mail address: [email protected] (A.C.N. Chen).Available online on ScienceDirect (www.sciencedirect.com).

1053-8119/$ - see front matter © 2008 Published by Elsevier Inc.doi:10.1016/j.neuroimage.2007.12.064

spectral field powers can characterize state features in EEG. The EEGdefault mode network (EEG-DMN) of spectral field powers at rest inthe respective EC or EO state is valued to serve as the basalelectrophysiological condition in human brain. In health, this EEG-DMN is deemed essential for evaluation of brain functions without taskdemands for gender difference, developmental change in age span, andbrain response to task activation. It is expected to define braindysfunction in disease at resting state and with consequences forsensory, affective and cognitive alteration in the human brain.© 2008 Published by Elsevier Inc.

Keywords: EEG field power; 90% effective power; Regional distribution;Default mode network; EC-EO state; Individual differences

Introduction

EEG and Brain Function/Dysfunction

EEG reflects vital brain activities, in fetal (Preissl et al., 2004)and from neonate (Vanhatalo and Kaila, 2004) to aging (Cumminsand Finnigan, 2007) in health and in disease. The function of EEGin the sleep/awake state (Chornet-Lurbe et al., 2002), perception(e.g. Basar et al., 2006; Andrade and Bhattacharya, 2003), attention(Vuilleumier and Driver, 2007), memory (Klimesch et al., 2006a;Wilding and Herron, 2006), emotion (Cacioppo, 2004), thinking(Srinivasan, 2007), as well as dysfunction in epilepsy (Benbadis,2006), autism (Kagan-Kushnir et al., 2005), dyslexia (Arns et al.,2007), AD/HD (Barry et al., 2007; Hobbs et al., 2007), anxiety(Smit et al., 2007), depression (Hunter et al., 2007), MCI (van derHiele et al., 2007; Buchem et al., 2007), dementia (Babiloni et al.,2006; Rossini et al., 2006), schizophrenia (Prichep, 2007), coma(Husain, 2006), vegetative state (Guerit, 2005), and even death(Sediri et al., 2007) is largely documented. In fact, EEG dynamics isclosely impacted in man's whole life. In additional to genetic factors(Begleiter and Porjesz, 2006), one of the main effects on the EEG isthe state of eyes-closed (EC) and eyes-open (EO) differentiation atrest. In the course of one's life, we often consciously orunconsciously lapse into an eyes-closed state, sometimes transiently

562 A.C.N. Chen et al. / NeuroImage 41 (2008) 561–574

within a second, briefly in minutes or even in fraction of hour notcounting napping-sleep. Yet the majority of time in the remaininghours we live in an eyes-open state. In state of eyes-open, the brain isbombarded with tremendous bits of visual stimuli. At the restingeyes-closed state, it was argued recently that the brain is in a “defaultmode” when devoid of any external task (Raichle et al., 2001).

“The Default Mode Brain State” defines a baseline state in thehuman brain lying quietly with eyes-closed (Raichle et al., 2001,Raichle and Synder, 2007) and is of great interest in that elementfor the default-mode network (DMN) or connectivity is readilymeasured from PET (Raichle et al., 2001; Geday et al., 2007) andfMRI (e.g. Damoiseaux et al., 2006). Measurement in health can berelated to: (a) stable oxygen extraction function of organized restingbrain function and suspended upon specific goal-directed behaviors(Raichle et al., 2001); (b) both medial prefrontal cortex and anteriorcingular cortex are embedded in an organized network for initiationof the functional “self” (Gusnard et al., 2001); (c) dorsal mPFCincreased in internal feeling judgment and ventral mPFC reduced inexternal cognitive judgment (Gusnard et al., 2001); (d) networkconnectivity of ventral anterior cortex and post cingulate cortexenhanced during resting state (Greicius et al., 2003), or the ventralmedial prefrontal cortex and posterior cingulate cortex exhibitinglargest activation at rest and deactivation upon tasks (Greicius andMenon, 2004); (e) slow oscillation between internal introspectionand external extrospection of brain state at rest (Fransson, 2005;Bellec et al., 2006), or slow respiratory activity (Birn et al., 2006);(f) intrinsic visual, auditory, somatosensory and olfactory sensorycortical activation in silence at rest (Hunter et al., 2006;Wilson et al.,2007); (g) attention and memory processing (Fransson, 2006; Es-posito et al., 2006); (h) age-related DMN (Grady et al., 2006;Damoiseaux et al., in press); (i) essential patterns in the DMN(Damoiseaux et al., 2006); and (j) light-sleep, even in the absence ofconscious awareness during light sleep (Horovitz et al., 2007).However, a different opinion on the nature of fMRI-DFN is alsorecently raised (Fransson, 2006).

Subsequently, the fMRI-DMN alternations in brain malfunctionsare now demonstrated in fragile X syndrome (Menon et al., 2004),attention disorder (Helps et al., in press; Castellanos et al., 2008),epilepsy (Laufs et al., 2007), Tourette's syndrome (Marsh et al.,2007), anxiety disorder (Zhao et al., 2007), bipolar disorder (Cal-houn et al., in press), major depression (Greicius et al., 2007),schizophrenia (Harrison et al., 2007; Garrity et al., 2007; Calhounet al., in press; Zhao et al., 2007), mild cognitive impairment (Rom-bouts et al., 2005) and Alzheimer's disease (Wang et al., 2006;Rombouts et al., in press; Sorg et al., 2007). In neuroimaging,most ofthe study conditions comparing the fMRI-DFN are relied on the ECbut little work on EO state.

Issues

When one closes his/her eyes, one never ceases feeling andthinking, unless under adequate anesthesia, in a dysfunctional modeof coma, vegetative state or death. From the literature on the defaultmode of brain function in fMRI, the distinction of the eyes-closedcondition is well studied but the eyes-open resting condition hasbeen infrequently made. In contrast since Hans Berger, our know-ledge of EC and EO in nearly 80 years of studying EEG is limited toa few transient classical “alpha block” (Pollen and Trachtenberg,1972) or contemporary “alpha desynchronization” (Pfurtscheller etal., 1996; Klimesch et al., 2007; Neuper et al., 2006) studies. EEGhas largely been dealt with as a set of isolated frequency fluctuations

in the temporal waveform domain, much less up to now as a set ofspectral field energy in the spatial domain. Awealth of information isperhaps under-exploited regarding the spectral energy changes atregional brain areas, other than the alpha-block studies, largely at thevisual field of the occipital brain upon EC to EO state.

Study aims

Taking into consideration the aforementioned, this study wasaimed at examining (a) the full spatial distribution of the spectralfield powers at the basic EC state, (b) how these regional spectralfield powers varies from the EC to EO state, (c) the individualcharacteristics of regional spectral field powers in both states, and(d) if some associated networks of regional spectral field powersexist in either EC or the EO state.

Materials and methods

Subjects

Fifteen healthy female volunteers (age: 20–26 years old) fromthe University community participated in the study. Subjects wereexcluded from the study if they had any physical/psychologicalproblem or were using medication. Written informed consent wasobtained from each subject in accordance with the Helsinki Decla-ration, and the study was approved by the local ethical committee.

Experimental design

Each subject was asked to sit in an armchair, throughout theduration of the experiment, in a quiet room (temperature: 22–24 °C)at shaded daylight. Each subject was asked to keep the eyes-closedfor 3 min and eyes-open for another 3 min, in that order, while EEGwas recorded. Each subject was not given any instruction but askedto stay fully relaxed.

EEG recordings and pre-treatment of EEG

A high-density 128-ch EEG recording, using Ag/AgCl electrodesand two channels of electro-oculogram (EOG) were continuouslygathered for 3 min during the EC and EO states, respectively with theA.N.T. EEG System (A.N.T. Enschede, Netherlands). The electrodeswere mounted according to the 10/5 montage system (Oostenveldand Praamstra, 2001). The 3D 128-ch montage and nomenclature isdepicted in Fig. 1.

Bilateral EOG was recorded from the horizontal and vertical sitesto monitor blinking or eye movements. The portion of EOG con-tamination of each scalp trace was removed in the off-line analysis.All the EEG channels were recorded in an averaged reference butwere offline bilaterally re-referenced (averaged A1+A2). EEG datawere sampled at 512 Hz and electrode impedance was kept lowerthan 5 kΩ. EEG data were filtered with 0.5–100 Hz bandpass filteroff-line andwere subjected to epoching (2 s each), linear-detrend, andartefact rejection pre-processing. The artifact rejection methodsconsisted of exclusion in epoch with large amplitude (over ±80 μV),DC bias, blinks, and slow eye movement coincident with EOG.

“Bad electrodes” were replaced with the extrapolated virtuevalues from the neighboring electrodes. After rejection of EOGcontamination and non-specific artifacts, each set of EEG data (2-sepoch) was subjected to Fast-Fourier Transform (FFT) analysis toobtain the absolute EEG band power (μv2) at each electrode in the

Fig. 1. EEG recording electrode sites and nomenclatures. Superior-anterior view (left panel) and superior-posterior view (right panel) perspectives of the 128-chEEG electrode nomenclature in 10/5 montage.

563A.C.N. Chen et al. / NeuroImage 41 (2008) 561–574

following 7 bands: Delta (0.5–3.5 Hz), theta (4–7 Hz), Alpha-1(7.5–9.5 Hz), Alpha-2 (10–12 Hz), Beta-1 (13–23 Hz), Beta-2 (24–34 Hz) and Gamma (35–45 Hz). These broad bands were definedby the conventional IFCN guideline (Nuwer et al., 1999) but notaccording to individuals' spectral characteristics (cf. Doppelmayret al., 1998; Klimsch, 1999).

Fig. 2. Topographic spectral mapping of low-frequency delta and theta field powersand the theta field power at the fronto-central area. Increase and expansion of delshown from EC to EO state.

For each EC and EO state, each 3-min period EEGwas analyzed in2 sec epochs, resulting in 90 epochs. Out of them, around 70–80 validepochs at the EC state and 60–80 epochs at the EO state across the 15subjects, on average, were subjected to further analyses. For each ECand EO state the valid epochs were averaged after procedure of fastFourier Transform (FFT) performed by ASA 3.0 software (A.N.T.

in EC vs. EO states. Delta field power was distributed over the prefrontal areata field power, but marked reduction of theta field power were respectively

Fig. 3. Left panels: t-tests of EC and EO field power showed significant increase of delta, but radical reduction of theta field power from EC to EO. Right panels:Correlation analyses indicated significant associations of the values between EC and EO states for delta and theta. Greater inter-individual variability was notedin EC than in EO state for both field powers. Subjects were ordinally listed from the EC state.

564 A.C.N. Chen et al. / NeuroImage 41 (2008) 561–574

Enschede, Netherlands). The grand averages of FFT maps of differentstages in topographic maps (256 hues) of the mean amplitude of thesurface EEG power were calculated on a 3D “quasi-realistic” corticalmodel by a spline interpolating function (Tandonnet et al., 2005;Babiloni et al., 2006; Perrin et al., 1987). For each of the 7 EEG bands,a scalp field power spectralmapwas interpolated from the 128-ch EEGand the site/electrode of maximal power was isolated with its valuecalculated. Through automatic scaling, the range of maximum andminimum values in EEG spectral power was established.

Region of Interest (ROI)

The total field power (sum of the electrode spectral power) ineach spectral EEG bands within a region of interest (ROI) was

Table 1Magnitudes of the individual differences at the EC and EO state (field powerin μV2)

EC max/min=ratio EO max/min=ratio EC/EO

Delta 845/151=5.91 582/90=10.86 5.91/10.86=0.54Theta 29.3/6.7=4.40 12.3/5.2=2.00 4.4/2.0=2.20Alpha-1 230/2.3=90.60 20.5/2.1=9.54 90.6/9.54=10.34Alpha-2 62.7/1.0=62.70 20.5/2.1=9.54 62.7/8.12=7.10Beta-1 20.3/4.1=5.00 11.6/3.7=3.15 5.0/3.15=1.59Beta-2 22.9/1.9=12.16 5.6/2.5=7.62 12.16/7.62=1.59Gamma 33.7/0.9=36.18 29.4/1.3=21.42 36.18/21.42=1.69

calculated, respectively in each of the 7 broad bands at the EC andEO state, as dependent measure. The ROI was spatially delimited bythe area exhibiting the band power value in electrodes exceeding acritical value that equaled to 90% of the effective value (90% EV),termed effective energy. As in the mapping of brain activity by PET-fMRI, EEG activity in this study is based on the threshold criterion.From the threshold criterion, the effective size of spatial area andmagnitude of activity can be defined and measured. The measuredarea in this study was the “region of interest”, while the measuredmagnitude was the “dominant EEG power”. From each spectralEEG band, separate spectral EEG power was calculated based on thesum of the measured power of all electrodes in the ROI. In the study,the 90% EV (effective value) was the sum of all EEG powers fromthe electrodes in the ROI, above 90% of the effective value. The EVwas defined by the 5th upper percentile ranking value of the 128electrodes, from the zero to maximum value of all the 128 EEG sites.Often, the maximal value has been used in the calibration range.However, we choose to use the 5th percentile ranking value, insteadof the maximum value (the EEG site of the highest rank), for the sakeof preventing the outlier electrode value, which might occur. In thismanner, the use of 5th percentile ranking value for EV was moreconservative and would effectively exclude the probable outliermaximum electrode (such as from a bad electrode site with aspurious effect on the EEG). Thus, we defined the ROI field poweras the dominant spectral EEG field power. In this study, the 90% EVof spectral EEG well delimited the dominant regional field powerrespectively for delta, theta, alpha-1, alpha-2, beta-1, beta-2 and

Fig. 4. Topographic spectral mapping of the middle-frequency alpha-1 and alpha-2 field powers. The alpha-1 field power was distributed mainly at the posteriorarea with an anterior extension, while the upper alpha-2 field power was restricted at the posterior area. Both showed marked reduction in the field power fromEC to EO state.

565A.C.N. Chen et al. / NeuroImage 41 (2008) 561–574

gamma broad bands in one of the two states. A grand average of theresting EC and EO states was computed then the statistical effects ofthe state change were conducted on the effective spectral field powerat the regional areas in each of the 7 bands. For statistical comparisonbetween states or across conditions, the ROI territory that containsthe largest electrode array was selected as a template. As a generalprinciple, the sum of the electrode values within the ROI in eachsubject was calculated for statistical analysis. In this manner, thedependent variable of interest was the field power of array of elec-trodes, but not a power value of a single electrode.

In this study, the spatial template of delta field power was appliedby the EO state since it exhibited higher 90% EV than that of the ECstate, while the rest of spectral powers were applied by the EC stateas these values were higher at the EC than EO state. The dominant90% EVof the spectral field power in an EEG delimited a spectralregional area with 10-spacings set in each spectral field power range:below 10% in blue and above 90% in red between the range of zeroand the top 90% EV (see the calibration bars in Figs. 2, 4, 6).

Statistical analysis

Statistical comparisons were performed on the scalp EEG fieldpower by paired t-test for comparison of EC and EO states. Weelected to use the raw values, i.e. without logarithmic transforma-tion of the EEG values (e.g. van Albada and Robinson, 2007), torely on the original recording for straightforward interpretation ofEEG values. In our past experience, log transformation of EEG

values cannot completely eliminate the problem of statisticalnormality in the distribution of EEG values. Thus, if the data setswere confirmed to normality distribution, dependent t-test of meansand the Pearsons correlation test on the association of spectral fieldpowers between these two states were used. If not, the Wilcoxon testfor nonparametric comparison of ranks and Spearman correlation ofranks were employed to evaluate the probability significance. In alltests, pb0.05 was accepted as significant. The statistical programSigmaStat 3.04v (SPSS Inc, Chicago, USA) was used for statisticalprocedures.

Results

Low-frequency EEG field power

Spatial MappingIn Fig. 1, the dominant delta field power was found to be at the

prefrontal area for both the EC and the EO states. In comparison,theta power was distributed at the fronto-central area, centered overthe vertex for both (EC and EO) states.

Effect of the change of statesFrom the EC to the EO state, the expansion of the delta field

power in the left fronto-lateral region was noted, but reduction oftheta field power in amplitude and space at the fronto-central areawas clearly shown. These effects were verified and presented inFig. 3. The t-tests independently demonstrated a marked enhance-

566 A.C.N. Chen et al. / NeuroImage 41 (2008) 561–574

ment of the delta power (EC: mean 354.9 μV2, EO: mean 407.4 μV2;t=-2.72, pb0.2), while a great reduction in the theta power (EC:mean 14.7 μV2, EO: mean 9.1 μV2; t=4.04, pb0.001) was found.

Individual's profileThe individuals varied greatly in EEG powers, largely in the

resting EC than EO state. The right panels in Fig. 3 depictthe magnitude of the variability reached as high as 5.91 folds in thedelta power between the maximum and minimum values in thesubjects at the EC state, but only 0.54 fold at the EO state (seeFig. 3, left panel; Table 1). For the theta field power, the ratios were4.4 folds at the EC and 2.00 at the EO state (see Fig. 3, left panel;Table 1). All of the comparisons of the inter-individual variabilityare listed in the Table 1. Nevertheless, the measured field powersalso displayed a set of significant correlations between EC and EOstates in both delta and theta powers (r=0.95, pb0.001 andr=0.77, pb0.001, respectively).

Middle-frequency EEG field power

Spatial MappingIn Fig. 4, the dominant alpha-1 field power was found to be at the

posterior area along with a frontal field extension much eminently atthe EC than the EO state. In comparison, the upper alpha-2 field

Fig. 5. Left panels:W-test (Wilcoxon rank test) and t-test of EC and EO power compfield power from EC to EO. Right panels: Correlation analyses indicated non-signifalpha-2 field powers. However, marked greater inter-individual variability was shordinally listed from the EC state.

power was restricted at the posterior area only with a clear contour atthe EC state. At the EO state, both alpha-1 and alpha-2 field powerswere not discernible in the regional topography under the samescaling at the EC state.

Effect of the statesFrom the EC to the EO state, the magnitude and spatial extent of

the alpha-1 field power was greatly reduced (Fig. 5). The t-testsindependently indicated a marked reduction of the alpha field power(EC: mean 54.9, median 41.6 μV2, EO: mean 6.7, median 4.3 μV2;W=120, pb0.001), while a similar large reduction of the upperalpha-2 field power was also evident (EC: mean 23.6 μV2, EO: mean5.6 μV2; t=4.12,b001).

Individuals profileAgain, the great inter-individual variability in EEG powers of

alpha-1 and alpha-2 was observed to be limited to the resting ECstate unlike the EO state. From the Fig. 5 right panels, the magnitudeof the variability reached to peak values of 90.6 folds in the alpha-1field power between the maximum and minimum values in thesubjects at the EC state, but 9.54 folds the EO state. Similar largevalues, a 62.7 folds increase in EC state and a 8.12 folds in the EOstate for the upper alpha-2 field power were noted (see Table 1). Nocorrelation was shown in either alpha-1 or alpha-2 between the ECand the EO states.

arison showed, respectively, significant decrease of both alpha-1 and alpha-2icant (ns) association of the values between EC and EO states for alpha-1 andown respectively at EC than EO state for both field powers. Subjects were

Fig. 6. Topographic spectral mapping of high-frequency beta-1 distributed at posterior area, beta-2 and gamma field powers (both distributed at the midlinefrontal and prefrontal areas) in EC vs. EO state. Large topographic change in the beta-1 field power was discerned.

567A.C.N. Chen et al. / NeuroImage 41 (2008) 561–574

High-Frequency EEG Field Power

Spatial mappingIn Fig. 6, the beta-1 field powerwas found to be over the posterior

area as in the upper alpha-2 field power and exhibited reduction fromthe EC to the EO state. In comparison, the high-frequency beta-2 andgamma field powers were limited mainly at the prefrontal area.

Effect of the statesThe observed statistical significant difference in beta-1 field

power between the EC and EO states is shown in the top-left panel ofFig. 7. The t-test independently indicated a reduction of the beta-1field power (EC: mean 12.9 μV2, EO: mean 6.8 μV2; t=4.12, pb0.001), while no significant effects were noted in both the beta-2 andgamma field powers.

Profile of the individualsThe individual distribution profile of the beta-1 field power was

greatly dissimilar and distinguished the beta-1 from those of thebeta-2 and gamma field powers. All correlations between the EC andthe EO states in these high-frequency beta-1, beta-2 and gamma fieldpowers were proved not significant (right panels, Fig. 7).

Spectral co-habilitation and correlations among the Field Powersat each EC and EO State

At each of the EC or EO state, the dominant regional EEG fieldpowers are shown to be a distinct set of spectral activitiesdistributed in the head space. In Fig. 8, a composite model in eachEC and EO state was respectively made from two viewingperspectives, superior-anterior (sa) and superior-posterior (sp) to

Fig. 7. Left panels: t-test of EC and EO field power comparison showed a significantly decrease of beta-1, butW-tests (Wilcoxon rank test) of EC and EO powercomparisons showed no significant (ns) effects of beta-2 and gamma field powers from EC to EO state. Right panels: Correlation analyses indicated non-significant (ns) association of the values between EC and EO states for all the high-frequency spectral field powers. Great inter-individual variability in eachspectral field power was noted. Subjects were ordinally listed from the EC state.

568 A.C.N. Chen et al. / NeuroImage 41 (2008) 561–574

illustrate the coexistence of these spectral activities in space. ThisFig. 8 was computerized by a virtual brain model in respective theEC and the EO state of 25 young male subjects studied in Denmark(data not published before).

Again, the study data revealed a dramatic inter-individual vari-ability in human EEG magnitude. Table 1 lists the maximum andminimum values of regional 7 spectral field powers. In 15 subjects ahigh end of 96.0 folds of alpha-1 and low end 4.4 of theta in thegroup at EC state was revealed. The ratios of maximum/minimummanifested to a lesser differentiation at the EO state, ranging from0.86 of the delta field power to 8.12 folds of alpha-1 (drasticallydifferent from that in the EC state) in the EO state. The magnitude ofthe individual differences, comparing the EC and EO states is

presented in the rightist column of Table 1. The ratios in variabilitybetween the EC and EO states for alpha-1 and alpha-2 were 10.34and 7.10 folds respectively, demonstrating markedly greatervariability in the group at the EC than in the EO state for alpha-1and alpha-2 field powers.

Table 2a,b list the calculated cross-correlations among the 7spectral field powers, respectively in EC and EO states. At the ECstate, correlation analyses indicated that only the high frequency beta-2 and gamma powers at the prefrontal head region were correlated(r=0.92, pb0.001). No other spatial EEG band powers in the ECstate showed significant correlation (Table 2a). At the EO state, incontrast, the prefrontal delta field power exhibited significant correl-ation with the posterior beta-1 (r=0.86, pb0.001) and prefrontal

Fig. 8. Constellation of spectral field powers measured in a 3-min period depicts the regional distribution of the respective seven frequency band powers on thebrain topography, at EC vs. EO state, viewed from superior-anterior (s. a.) and superior-posterior (s. p.) perspectives. The resting EEG DFN was composed fromdata gathered in 25 healthy college young males from Aalborg, Denmark, while the author Chen was directing the Human Brain Mapping Laboratory. Data werenot published previously.

569A.C.N. Chen et al. / NeuroImage 41 (2008) 561–574

gamma field powers (r=0.554, p=0.036). In addition, the fronto-central theta field power was correlated with the anterior-posterioralpha-1 field power (r=0.52, p=0.046), while the prefrontal beta-2and gamma field powers were significantly correlated (Table 2b).

Discussion

EEG DMN

The main study findings indicate that the resting EEG in eyes-closed state devoid of stimulation or task is composed of a definedset of regional spectral activities in the classic 7 broad bands,termed EEG default mode network (EEG DMN). The distributionmaps of spatial regional field powers are consistent with a previousobservation using Danish subjects (Chen et al., 2006). This spatialdistribution of the field EEG powers has not fully been appreciated.After completing the current work, an important off-press publica-tion indicated that bold fMRI can be correlated with EEG powers insix widely distributed resting state networks, characterized by EEGfeatures involved in combination with different brain rhythms

(Mantini et al., 2007). In EEG studies, it is necessary to define notonly the spectral information but also the spatial characteristics ofthe EEG of subjective experience. Extending from current work in apreliminary report (Feng and Chen, submitted for publication), therelations of regional EEG field powers in listening to musiccompared to lecture (speech, 3 min eyes-closed) and variety ofemotional ratings are effectively revealed, as to validate the nature ofEEG DMN and brain functions.

Dominant EEG Spectral Field Power and Function

The spatial nature of the field EEG power may reflect someunderlying co-activity in structure-function relations. For example,the eminent prefrontal delta EEG field power (see Fig. 2) couldindicate some prefrontal metabolic/functional significance. Thoughit is tempted to categorize it as some residue of blinking/eyes-movement generated in the prefrontal subcortical site, two factsargue against it. The likelihood of a blinking artifact on the pre-frontal delta EEG was diminished since our artifact algorithm wasset at a threshold to eliminate all epochs exceeding 80 μV such as

Table 2aEC state: cross correlations of the regional spectral field powers (correlationvalues in bold, p-values in italics; significant values in shaded field)

able 2bO state: cross correlations of the regional spectral field powers (correlationalues in bold, p-values in italics; significant values in shaded field)

570 A.C.N. Chen et al. / NeuroImage 41 (2008) 561–574

the blinking on the EEG (150 μV). This argument was also nottenable as nearly no blinking episode was found in the EEG recordduring the EC state (only one blink across the 15 subjects). Thisconjecture may be viable with the EO state, which exhibited highfrequency of blinking (mean: 37.4/min). However, no correlationof the blinking rate and the prefrontal EEG field power could beestablished (r=0.344, p=0.21). Recently, it has been shown thatEEG is associated with cerebral metabolic rate, as it may explain65% of the variance in the delta power (Boord et al., 2007). Also,the delta (1.5–4 Hz) power correlates with brain atrophy in Al-zheimer's disease patients (Fernandez et al., 2005) as well as theamplitude of delta (2–4 Hz) sources with the prefrontal white matteracross the mild cognitive impairment and Alzheimer's patients (Ba-biloni et al., 2006). At resting state, a positive correlation of medialfrontal cortical metabolism with delta EEG power has beenestablished (Alper et al., 2006).

In the theta EEG power, 53% of the variance over age span isreported to be explained by cerebral metabolic rate (Boord et al.,2007). Our finding of the theta EEG field power distributed emi-nently in the fronto-central area (see Fig. 2) is consistent with theknown frontal midline theta EEG activity (Inanaga, 1998). It isgenerated in the medial prefrontal cortex and/or anterior cingulatecortex in monkey (Tsujimoto et al., 2006) and in man (Cahn andPolich, 2006). The regional preeminent theta EEG has been relatedto infant arousal (Grigg-Damberger et al., 2007), attention (Sausenget al., 2007), perception (Basar et al., 2006), memory (Onton et al.,2005), emotion (Guntekin and Basar, 2007), music (Feng and Chen,submitted for publication), and cognitive aging (Brickman et al.,2005).

In the middle-frequency alpha EEG, we observed a frontal ex-tension along with a more prominent posterior distribution in thelower alpha-1 field power (Fig. 4) either in congruent or at variancewith the resting anterior alpha asymmetry in emotion (Cacioppo,2004). The spectral field power in lower alpha-1 differentiated theupper alpha-2 field power with no anterior extension and only thesimilar posterior portion is of interest. The lower and upper alpha inEEG literature (Klimesch, 1997; Klimesch et al., 2006b) has beendescribed in function but little is known on the spatial differentiation.Recently, the neural source of alpha EEG is revealed in the brainregion of distinct cortical areas, such that the sleep spindle alpha-EEG is likely from thalamus, the mu-rhythm in motor function isfrom central sulcus, and visual alpha EEG in perceptual function isfrom occipital lobes (Goncalves et al., 2006).

Correlations of the averaged power of alpha EEG (8–12 Hz) andbold fMRI signals can be found in occipital, parietal and even frontallobes (Goncalves et al., 2006) but with wide individual variability atrest (de Munck et al., 2007). Nevertheless, the regional territory ofthe eminent alpha-1 and alpha-2 field powers reside more on theparietal area than occipital area (see Fig. 4). This observed alpha-1and alpha-2 field powers in territory may lead to reconsider theimportant aspect of frontal and parietal alpha EEG in source andfunction, extending from the conventional occipital alpha EEG.

In high frequency beta-1, beta-2 and gamma EEG field powers,the observed beta-1 field power in regional distribution, similar tothat of alpha-2 at the posterior scalp area (Fig. 6), suggests that beta-1as a spectral lingering of alpha-2 activity and belongs to the samefamily. It is proposed to call beta-2 as the high-frequency alpha-3, aplausibility that further remains to be proven. The great similarity indistribution at the bilateral temporal area and the prefrontal area in thebeta-2 and gamma EEG field powers (Fig. 6) and their strong corre-lation at both the resting EC and EO states (Table 2a,b) renders thesebeta-2 and gamma EEG in the same family. In the past, little spatialinformation of the beta and gamma EEG field in human brain hasbeen reported in literature. Due to their tiny EEG power, in the rangeless than 3 μVin amplitude, the distribution of these activities at somediscrete field opens to new challenge for understanding the structure-function relations in beta-gamma EEG, thought to reflect binding ofperception in brain function (Bertrand and Tallon-Baudry, 2000). Thetransient beta activity in motor function is associated with fMRI boldactivation of the sensorimotor cortex (Parkes et al., 2006), while thetransient gamma activity in visual perception with bold activation ofthe calcarine sulcus (Hoogenboom et al., 2006). Of special interest isthe site of eminent gamma field power at the dorso-lateral frontalareas in the left hemisphere, which is noted in the differentiation ofmusic from speech perception (Feng and Chen, submitted forpublication). We now are in the process to validate the cortical andsubcortical structures of the observed regional scalp EEG field pow-ers described in this study, by current source imaging of LORETA onspectral EEG field powers in Talairach standard brain modeling.

State changes

The spectral topographic distributions showing little changesattest the stability of the underlying structures associated with theEEG field powers. Instead, the major changes largely in reductionof the field power, except the increase of prefrontal delta, are the

TEv

571A.C.N. Chen et al. / NeuroImage 41 (2008) 561–574

main manifestation of the EC to EO state. Such results of strongmodulation at various spectral EEG field power greatly expand ourunderstanding of the limited classical “alpha-block” in EEG fromEC to EO (Pollen and Trachtenberg, 1972) or modern “alpha de-synchronization” (Rihs et al., 2007) from the state transaction. In fact,the observed changes in spectral EEG field powers at differentregional distribution attests to the complex and consorted EEGchanges from EC to EO at rest. Since no task or instruction was givento the subjects at the resting states, these changes may occur in-trinsically from the EC to the EO state both in terms of physiologicalvisual information processing and psychological terms of visualperception in seeing. This EC and EO differential states have recentlynot explicitly elucidated in the fMRI literature as the defaultmode of brain function, largely defined by the lack of explicit taskdemands on the brain in subjects in the EC state, lying at rest (Raichleet al., 2001). Several reports of the brain state in the default modefrom fMRI studies include “self-referential mental activity” (Gusnardet al., 2001), “exteroceptive and interoceptive deployment ofattention” (Nagai et al., 2004), “mind-readiness alert” (Fransson2005), and “inner-speech” (Hunter et al., 2006). Our observedspectral EEG field powers in the regional distribution between the ECand EO states, if is corresponding to the above hypothesis, remain tobe examined.

Inter-individual variability in EEG field powers and the inter-stateconsistency

Two aspects of the inter-subject variability deserve special at-tention. Firstly, the large differences in alpha field powers noted, asmuch as 90 folds in alpha-1 and 60 folds in alpha-2 in the group of 15subjects, are in magnitudes higher than expected. Secondly, theseinter-individual difference ratios are 10 folds higher at the resting ECthan EO state. Though there is some knowledge that a 10–15% of thenormal healthy subjects have little or nearly no alpha activity, themagnitude of such differences comes as a surprise. These resultswarrant consideration as a practical issue and call for re-conside-ration in statistical parametric comparisons of means. It seems thatnon-parametric Wilcox-test of ranks can be justified and similarlynon-parametric Spearman correlation test is appropriate for evalu-ating statistical features. In the current study, little spectral EEG fieldpowers (except delta and theta) showing significant correlationsmayindicate two divergent networks between the EC and EO state. Therobust correlations of high-frequency beta-2 and gamma fieldpowers in both states do suggest some invariant /function of theseactivities at the prefrontal area, regardless of the state.

Implication

Methodologically, our approach in the delimiting the dominantregional distribution of spectral EEG field power is an importantextension of the conventional quantitative EEG, qEEG, mapping(Duffy et al., 1994). The emphasis is on the field potential, in adiscrete area, over conventional use of selective electrode spectralpower in an interpolated map. Using high-density 128-ch montage,the spectral dominant EEG field power can be extracted in regionof interest (ROI) for robust comparisons and is based on the“effective-energy” concept. In current study, we selected the 90%EV (see Methods) as a sensible threshold to delimit a referencecluster of field power for comparison of states. This work furtherentrusts the steady-state comparison of experience episode in thebrain, expressed in EEG state. In general, the EEG DMN is aimed

at defining different EEG states. Our study on listening to ‘happy'music /'sad' music compared to that of lecture speech bears it out(Feng and Chen, submitted for publication).

Apparently, many variables used for processing the EEG (elec-trode density, recoding, reference, time-episode, epoch, FFT, fre-quency bands, effective-energy threshold) require standardization tofacilitate useful comparison across conditions and among labo-ratories. Additionally, the subject variables include the choice of theresting basal state, EC or EO, in the study. The observed 10 folds ofalpha variability in EC over EO shall be taken into account. Thereason for this in neurophysiology remains totally unknown. Giventhe high magnitude of spectral field powers in the EC state, itprovides an advantage in invoking states with effective EEG re-duction. Given the low magnitude of field powers in the EO state, itis an advantage to invoke states which may enhance field powers.The surprising markedly inter-individual variability, as much as 90folds in alpha-1 and 62 folds in alpha-2 field powers (see Table 1)shall not deter scientific investigation in EEG research. By consi-dering the individual normalization, relative EEG transformationand use of non-parametric statistics with adequate sample size,sensible and viable EEG/brain functions are deemed fully worthy forfuture investigation. In the field of EEG research, effort should bedirected toward the innovative method for standardized automaticquantification of spectral EEG powers and differentiation of EEGstates (Chen et al., 2005).

To this end, the current study strategy is ready and opens up anew avenue for the examination of the EEG DMN in gender dif-ferences, on the development of age-span in normal brain function aswell as their changes upon experimental stimulation or task demand.Likewise, the EEG DMN is valuable in the examination of alter-ations in patients with diseases and the dysfunctional responses totheir environment (cf. Laufs et al., 2007; Babiloni et al., 2006).

Conclusion

In a group of 15 healthy females, the EEG default mode network(DFN) at rest with eyes-closed of the dominant spectral field powersentails a constellation of large low-frequency delta activity at theprefrontal area, much smaller theta activity at fronto-central area,alpha-1 activity at the anterior-posterior area, alpha-2 and beta-1 at theposterior area, and high-frequency beta-2 and gamma activities at theprefrontal area. Compared to the eyes-open resting state, the deltafield power is enhanced at the prefrontal area while the theta, alpha-1,alpha-2 and beta-1 powers are reduced in the respective areas. Greaterinter-individual variability in field power can be seen in the eyes-closedthan the eyes-open state. In the eyes-closed state, the prefrontal deltafield power is correlated with both beta-1 and gamma field powers,while these two high-frequency field powers are mutually correlated atthe same area. The EEG DMN of the spectral field power in the eyes-closed or in the eyes-open state at rest, devoid of task demands, isreadily served as basal brain condition for the comparison of activationchallenge or dysfunctional alteration in diseases of the brain.

Acknowledgments

This study was supported by several grants respectively from theChinese National Science Foundation for a project on “Brain andAnesthesia” NSF-30770691, Beijing Municipal Government forAdvancement of Sciences, and Capital Medical University forInnovation Awards.

572 A.C.N. Chen et al. / NeuroImage 41 (2008) 561–574

References

Alper, K.R., John, E.R., Brodie, J., Günther, W., Daruwala, R., Prichep, L.S.,2006. Correlation of PET and qEEG in normal subjects. Psychiatry Res.146, 271–282.

Andrade, P.E., Bhattacharya, J., 2003. Brain tuned to music. J. R. Soc. Med.96, 284–287.

Arns, M., Peters, S., Breteler, R., Verhoeven, L., 2007. Different brainactivation patterns in dyslexic children: evidence from EEG power andcoherence patterns for the double-deficit theory of dyslexia. J. Integr.Neurosci. 6, 175–190.

Babiloni, C., Binetti, G., Cassetta, E., Dal Forno, G., Del Percio, C., Ferreri,F., Ferri, R., Frisoni, G., Hirata, K., Lanuzza, B., Miniussi, C., Moretti,D.V., Nobili, F., Rodriguez, G., Romani, G.L., Salinari, S., Rossini, P.M.,2006. Sources of cortical rhythms change as a function of cognitiveimpairment in pathological aging: a multicenter study. Clin. Neurophy-siol. 117, 252–268.

Barry, R.J., Clarke, A.R., McCarthy, R., Selikowitz, M., 2007. EEG cohe-rence in children with attention-deficit/hyperactivity disorder and co-morbid oppositional defiant disorder. Clin. Neurophysiol. 118, 356–362.

Basar, E., Guntekin, B., Oniz, A., 2006. Principles of oscillatory braindynamics and a treatise of recognition of faces and facial expressions.Prog. Brain Res. 159, 43–62.

Begleiter, H., Porjesz, B., 2006. Genetics of human brain oscillations. Int. J.Psychophysiol. 60, 162–171.

Bellec, P., Perlbarg, V., Jbabdi, S., Pélégrini-Issac, M., Anton, J.L., Doyon,J., Benali, H., 2006. Identification of large-scale networks in the brainusing fMRI. NeuroImage 29, 1231–1243.

Benbadis, S.R., 2006. The EEG in nonepileptic seizures. J. Clin. Neuro-physiol. 23, 340–352.

Bertrand, O., Tallon-Baudry, C., 2000. Oscillatory gamma activity in hum-ans: a possible role for object representation. Int. J. Psychophysiol. 38,211–223.

Birn, R.M., Diamond, J.B., Smith, M.A., Bandettini, P.A., 2006. Separatingrespiratory-variation-related fluctuations from neuronal-activity-relatedfluctuations in fMRI. NeuroImage 31, 1536–1548.

Boord, P.R., Rennie, C.J., Williams, L.M., 2007. Integrating “brain” and“body”measures: correlations between EEG and metabolic changes overthe human lifespan. J. Integr. Neurosci 6, 205–218.

Brickman, A.M., Paul, R.H., Cohen, R.A., Williams, L.M., MacGregor,K.L., Jefferson, A.L., Tate, D.F., Gunstad, J., Gordon, E., 2005. Categoryand letter verbal fluency across the adult lifespan: relationship to EEGtheta power. Arch. Clin. Neuropsychol. 20, 561–573.

Buchem, M.A., van Dijk, J.G., Middelkoop, H.A., 2007. EEG correlates inthe spectrum of cognitive decline. Clin. Neurophysiol. 118, 1931–1939.

Cacioppo, J.T., 2004. Feelings and emotions: roles for electrophysiologicalmarkers. Biol. Psychol. 67, 235–243.

Cahn, B.R., Polich, J., 2006. Meditation states and traits: EEG, ERP, andneuroimaging studies. Psychol. Bull. 132, 180–211.

Calhoun, V.D., Maciejewski, P.K., Pearlson, G.D., Kiehl, K.A., in press.Temporal lobe and “default” hemodynamic brain modes discriminatebetween schizophrenia and bipolar disorder. Hum. Brain Mapp.

Castellanos, F.X., Margulies, D.S., Kelly, C., Uddin, L.Q., Ghaffari, M.,Kirsch, A., Shaw, D., Shehzad, Z., Di Martino, A., Biswal, B., Sonuga-Barke, E.J., Rotrosen, J., Adler, L.A., Milham, M.P., 2008. Cingulate-Precuneus Interactions: A New Locus of Dysfunction in AdultAttention-Deficit/Hyperactivity Disorder. Biol. Psychiatry 63, 332–337.

Chen,A.C.,Wang, L., Arendt-Nielsen, L., Krupimski, R., 2005. Standardisedautomatic quantification method in EEG-ERP mapping. Organ. Hum.Brain Mapp. 271.

Chen, A.C., Liu, F.J., Wang, L., Arendt-Nielsen, L., 2006. Mode and site ofacupuncture modulation in the human brain: 3D (124-ch) EEG powerspectrum mapping and source imaging. NeuroImage 29, 1080–1091.

Chornet-Lurbe, A., Oteo, J.A., Ros, J., 2002. Characteristic times in sleep-waking electroencephalograms. Rev. Neurol. 35, 415–419.

Cummins, T.D., Finnigan, S., Ros, J., 2007. Theta power is reduced inhealthy cognitive aging. Int. J. Psychophysiol. 66, 10–17.

Damoiseaux, J.S., Rombouts, S.A., Barkhof, F., Scheltens, P., Stam, C.J., Smith,S.M., Beckmann, C.F., 2006. Consistent resting-state networks acrosshealthy subjects. Proc. Natl. Acad. Sci. U. S. A. 103, 13848–13853.

Damoiseaux, J.S., Beckmann, C.F., Arigita, E.J., Barkhof, F., Scheltens,P., Stam, C.J., Smith, S.M., Rombouts, S.A. in press. Reduced resting-state brain activity in the “default network” in normal aging. Cereb.Cortex 2.

de Munck, J.C., Goncalves, S.I., Huijboom, L., Kuijer, J.P., Pouwels, P.J.,Heethaar, R.M., Lopes da Silva, F.H., 2007. The hemodynamic responseof the alpha rhythm: an EEG/fMRI study. NeuroImage 35, 1142–1151.

Doppelmayr, M.M., Klimesch, W., Pachinger, T., Ripper, B., 1998. Thefunctional significance of absolute power with respect to event-relateddesynchronization. Brain Topogr. 11, 133–140.

Duffy, F.H., Hughes, J.R., Miranda, F., Bernad, P., Cook, P., 1994. Status ofquantitative EEG (QEEG) in clinical practice. Clin. Electroencephalogr.25, 6–22.

Esposito, F., Bertolino, A., Scarabino, T., Latorre, V., Blasi, G., Popolizio,T., Tedeschi, G., Cirillo, S., Goebel, R., Di Salle, F., 2006. Independentcomponent model of the default-mode brain function: Assessing theimpact of active thinking. Brain Res. Bull. 70, 263–269.

Feng, W., Chen, A.C., submitted for publication. Spectral EEG mapping ofhuman emotion in processing of fast (Superman) vs. slow (Schindler'sList) music of movie episodes compared to a lecture speech.

Fernandez, A., Rodriguez-Palancas, A., Lopez-Ibor, M., Zuluaga, P.,Turrero, A., Maestu, F., Amo, C., Lopez-Ibor Jr., J.J., Ortiz, T., 2005.Increased occipital delta dipole density in major depressive disorderdetermined by magnetoencephalography. J. Psychiatry Neurosci. 30,17–23.

Fransson, P., 2005. Spontaneous low-frequency BOLD signal fluctuations:an fMRI investigation of the resting-state default mode of brain functionhypothesis. Hum. Brain Mapp. 26, 15–29.

Fransson, P., 2006. How default is the default mode of brain function?Further evidence from intrinsic BOLD signal fluctuations. Neuropsy-chologia 44, 2836–2845.

Garrity, A.G., Pearlson, G.D., McKiernan, K., Lloyd, D., Kiehl, K.A.,Calhoun, V.D., 2007. Aberrant “default mode” functional connectivity inschizophrenia. Am. J. Psychiatry 164, 450–457.

Geday, J., Kupers, R., Gjedde, A., 2007. As time goes by: temporalconstraints on emotional activation of inferior medial prefrontal cortex.Cereb. Cortex. 17, 2753–2759.

Goncalves, S.I., de Munck, J.C., Pouwels, P.J., Schoonhoven, R., Kuijer,J.P., Maurits, N.M., Hoogduin, J.M., Van Someren, E.J., Heethaar, R.M.,Lopes da Silva, F.H., 2006. Correlating the alpha rhythm to BOLD usingsimultaneous EEG/fMRI: inter-subject variability. NeuroImage 30,203–213.

Grady, C.L., Springer, M.V., Hongwanishkul, D., McIntosh, A.R., Winocur,G., 2006. Age-related changes in brain activity across the adult lifespan.J. Cogn. Neurosci. 18, 227–241.

Greicius, M.D., Flores, B.H., Menon, V., Glover, G.H., Solvason, H.B.,Kenna, H., Reiss, A.L., Schatzberg, A.F., 2007. Resting-state functionalconnectivity in major depression: abnormally increased contributionsfrom subgenual cingulate cortex and thalamus. Biol. Psychiatry 62,429–437.

Greicius, M.D., Krasnow, B., Reiss, A.L., Menon, V., 2003. Functionalconnectivity in the resting brain: a network analysis of the default modehypothesis. Proc. Natl. Acad. Sci. U. S. A. 100, 253–258.

Greicius, M.D., Menon, V., 2004. Default-mode activity during a passivesensory task: uncoupled from deactivation but impacting activation.J. Cogn. Neurosci. 16, 1484–1492.

Guerit, J.M., 2005. Neurophysiological patterns of vegetative and minimallyconscious states. Neuropsychol. Rehabil. 15, 357–371.

Guntekin, B., Basar, E., 2007. Emotional face expressions are differentiatedwith brain oscillations. Int. J. Psychophysiol. 64, 91–100.

Grigg-Damberger, M., Gozal, D., Marcus, C.L., Quan, S.F., Rosen, C.L.,Chervin, R.D., Wise, M., Picchietti, D.L., Sheldon, S.H., Iber, C., 2007.The visual scoring of sleep and arousal in infants and children. J. Clin.Sleep Med. 3, 201–240.

573A.C.N. Chen et al. / NeuroImage 41 (2008) 561–574

Gusnard, D.A., Akbudak, E., Shulman, G.L., Raichle, M.E., 2001. Medialprefrontal cortex and self-referential mental activity: relation to a defaultmode of brain function. Proc. Natl. Acad. Sci. U. S. A. 98 , 4259–4264.

Harrison, B.J., Yücel, M., Pujol, J., Pantelis, C., 2007. Task-induced deacti-vation of midline cortical regions in schizophrenia assessed with fMRI.Schizophr. Res. 91, 82–86.

Helps, S., James, C., Debener, S., Karl, A., Sonuga-Barke, E.J. in press. Verylow frequency EEG oscillations and the resting brain in young adults: apreliminary study of localisation, stability and association with symp-toms of inattention. J. Neural. Transm.

Hobbs, M.J., Clarke, A.R., Barry, R.J., McCarthy, R., Selikowitz, M., 2007.EEG abnormalities in adolescent males with AD/HD. Clin. Neurophy-siol. 118, 363–371.

Hoogenboom, N., Schoffelen, J.M., Oostenveld, R., Parkes, L.M., Fries, P.,2006. Localizing human visual gamma-band activity in frequency, timeand space. NeuroImage 29, 764–773.

Horovitz, S.G., Fukunaga, M., de Zwart, J.A., van Gelderen, P., Fulton, S.C.,Balkin, T.J., Duyn, J.H., 2007. Low frequency BOLD fluctuationsduring resting wakefulness and light sleep: A simultaneous EEG-fMRIstudy. Hum. Brain Mapp. 38, 306–320.

Hunter, A.M., Cook, I.A., Leuchter, A.F., 2007. The promise of the quantitativeelectroencephalogram as a predictor of antidepressant treatment outcomesin major depressive disorder. Psychiatr. Clin. North Am. 30, 105–124.

Hunter, M.D., Eickhoff, S.B., Miller, T.W., Farrow, T.F., Wilkinson, I.D.,Woodruff, P.W., 2006. Neural activity in speech-sensitive auditory cor-tex during silence. Proc. Natl. Acad. Sci. U. S. A. 103, 189–194.

Husain, A.M., 2006. Electroencephalographic assessment of coma. J. Clin.Neurophysiol. 23, 208–220.

Inanaga, K., 1998. Frontal midline theta rhythm and mental activity. Psy-chiatry Clin. Neurosci. 52, 555–566.

Kagan-Kushnir, T., Roberts, S.W., Snead III, O.C., 2005. Screening electro-encephalograms in autism spectrum disorders: evidence-based guide-line. J. Child Neurol. 20 (3), 197–206.

Klimesch, W., 1997. EEG-alpha rhythms and memory processes. Int. J.Psychophysiol 26, 319–340.

Klimesch, W., 1999. EEG alpha and theta oscillations reflect cognitive andmemory performance: a review and analysis. Brain Res. Rev. 29, 169–195.

Klimesch, W., Doppelmayr, M., Hanslmayr, S., 2006a. Upper alpha ERDand absolute power: their meaning for memory performance. Prog. BrainRes. 159, 151–165.

Klimesch, W., Sauseng, P., Hanslmayr, S., 2006b. EEG alpha oscillations:the inhibition-timing hypothesis. Brain Res. Rev. 53, 63–88.

Klimesch, W., Sauseng, P., Hanslmayr, S., Gruber, W., Freunberger, R.,2007. Event-related phase reorganization may explain evoked neuraldynamics. Neurosci. Biobehav. Rev. 31 (7), 1003–1016.

Laufs, H., Hamandi, K., Salek-Haddadi, A., Kleinschmidt, A.K., Duncan,J.S., Lemieux, L., 2007. Temporal lobe interictal epileptic dischargesaffect cerebral activity in “default mode” brain regions. Hum. BrainMapp 25, 894–901.

Mantini, D., Perrucci, M.G., Del Gratta, C., Romani, G.L., Corbetta, M.,2007. Electrophysiological signatures of resting state networks in thehuman brain. Proc. Natl. Acad. Sci. U. S. A. 104, 13170–13175.

Marsh, R., Zhu, H., Wang, Z., Skudlarski, P., Peterson, B.S., 2007. A devel-opmental fMRI study of self-regulatory control in Tourette's syndrome.Am. J. Psychiatry 164, 955–966.

Menon, V., Leroux, J., White, C.D., Reiss, A.L., 2004. Frontostriatal deficitsin fragile X syndrome: relation to fMR1 gene expression. Proc. Natl.Acad. Sci. U. S. A 101, 3615–3620.

Nagai, Y., Critchley, H.D., Featherstone, E., Trimble, M.R., Dolan, R.J.,2004. Activity in ventromedial prefrontal cortex covaries with sympa-thetic skin conductance level: a physiological account of a “defaultmode” of brain function. NeuroImage 22 (1), 243–251 May.

Neuper, C., Wortz, M., Pfurtscheller, G., 2006. ERD/ERS patterns reflectingsensorimotor activation and deactivation. Prog. Brain Res. 159, 211–222.

Nuwer, M.R., Comi, G., Emerson, R., Fuglsang-Frederiksen, A., Guerit, J.M.,Hinrichs, H., Ikeda, A., Luccas, F.J., Rappelsberger, P., 1999. IFCNstandards for digital recording of clinical EEG. The International Fede-

ration of Clinical Neurophysiology. Electroencephalogr. Clin. Neurophy-siol. 52, 11–14.

Onton, J., Delorme, A., Makeig, S., 2005. Frontal midline EEG dynamicsduring working memory. NeuroImage 27, 341–356.

Oostenveld, R., Praamstra, P., 2001. The five percent electrode system forhigh-resolution EEG and ERP measurements. Clin. Neurophysiol. 112,713–719.

Parkes, L.M., Bastiaansen, M.C., Norris, D.G., 2006. Combining EEG andfMRI to investigate the post-movement beta rebound. NeuroImage 29,685–696.

Perrin, F., Pernier, J., Bertrand, O., Giard,M.H., Echallier, J.F., 1987.Mappingof scalp potentials by surface spline interpolation. Electroencephalogr.Clin. Neurophysiol. 66, 75–81.

Pfurtscheller, G., Stancak, A.J., Neuper, C., 1996. Event-related synchro-nization (ERS) in the alpha band-an electrophysiological correlate ofcortical idling: a review. Int. J. Psychophysiol. 24, 39–46.

Pollen, D.A., Trachtenberg, M.C., 1972. Some problems of occipital alphablock in man. Brain Res. 41, 303–314.

Preissl, H., Lowery, C.L., Eswaran, H., 2004. Fetal magnetoencephalo-graphy: current progress and trends. Exp. Neurol. 190, 28–36.

Prichep, L.S., 2007. Quantitative EEG and electromagnetic brain imaging inaging and in the evolution of dementia. Ann.N.Y.Acad. Sci. 1097, 156–167.

Raichle, M.E., MacLeod, A.M., Snyder, A.Z., Powers, W.J., Gusnard, D.A.,Shulman, G.L., 2001. A default mode of brain function. Proc. Natl. Acad.Sci. U. S. A. 98, 676–682.

Raichle, M.E., Snyder, A.Z., 2007. A default mode of brain function: a briefhistory of an evolving idea. NeuroImage 237, 1083–1090.

Rihs, T.A., Michel, C.M., Thut, G., 2007. Mechanisms of selective inhibitionin visual spatial attention are indexed by alpha-band EEG synchroniza-tion. Eur. J. Neurosci. 25, 603–610.

Rombouts, S.A., Barkhof, F., Goekoop, R., Stam, C.J., Scheltens, P., 2005.Altered resting state networks in mild cognitive impairment and mildAlzheimer's disease: an fMRI study. Hum. Brain Mapp. 26, 231–239.

Rombouts, S.A., Damoiseaux, J.S., Goekoop, R., Barkhof, F., Scheltens, P.,Smith, S.M., Beckmann, C.F., in press. Model-free group analysis showsaltered BOLD FMRI networks in dementia. Hum. Brain Mapp.

Rossini, P.M., Del Percio, C., Pasqualetti, P., Cassetta, E., Binetti, G., Dal Forno,G., Ferreri, F., Frisoni, G., Chiovenda, P., Miniussi, C., Parisi, L., Tombini,M., Vecchio, F., Babiloni, C., 2006. Conversion from mild cognitiveimpairment toAlzheimer's disease is predicted by sources and coherence ofbrain electroencephalography rhythms. Neuroscience 143, 793–803.

Sauseng, P., Hoppe, J., Klimesch, W., Gerloff, C., Hummel, F.C., 2007.Dissociation of sustained attention from central executive functions: localactivity and interregional connectivity in the theta range. Eur. J. Neurosci.25, 587–593.

Sediri, H., Bourriez, J.L., Derambure, P., 2007. Role of EEG in the diagnosisof brain death. Rev. Neurol. 163, 248–253.

Smit, D.J., Posthuma, D., Boomsma, D.I., De Geus, E.J., 2007. The relationbetween frontal EEG asymmetry and the risk for anxiety and depression.Biol. Psychol. 74, 26–33.

Sorg, C., Riedl, V., Mühlau, M., Calhoun, V.D., Eichele, T., Läer, L.,Drzezga, A., Förstl, H., Kurz, A., Zimmer, C.,Wohlschläger, A.M., 2007.Selective changes of resting-state networks in individuals at risk forAlzheimer's disease. Proc. Natl. Acad. Sci. U. S. A. 104, 18760–18765.

Srinivasan, N., 2007. Cognitive neuroscience of creativity: EEG basedapproaches. Methods 42, 109–116.

Tandonnet, C., Burle, B., Hasbroucq, T., Vidal, F., 2005. Spatial enhance-ment of EEG traces by surface Laplacian estimation: comparison be-tween local and global methods. Clin. Neurophysiol. 116, 18–24.

Tsujimoto, T., Shimazu, H., Isomura, Y., 2006. Direct recording of thetaoscillations in primate prefrontal and anterior cingulate cortices. J. Neuro-physiol. 95, 2987–3000.

van Albada, S.J., Robinson, P.A., 2007. Transformation of arbitrary distri-butions to the normal distribution with application to EEG test-retestreliability. J. Neurosci. Methods 16, 205–211.

van der Hiele, K., Vein, A.A., Reijntjes, R.H., Westendorp, R.G., Bollen,E.L., van Buchem, M.A., van Dijk, J.G., Middelkoop, H.A., 2007. EEG

574 A.C.N. Chen et al. / NeuroImage 41 (2008) 561–574

correlates in the spectrum of cognitive decline. Clin. Neurophysiol. 118,1931–1939.

Vanhatalo, S., Kaila, K., 2006. Development of neonatal EEG activity:from phenomenology to physiology. Semin. Fetal Neonatal. Med. 11,471–478.

Vuilleumier, P., Driver, J., 2007. Modulation of visual processing byattention and emotion: windows on causal interactions between humanbrain regions. Philos. Trans. R. Soc. Lond., B Biol. Sci. 362, 837–855.

Wang, L., Zang, Y., He, Y., Liang, M., Zhang, X., Tian, L., Wu, T., Jiang, T.,Li, K., 2006. Changes in hippocampal connectivity in the early stages of

Alzheimer's disease: evidence from resting state fMRI. NeuroImage 31,496–504.

Wilding, E.L., Herron, J.E., 2006. Electrophysiological measures of episodicmemory control andmemory retrieval. Clin. EEGNeurosci. 37, 315–321.

Wilson, S.M., Molnar-Szakacs, I., Iacoboni, M., 2007. Beyond SuperiorTemporal Cortex: Intersubject Correlations in Narrative Speech Compre-hension. Cereb. Cortex 15 May.

Zhao, X.H., Wang, P.J., Li, C.B., Hu, Z.H., Xi, Q., Wu, W.Y., Tang, X.W.,2007. Altered default mode network activity in patient with anxietydisorders: An fMRI study. Eur. J. Radiol 63, 373–378.