ORIGINAL RESEARCH

published: 22 July 2016

doi: 10.3389/fnhum.2016.00372

Temporal Dynamics of the Default

Mode Network Characterize
Meditation-Induced Alterations in
Consciousness
Rajanikant Panda 1,2†, Rose D. Bharath 1,2* †, Neeraj Upadhyay 3, Sandhya Mangalore
2
, Srivas Chennu 4,5 and Shobini L. Rao 1
1
Cognitive Neuroscience Center, National Institute for Mental Health and Neurosciences, Bangalore, India, 2 Department
of Neuroimaging and Interventional Radiology, National Institute for Mental Health and Neurosciences, Bangalore, India,
3
Department of Neurology and Psychiatry, Sapienza University of Rome, Rome, Italy, 4 School of Computing, University
of Kent, Chatham Maritime, UK, 5 Department of Clinical Neurosciences, University of Cambridge, Cambridge, UK

Current research suggests that human consciousness is associated with complex,

synchronous interactions between multiple cortical networks. In particular, the default
mode network (DMN) of the resting brain is thought to be altered by changes
in consciousness, including the meditative state. However, it remains unclear how
meditation alters the fast and ever-changing dynamics of brain activity within this
network. Here we addressed this question using simultaneous
electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) to
compare the spatial extents and temporal dynamics of the DMN during rest and
Edited by:
Claudia Voelcker-Rehage,
meditation. Using fMRI, we identified key reductions in the posterior cingulate hub of
Chemnitz University of Technology, the DMN, along with increases in right frontal and left temporal areas, in
Germany
experienced meditators during rest and during meditation, in comparison to healthy
Reviewed
by:
controls (HCs). We employed the simultaneously recorded EEG data to identify the
Tao Liu, topographical microstate corresponding to activation of the DMN. Analysis of the
Zhejiang University, China
Márk Molnár,
temporal dynamics of this microstate revealed that the average duration and
Hungarian Academy of Sciences, frequency of occurrence of DMN microstate was higher in meditators compared to
Hungary
HCs. Both these temporal parameters increased during meditation, reflecting the
*Correspondence:
Rose D. Bharath
state effect of meditation. In particular, we found that the alteration in the duration of
[email protected] the DMN microstate when meditators entered the meditative state correlated
[email protected]
negatively with their years of meditation experience. This reflected a trait effect of
† These authors

have contributed meditation, highlighting its role in producing durable changes in temporal dynamics of
equally to this work. the DMN. Taken together, these findings shed new light on short and long- term
Received: 08 May 2016
Accepted: 11 July 2016 consequences of meditation practice on this key brain network.
Published: 22 July 2016
Keywords: default mode network, microstate, DMN-microstate, simultaneous EEG-fMRI, meditation
Citation:
Panda R, Bharath RD, Upadhyay N, INTRODUCTION
Mangalore S, Chennu S and Rao SL
(2016) Temporal Dynamics of the The grand challenge of characterizing the dynamic neural substrate underlying
Default Mode Network Characterize
human consciousness has captured the interest of many researchers cutting across disciplinary
Meditation-Induced
Alterations in Consciousness. boundaries and covering altered states ranging from sleep, meditation, hypnosis, anesthesia,
Front. Hum. Neurosci. 10:372. coma, disorders of consciousness, delirium tremens, psychoses, etc. Normal consciousness is
doi: 10.3389/fnhum.2016.00372 thought to require both wakefulness and arousal, and several neuro scientifc studies
conceptualize wakefulness as a

Frontiers in Human Neuroscience | www.frontiersin.org 1 July 2016 | Volume 10 | Article 372

Panda et al. Temporal Dynamics of the DMN and Meditation

continuum with different levels of awareness (Grill-Spektor have durations of 100–200 ms (Brandeis and Lehmann, 1989
et al., 2000; Bar et al., 2001; Kouider et al., 2010). Resting ; Pascual-Marqui et al., 1995; Michel et al., 2004; Lehmann et
state functional magnetic resonance imaging (rsfMRI) has al., 2010; Baker et al., 2014). The concept of microstates was
been used to study the neural correlates of conscious first proposed and demonstrated by Lehmann et al. (1987)
awareness in normal and altered states of consciousness. In when they described brain states that remain stable for 80–
particular, this research enterprise has highlighted and 120 ms before rapidly evolving into another quasi-stable
progressively refined our understanding of the so-called microstate. The most common parameters used to
default mode network (DMN) of the brain, consisting of quantify microstate dynamics are duration or lifespan, which
precuneus/posterior cingulate cortex (PCC), medial frontal is the average length of time each microstate remains stable
cortex (mPFC), the temporoparietal junction (TPJ) and whenever it appears. Another useful parameter is frequency
hipocampal formation including parahipocampal cortex, as a of occurrence, which is the average number of times per
key neural correlate of consciousness (Buckner et al., 2008). second that the microstate becomes dominant (Lehmann et
In particular, researchers have studied changes in the DMN al., 1987). Most microstate studies reports four classic
as a function of meditative and introspective cognitive microstates which can explain more than 70% of the variation
states like day dreaming, mind wandering, and in the scalp topographies manifesting in EEG time series (
autobiographical memory retrieval (Baerentsen et al., 2010; Tei et al., 2009; Khanna et al., 2015), which have been
Baron Short et al., 2010; Hasenkamp et al., 2012; found to be correlated with rsfMRI networks associated
Garrison et al., 2013). These investigations into the with phonological processing, visual processing, the salience
neurophenomenology of meditation have found PCC network, and attentional switching (Mantini et al., 2007; Britz
deactivation to be associated with ‘‘undistracted et al., 2010). Other studies have reported a higher number of
awareness’’ and ‘‘effortless doing’’, with PCC activation microstates (10–13) with different analysis methods (Musso
being linked to ‘‘distracted awareness’’ and ‘‘controlling’’ et al., 2010; Yuan et al., 2012), some of which correlate
in experienced meditators (Garrison et al., 2013). This with the DMN. Here, we refer to the ‘‘DMN microstate’’ as
evidence is in line with the contrasting approaches in two the EEG microstate which correlated maximally with the DMN
major types of meditation, which either emphasize focused identified with fMRI. Such microstate analyses have been
attention or open monitoring. Focused attention meditation applied to the study of meditation, where increased duration
aims to reduce mind wandering by concentrating on tasks of microstates has been reported in EEG-based studies in
like breath, sounds or mental imagery, while open monitoring Chan-meditators and Ch’anMo’chao, or Vipassana meditators
meditation practice encourages mind wandering, and makes (Faber et al., 2005; Lo and Zhu, 2009; Tei et al., 2009).
one aware of this process (Xu et al., 2014). Extensive study of Microstate parameters have also been shown to be modulated
the DMN in health and disease have also found correlations by psychiatric disorders (Dierks et al., 1997; Lehmann et al.,
between DMN connectivity and sleep (Fukunaga et al., 2006 2005; Kikuchi et al., 2011; Nishida et al., 2013) and even by
; Horovitz et al., 2008; Picchioni et al., 2008), anesthesia ( personality type (Schlegel et al., 2012).
Vincent et al., 2007), disorders of consciousness ( In this study, we analyzed the DMN microstate
Fernández-Espejo et al., 2012; Guldenmund et al., 2012). to understand the mechanisms of meditation-induced
While rsfMRI has enabled us to have a fine-grained alterations in consciousness. By contrasting healthy
spatial understanding of the DMN in these states of controls (HCs) at rest against expert meditators at rest and
consciousness, conventional analytical approaches are not during meditation, we explored both state and trait changes
best suited to measure the temporal dynamics of its activity, in DMN-microstate dynamics produced by meditation with a
especially during meditation-induced alterations in the hypothesis that these could cause differential alterations in its
state of consciousness. To better understand the temporal duration and frequency. The state changes felt during
dynamics of the DMN, researchers have developed a range meditation are usually described as a deep sense of calm
of techniques, including time resolved resting fMRI analysis peacefulness, cessation or slowing of mind’s internal dialog
where DMN connectivity is assessed multiple times using and conscious awareness merging completely with the
sliding temporal windows (Chang and Glover, 2010; object of meditation (Brown, 1977; Wallace, 1999).
Hutchison et al., 2013; Leonardi et al., 2013; Allen et al., Alongside, long-term expertise in meditation also
2014; Zalesky et al., 2014). Alternative techniques to obtain produces durable changes in neural dynamics, with
temporal detail with fMRI include identification of improvements in mental and physical health presumably due
temporal functional modes using temporal independent to its trait effects (Chiesa and Serretti, 2010, 2011; Hofmann
component analysis (ICA; Smith et al., 2012), modified seed et al., 2010). Here, we describe changes in the spatial
to voxel based connectivity to define coactivation patterns ( configuration of the DMN as a function of meditation, and
Liu and Duyn, 2013) and the recent innovation driven show that state and trait influences on the temporal dynamics
coactivation pattern (Karahanoglu and Van De Ville, 2015). of the DMN microstate
MATERIALS ANDcan indeed be dissociated.
METHODS
While these methods have attempted to capture the
temporal dynamics of the DMN, the time scales of the Participants
observed fluctuations in its activity vary from tens of This was a prospective study conducted at a tertiary
seconds to few minutes in fMRI based studies (Chang and neurological institute, the National Institute of Mental
Glover, 2010; Handwerker et al., 2012; Hutchison et al., 2013). Health and
In contrast, modeling of electromagnetic brain dynamics using
electroencephalography
Frontiers (EEG)
in Human Neuroscience and MEG suggest that these 2
| www.frontiersin.org July 2016 | Volume 10 | Article 372
neuronal fluctuations called microstates
Panda et al. Temporal Dynamics of the DMN and Meditation

Neurosciences (NIMHANS) in Bangalore, India. The study Data Acquisition

was performed after obtaining informed written consent EEG Data Acquisitions
from the participants. They were recruited as healthy EEG data were recorded using a 32-channel MR-
participants in the multi-institutional study on cognitive compatible EEG system (Brain Products GmbH, Gilching,
networks. Ethical approval was obtained from the Germany). The EEG cap (BrainCap MR, Brain Products)
institutional ethics committee for studies involving humans, consisted of 31 scalp electrodes placed according to the
convened by NIMHANS (No.NIMHANS/69th IES/2010). international 10–20 system electrode placement and one
The meditator cohort included 20 Raja Yoga expert additional electrode dedicated to the ECG which was
meditators (male; age: placed on the back of the subject. Data were recorded
35 ± 7.9 years, years of education: 15.4 ± 1.6 years, right relative to an FCz reference and a ground electrode
handed) from the Brahma Kumaris Spiritual Organization, (GND) according to 10–20 electrode system. Data was
all of them with more than 10 years of Raja yoga sampled at 5000 Hz to enable removal of MRI gradient
meditation practice. Raja yoga meditation involves artifact. The impedance between electrode and scalp was
internally visualizing a glowing star as rays emerging kept below 5 kΩ. EEG was recorded using the Brain
between the eye brows and thus could be considered a
Recorder software (Version 1.03, Brain Products). To prevent
type of focused attention type of meditation practice. All
the movement of the subjects’ head, we placed sufficient
participants reported that they spent 1.68 ± 0.59 h in
padding between the head and the head coil of the MRI
meditation per day in the last 10–22 years (15.2 ± 3.54).
scanner. Total time for each rest EEG recoding was same
They also reported having a cumulative experience of
as rsfMRI recording (9.24 min), and time locked to each
11332.5 ± 6009.86 h of meditation practice (cumulative
other.
experience calculated by combining the numbers of hours
fMRI Data Acquisition
per day with years of meditation practice) in their life.
rsfMRI was acquired using a 3T scanner (Skyra, Siemens,
Twenty HCs were also recruited for the study. The control
Erlangen, Germany). One hundred and eighty-five
participants were matched with the meditator cohort by
volumes of gradient-echo, Echo-Planar Images (EPI) were
age, gender, education and ethnicity (male; age:
obtained using the following EPI parameters: 36 slices, 4
29 ± 6.8 years, years of education: 16.1 ± 1.1 years, right
mm slice thickness in interleaved manner with an field of
handed) and both the groups were comparable. None of
view (FOV) of 192 × 192 mm, matrix 64 × 64,
controls had experience in any type of regular meditative
repetition time (TR) 3000 ms, echo time (TE) 35 ms,
practices. Both meditators and control participants were
re-focusing pulse 90◦, matrix- 256 × 256 × 114, voxel
multilingual (languages known 3.38 ± 0.58), with
size-3 × 3 × 4 mm. We also acquired a three
kannada as first language. None of the participants had
dimensional magnetization prepared rapid acquisition
any history of neurological or psychiatric illnesses, or prior
gradient echo (3D MPRAGE) sequence for anatomical
trauma, and were not on any chronic medications that
could affect the experiment. information (with the voxel size 1 × 1 × 1 mm, 192 × 192
× 256 matrix) for better registration and overlay of brain
activity.
Experiment Design
The study design was explained to the subjects and Data Analysis
instructions were given before performing the EEG-fMRI EEG Artifact Removal and Preprocessing
data collection in the MRI scanner. The participants Raw EEG data was processed offline using BrainVision
(both controls and meditators) were instructed to lie Analyzer software version 2 (Brain Products GmbH, Gilching,
awake, with their eyes closed in a relaxed state within Germany). The gradient artifacts were corrected according to
the MRI gantry. They were advised to refrain from any Allen et al. (2000). A moving average width of 20 MR
cognitive, language or motor tasks during the acquisition. volumes (TRs) was used for gradient correction. Corrected
Ear plugs were given to reduce scanner induced noise. EEG data were filtered using a high-pass filter at 0.03 Hz and
Initially a 4.24 min structural MRI was recorded. This a low-pass filter at 75 Hz. The data was then down-sampled to
allowed them to get familiarized with the MRI 250 Hz. Ballistocardiogram (BCG) artifacts were removed from
environment before starting the rest and meditation the EEG using an averaged subtraction method using
session. A single resting EEG-fMRI was obtained in HCs heartbeat events (R peaks; Allen et al., 1998, 2000; Goldman et
(9.24 min). However in meditators two serial (9.24 min al., 2000), implemented in BrainVision Analyzer 2. Once
each) resting EEG-fMRI were recorded, one during gradient and BCG artifacts were removed, the data were
resting wakefulness, followed immediately by one downsampled to 250 Hz. They were visually inspected for
during meditation. During a post hoc interview, all meditators artifacts resulting from muscular sources or head movement
reported that they could satisfactorily achieve the artifact and any epoch containing any channel varying
meditative state inside the scanner. This was also more than 150 µV was rejected. Finally, the signal was
subsequently confirmed with analysis of the EEG wave filtered with a band-pass of 0.01–45 Hz. To confirm that the
form, where we visually inspected the EEG time series meditators were in the meditative state during the second
in the time and spectral domains to identify an increase session, EEG spectral analysis was carried out for the entire
in alpha oscillations during meditation, consistent with 9.24 min for resting state and for the entire 9.24 min for
previous studies which have reported this increase ( meditative state. A fast fourier
Takahashi
Frontiers et al.,
in Human 2005; Lagopoulos
Neuroscience et al., 2009).
| www.frontiersin.org 3 July 2016 | Volume 10 | Article 372
Panda et al. Temporal Dynamics of the DMN and Meditation

transform (FFT) was used to calculate spectral power response function using customized three column format in
density (µV2). FEAT Version 5.98, part of FSL (FMRIB’s Software Library2).
The GLM was modeled as an event related design, where
EEG Feature Extraction EEG microstates were considered as explanatory variables
The EEG microstates were derived from the resting EEG (regressors) for rsfMRI analysis. This procedure, enabled us
data using sLORETA software. To compute microstates the to combine EEG and fMRI time series despite their very
local maxima of the global field power (GFP; Lehmann different sampling rates. To do so, the onset time and the
and Skrandies, 1980) were identified. Since scalp duration of each EEG microstate were provided as inputs
topography remains stable around peaks of GFP as a result to the GLM and convolved with a gamma hemodynamic
of temporal smoothing in sLORETA, they are representative response function (Musso et al., 2010). This rsfMRI, thus
of the transient microstates (Pascual-Marqui et al., 1995; informed by the EEG microstates, was analyzed at the
Koenig et al., 2002). The algorithm implemented for individual level to derive subject-wise Z-stats maps for
estimating microstates is based on a modified version of each microstate. We examined the spatial correlation (R)
the classical k-means clustering method in which cluster between the spectral Z-stats map and the DMN component
orientations are estimated. An optimal number of 13 estimated by ICA of blood-oxygen- level dependent (BOLD)
clusters for this method was determined by means of activity (see below). Using the FSL utility ‘‘fslcc’’, the Z-
a cross-validation criterion (Pascual-Marqui et al., 1995). stats maps which correlated with BOLD ICA map of the
The entire EEG data at each time point was DMN with a correlation of at least 0.3 were selected for
decomposed into a temporal sequence of one of these 13 further analysis (Kiviniemi et al., 2011). When more than
EEG microstates. We used this EEG microstate one microstate Z-map correlated with the DMN the one
decomposition to guide the analysis of the rsfMRI data at with highest correlation was selected. This subject-wise
the single subject level. spectral Z-stats map of the DMN microstate was included in
further group analysis. Group level analysis was done
rsfMRI Preprocessing using the higher level FEAT tool, where three group mean
The rsfMRI analysis was performed using FSL software effects (controls at rest, meditators at rest and meditation
(FMRIB’s Software Library1), in particular with the fMRI state) were extracted by specifying the contrast (1 0 0; 0
Expert Analysis Tool (FEAT) and multivariate exploratory 1 0; 0 0 1) respectively.
linear decomposition into independent components Independent Components (ICs)
(MELODIC) modules. The first five functional images (EPI Functional connectivity was ascertained using ICA which
volumes) were discarded from each of the subjects’ decomposes the rsfMRI data into statistically independent
rsfMRI data to allow for signal equilibration, giving a spatial maps (and associated time series). This is a data
total of 180 volumes used in analysis. We conducted driven approach by which we can extract the functional
motion correction using MCFLIRT (Jenkinson et al., 2002), networks in a voxel-wise manner based on their spatial
and non-brain tissue (Scalp, CSF, etc.) removal using the and temporal signal fluctuations. ICA of rsfMRI was
Brain Extraction Toolbox (Smith, 2002). The average carried out with the probabilistic independent components
head motion of the subjects was also not found to analysis (PICA) using FSL’s MELODIC method. This single-
significantly differ between the groups. Spatial smoothing subject ICA was carried out with FEAT MELODIC data
was performed using a Gaussian kernel of 5 mm full width exploration, in which the rsfMRI data was extracted into
at half maximum (FWHM), followed by a mean based the independent components (ICs) during the preprocessing
intensity normalization of all volumes by the same step. All extracted IC were visually inspected. Non-motion
factor, and then a high temporal band-pass filtering related noise sources such as eye related artifacts, field in
with sigma 180 s. These temporal filtering parameters were homogeneity, high frequency noise, slice dropout and
based on previously validated methodology employed by gradient instability were removed on the basis of existing
Beckmann and Smith (Smith et al., 2004; Beckmann and literature (Smith et al., 2004; Beckmann and Smith, 2005;
Smith, 2005; Beckmann et al., 2005). We co-registered the Beckmann et al., 2005). ICs that showed clearly
EPI volumes with the individual subject’s high-resolution identifiable patterns in both spatial and frequency
anatomical volume in MNI152 standard space template domains, and activations in frequency plots below 0.1 Hz
and re-sampled the filtered data into standard space, were removed to reduce respiration and heart rate related
keeping the re-sampled resolution at the fMRI resolution artifacts. The selected noise components were removed using
(3 mm) using FNIRT, the FMRIB non-linear image the command ‘‘fsl_regfilt’’ defined in FSL MELODIC.
registration tool (Klein et al., 2009). Two participants from Multivariate group PICA was then carried out using
each group were excluded due to uncorrectable cardio- FSL MELODIC (Beckmann and Smith, 2005; Beckmann
ballistic and motion artifacts during data preprocessing. et al., 2005), to derive maximally spatially ICs across all
datasets (18 control and 18 meditator × 2 sessions),
Identification of DMN-Microstate which were temporally concatenated. The data set was
The spectral Z-stats maps of the EEG microstates were created decomposed into 15 sets of independent vectors which
1www.fmrib.ox.ac.uk/fsl
by convolving all the 13 microstates with the gamma describe signal variation across the
2http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/

hemodynamic

Frontiers in Human Neuroscience | www.frontiersin.org 4 July 2016 | Volume 10 | Article 372

Panda et al. Temporal Dynamics of the DMN and Meditation

temporal (timecourses) and spatial (maps) domain by When the voxel wise differences between the groups were
optimizing for non-Gaussian spatial source distributions higher, the term ‘‘increased connectivity’’ was used and
using the Fast ICA algorithm. The resulting maps were vice versa.
threshold using an alternative hypothesis test with a
significance level of p > 0.5 (i.e., an equal weight is placed RESULTS
on false positives and false negatives) by fitting a
Gaussian/Gamma mixture model to the histogram of Despite being in an MRI scanner with gradient acoustic
intensity values (Smith et al., 2004; Beckmann and Smith, noise, with a fixed head and body position, all meditators
2005). Finally, the components representing the DMN reported that they were able to stay awake in relaxed state
were visually identified from the set of group IC maps and also enter into a meditative state as instructed. Visual
by comparing their constituent cortical sources with the inspection of the EEG for sleep spindles and K-complexes
regions of the DMN reported in the literature ( confirmed that none of the subjects fell asleep during
Beckmann and Smith, 2005; Beckmann et al., 2005; rsfMRI data acquisition. In all the results presented below,
Damoiseaux et al., 2006). three conditions are compared, controls at rest, meditators
at rest, and the same meditators during meditation.
DMN-Microstate Statistical Analysis
The mean duration and number of occurrences per second Group Level EEG Frequency Differences
of the DMN microstates were calculated for each subject. During Meditation
The duration of the DMN-microstate was defined as a Significant changes were noted in spectral power of
continuous time epoch (milliseconds) during which all EEG
EEG GFP maps were assigned to the the DMN-microstate frequency bands between the meditation and rest states in
class by the k-means clustering algorithm. The frequency of meditators. Meditators had increased power particularly in
occurrence of the DMN- microstate is defined the number the alpha, theta and beta frequencies in the meditation
of such occurrences per second. Group differences were state compared to rest state, which is depicted in the Figure 1
calculated for both the DMN- microstate duration and . This also ascertained that the meditators did not go into
frequency of occurrence for all subjects. Independent samples meditative state during resting state and performed meditation
t-tests were carried out to compare the difference between in meditative state as instructed.
meditators and controls at rest, and a paired sample t-test
was carried out between the meditator at rest and during Spatial Analysis of DMN-Microstate and
meditation. At a 95% confidence interval (CI), a p-value of
< 0.05 was considered significant. Furthermore, the number
DMN During Meditation
Figure 2 depicts the average DMN-microstate, the
of years of meditation practice were correlated with the
corresponding
duration and occurrence of DMN microstates with a
Z-stats map and rsfMRI-derived DMN ICA component in
pearson correlation. Multiple comparisons were compensated
HCs, meditators at rest (meditator) and during meditation.
for by FDR correction.
Dual regression analysis (Figure 3) of the rsfMRI DMN
revealed that meditators were different from HCs (p <
DMN Dual Regression Analysis
0.05) already at rest, as they had decreased posterior
Groupwise comparison of rsfMRI ICs was performed using
cingulate (number of voxels: 223, T-max: 0.987, coordinates:
dual regression, which allows for voxel-wise comparisons3.
−4, −40, 26) connectivity, increased right middle frontal
The set of spatial maps from the group-mean analysis
gyrus (MFG; number of voxels: 263, T-max: 0.956,
(GroupICs) was used to generate subject-specific-spatial
coordinates: 42, 8, 46) and left middle temporal gyrus
maps and associated time series (Beckmann et al., 2009;
(MTG; number of voxels: 53, T-max: 0.963, coordinates:
Filippini et al., 2009). DMN component maps from all
−52, −48, 6) connectivity. This
subjects were arranged into a single 4D file and voxel-
wise differences were estimated. Statistical differences were
assessed using randomized, on- parametric permutation,
incorporating the threshold-free cluster enhancement (TFCE)
technique (Smith et al., 2009). Separate null distributions of
t-values were derived for the contrasts reflecting the inter-
group effects by performing 5000 random permutations and
testing the difference between groups or against zero for
each iteration (Nichols and Holmes, 2002). To estimate group
mean effects, HC > meditator, meditator > HCs, meditator >
meditation and mediation > meditator contrasts were used
and the resultant statistical maps were threshold at p < FIGURE 1 | Power spectra. Power spectral density distribution for
0.05 (family-wise error (FWE) corrected). Following this, meditators at meditative state (red line) and at rest (blue line), control at rest
inter-group effects were threshold by at p < 0.05
3http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/DualRegression (black line).
(false discovery rate (FDR) corrected; Filippini et al., 2009).

Frontiers in Human Neuroscience | www.frontiersin.org 5 July 2016 | Volume 10 | Article 372

Panda et al. Temporal Dynamics of the DMN and Meditation

FIGURE 2 | Default mode network (DMN) microstates during rest and meditation. (A) Group averaged electroencephalography (EEG) microstate in controls
at rest (control), meditator at rest state (meditator) and meditator during meditation (meditation). (B) Group averaged Z-stats maps corresponding to the DMN
microstates. (C) Corresponding DMN components derived from resting state functional magnetic resonance imaging (rsfMRI) using independent components
analysis (ICA). The striking spatial similarity between the EEG-derived DMN Z-stats map (B) and rsfMRI-derived DMN (C) is evident. Meditators showed evidence of
decreased connectivity in precuneus/posterior cingulate cortex (PCC) and medial prefrontal cortex (mPFC) at rest, which decreased further during meditation (C).

relatively lower cingulate connectivity was further observed increased left MTG (number of voxels: 122, T-
significantly reduced relative to rest, when the meditators max: 0.992, coordinates: −58, −46, 8) and right MFG
entered into the meditative state (number of voxels: 107, (number of
T-max: 0.998, coordinates: −2, −40, 26). Alongside this voxels: 424, T-max: 0.998, coordinates: 40, 8, 46)
reduction, we also connectivity during meditation.

Frontiers in Human Neuroscience | www.frontiersin.org 6 July 2016 | Volume 10 | Article 372

Panda et al. Temporal Dynamics of the DMN and Meditation

FIGURE 3 | Dual regression analysis of group main and between-group effects in the rsfMRI DMN component. Between-group effects showed significant
(p < 0.05 false discovery rate (FDR) corrected) decreased PCC connectivity in meditators compared to controls, which further decreased during meditation.
(A) Meditators had higher connectivity in right middle frontal gyrus (MFG) and left middle temporal gyrus (MTG) than controls, which again increased during
meditation (B).

Temporal Analysis of the DMN-Microstate meditators with their meditation experience in years. The
The meditators at rest had increased duration and results are summarized in Figure 5. There was a very strong,
higher frequency of occurrence of the DMN-microstate significant positive correlation in the duration of DMN-
compared to controls (Figure 4).The mean duration of the microstate in meditators at rest (r = 0.96, p = 7.3E-11) and
DMN-microstate was 93.05 ± 15.18 ms in HCs and in the meditative state (r = 0.51, p = 0.03) with years of
118.88 ± 12.78 ms in meditators at rest (Unpaired t-test, t = meditation practice (Figure 5A). Interestingly, in each
−3.55; df = 34; p = 3.6E- 06); the mean frequency of its meditator, the net increase in the duration of the DMN-
occurrence was 3.15 ± 0.66/s in controls and 3.84 ± 0.48/s microstate during meditation relative to rest actually
in meditators (Unpaired t-test, t = −3.56; df = 34; p = correlated negatively with the years of meditation practice (
0.001). During meditation the duration and frequency of Figure 5C, r = −0.549, p = 0.023). This implied that the
occurrence increased significantly further. Specifically, more experienced meditators had a DMN-microstate
during meditation, the mean duration of DMN- microstate duration akin to being in a meditative state even at rest,
was 134.27 ± 11.21 ms (Paired t-test, t = −5.06; df = 17; highlighting how the state effect of meditation on DMN
p = 9.5E-05) and the mean frequency of occurrence was dynamics gradually transitions into a trait effect that
4.03 ± 0.46/s (Paired t-test, t = −5.06; df = 17; p = 1.3E- potentially permanently alters it over many years of
06; practice. However, this effect could potentially be
Figure 4). explained away by the age of the meditators rather than
Correlation of DMN-Microstate Dynamics their years of experience, as participants with longer years
With Years of Experience in Meditation of experience were older than those with lesser experience.
The observed differences in the temporal properties of the To address this potential confound, we regressed out this
DMN- effect by including age of the meditator as a covariate in the
microstate suggested a combination of state and trait- correlation analysis. Having accounted for the influence of
based influences. To further explicate the influence of age, we still found a significant negative correlation between
long-term meditation experience in engendering trait the state-induced change in DMN-microstate duration and
differences between meditators and controls, we years of mediation experience (r = −0.549, p = 0.03). This
correlated the duration and frequency of occurrence of the confirmed that there was indeed a trait-level modulation of
DMN-microstate in individual the DMN’s dynamics produced
Frontiers in Human Neuroscience | www.frontiersin.org 7 July 2016 | Volume 10 | Article 372
Panda et al. Temporal Dynamics of the DMN and Meditation

FIGURE 4 | Temporal dynamics of the DMN-microstate. Bar charts plot the duration (A) and occurence/s (B) of the DMN-microstate in healthy controls (HCs),
meditators at rest and during meditation. Both measures were significantly higher in meditators than in controls, and increased further during meditation.

by meditation experience, that could not be explained entering into a meditative state was associated with
away by aging. significant increases both in the duration and frequency of
Further, we found that this pattern of effects were the DMN- microstate. As experience with meditation
specific to the duration of the DMN-microstate. Specifically, increased, there was a progressively higher duration of the
the effect of meditation experience on the frequency of DMN-microstate at rest and entering a state of meditation
occurrence of the DMN-microstate was not statistically only brought about an increase in its frequency of
significant, either at rest (r = 0.43, p = 0.06) or in the occurrence. Taken together, these findings elucidate
meditative state (r = 0.46, p = 0.053), though there was a salient spatiotemporal mechanisms of meditation, and
trend towards increasing DMN-microstate occurrence with particularly highlight its role in altering the temporal
increasing years of meditation practice (Figure 5B). Further, dynamics of brain activity over short and long time scales.
the meditation state induced increase in the frequency of Within the large body of fMRI literature on resting
occurrence did not correlate with years of practice (r = state networks, the DMN has been a particular focus of
−0.004, p = 0.98). As the frequency of occurrence of the attention (Raichle et al., 2001; Buckner et al., 2008) as a
DMN-microstate was unaffected by the years of key brain network that underlies awareness of the self
experience, it was distinct from duration and is probably within the environment. Our finding of decreased PCC
indicative of the state effect of meditation, as seen in Figure connectivity and increased temporal and frontal
4B. connectivity within the DMN during meditation has been
reported previously in seed to voxel connectivity studies (
DISCUSSION Farb et al., 2007; Brewer et al., 2011). Alongside, the
significance of the temporo-frontal regions during
We have investigated dynamic alterations of the DMN meditation has been consistently highlighted in previous
using simultaneous EEG-fMRI, with the aim of task based fMRI studies (Brefczynski-Lewis et al., 2007;
characterizing the spatial and temporal changes caused by Lagopoulos et al., 2009). Indeed, our findings mirror those
meditation induced alterations of consciousness. Using from recent studies on fMRI of meditation (Baijal and
spatial ICA analysis of our fMRI data, we found that the Srinivasan, 2010; Cahn et al., 2010; Hasenkamp and Barsalou
posterior cingulate hub of the DMN was less strongly , 2012; Hasenkamp et al., 2012). In particular, Hasenkamp
connected in meditators, which became further reduced and Barsalou (2012) studied the effects of meditation
during meditation. In contrast, the right middle frontal and experience on DMN functional connectivity, and
left middle temporal gyri were more active in meditators and demonstrated increased connectivity in frontal areas during
during meditation. By linking the fMRI with EEG, we were meditation in meditators with more experience, which they
able to identify the DMN microstate in all participants, by interpreted as evidence of greater attentional control gained
matching its spatial configuration to their respective DMN with such experience. In our meditator cohort, these
ICA components. The temporal properties of this DMN- attentional control networks could also be similarly more
microstate highlighted its significantly higher duration active in the expert meditators. This increased attentional
and frequency of occurrence, in meditators at rest and control, along with decreased PCC connectivity, is
during meditation. Further, the relationship between these hypothesized to undergo neuroplastic changes that enable
temporal properties and years of meditation experience them to gain better control of mind-wandering. Based on
enabled us to explore state and trait influences on DMN evidence of task positive networks being recruited at rest,
dynamics. This analysis suggested that in less experienced it has been argued that continuous and regular
meditators,
Frontiers in Human Neuroscience | www.frontiersin.org 8
meditation July 2016 | Volume 10 | Article 372
Panda et al. Temporal Dynamics of the DMN and Meditation

FIGURE 5 | Correlation of DMN-microstate dynamics with meditation experience. The duration (A) and frequency of occurrence (B) of the DMN-microstate
were positively correlated with years of experience. The blue and red lines in dicate the linear fits of these DMN-microstate properties measured during meditation
and rest, respectively. Duration was less altered as a function of state in more experienced meditators. Corroborating this trait effect, the difference in
DMN-microstate duration between meditation and rest was negatively correlated with years of experience (C). No such trait effect was observed in the frequency of
occurrence of the DMN-microstate (D).

practice transforms the rest state of experienced meditators noting that the focus of the analysis here has been on the
(Brewer et al., 2011). However in the current study the DMN- microstate, as that was the only microstate that was
right frontal and left temporal areas described are part consistently present in all subjects. This is similar to
of the DMN and do not represent task positive networks. findings reported by Musso et al. (2010), whose group
However, it is possible that this neuroplasticity is analysis only revealed a fronto-occipital network that
mediated by task positive networks, and future research overlapped with visual networks and the DMN. This is
could elucidate how meditation modulates the relationship despite the fact that all the subjects in their study had
between these different brain networks. microstates which correlated with several other resting state
Within this research context, our finding have generated networks. With the continual advancement in methodological
new evidence delineating temporal changes in DMN tools for simultaneous EEG-fMRI analysis, we might be
activity that accompanies these spatial changes. Our able parse spatiotemporal brain dynamics with the requisite
finding of increasing frequency of occurrence of the detail and consistency for mapping all the resting state
DMN-microstate during meditiation implies that there is a networks.
decreased occurrence of other microstates, thus resulting in Our finding of experienced meditators having a higher
an overall decrease in variability in the temporal dynamics of duration of DMN microstate that did not change much in
brain activity. According to large scale dynamical models of the meditative state is consistent with the philosophical
consciousness (Deco et al., 2013), resting brain can be perspective (Lutz et al., 2008), behavioral (Carter et al., 2005;
divided into three states, namely the spontaneous state, Srinivasan and Baijal, 2007) and brain structural changes (
multistable attractor states and unstable spontaneous state, Lazar et al., 2005) related to the trait effects of meditative
differing in their coupling strengths. The state with least practice. Because of the synergistic interaction between state
coupling strength is the spontaneous state and that with and trait effects it has been largely difficult to design studies
highest coupling strength is the unstable spontaneous state to separate these effects (Cahn and Polich, 2006) and hence
often associated with a task. Human consciousness is relatively few studies that have exclusively looked at the
postulated to be positioned at the verge of instability defined state effect of meditation. In future research, the study of
to lie between the multistable attractor state and the neurophenomenology of meditation, which correlates moment-
unstable spontaneous state. Within this framework, decreased to-moment first person reports of internal experience to
variability within such a dynamical system is compatible with guide neuroimaging analysis might be able to separate
our finding of increased frequency of the DMN microstate state effect from trait effects of meditation (Jack and Shallice
during meditation. It is worth , 2001; Jack and Roepstorff, 2002; Lutz et al., 2002).
Frontiers in Human Neuroscience | www.frontiersin.org 9 July 2016 | Volume 10 | Article 372
Panda et al. Temporal Dynamics of the DMN and Meditation

CONCLUSION followed by RDB, SC, NU revised the manuscript critically

for important intellectual content. Final approval and
Our investigation of the spatial configuration of the DMN agreement to be accountable for all aspects of the work in
in meditators and during meditation revealed a decrease in ensuring that questions related to the accuracy or integrity
posterior cingulate connectivity, and a complementary of any part of the work are appropriately investigated and
increase in middle frontal and temporal connectivity. Using resolved by all authors.
temporal microstate analysis applied to simultaneous EEG-
fMRI data, we found that the duration and frequency of
FUNDING
occurrence of the DMN-microstates are useful metrics of
meditation- induced altered consciousness, and increase We acknowledge the support of the Department of Science
systematically across controls, meditators during rest and and Technology, Govt. of India, India for providing the 3T
during meditation. We report complementary roles of these MRI scanner exclusively for research in the field of
metrics, where duration primarily reflected the trait effect neurosciences.
and occurrence represented the state effect of meditation. Our
results shed new light on the neurobiology of meditation, and ACKNOWLEDGMENTS
its effect on the spatiotemporal properties of the DMN in
We thank Dr. S. R. Chandra for ascertaining the quality of
particular.
EEG (Department of Neurology, National Institute of
AUTHOR CONTRIBUTIONS Mental Health and Neurosciences (NIMHANS), India).
We thank BK. Ambika, BK. Sneha, BK. Sushilchandra and
RP and RDB have contributed equally for this article. BK. Srikant from Spiritual Application Research Center
Contributions to the conception or design of the work by (SpARC), Prajapita Brahamakumari Iswariya Viswa
RP, RDB, SLR. Data acquisition done by RP and NU and Vidyalaya for facilitating the participation of expert
analysis was carried out by RP, RDB, NU, SM. RDB, SC meditators. We are grateful to the staff especially the
interpreted the results of the study. Manuscript drafted by radiographers, for their support. Finally, we thank all the
RP participants for their time.

REFERENCES Beckmann, C. F., and Smith, S. M. (2005). Tensorial extensions of

independent component analysis for multisubject fMRI analysis.
Allen, E. A., Damaraju, E., Plis, S. M., Erhardt, E. B., Eichele, T., and Neuroimage 25, 294–311. doi: 10.1016/j.neuroimage.2004.10.043
Calhoun, Brandeis, D., and Lehmann, D. (1989). Segments of event-related potential
V. D. (2014). Tracking whole-brain connectivity dynamics in the resting state. map series reveal landscape changes with visual attention and subjective
Cereb. Cortex 24, 663–676. doi: 10.1093/cercor/bhs352 contours. Electroencephalogr. Clin. Neurophysiol. 73, 507–519. doi:
Allen, P. J., Josephs, O., and Turner, R. (2000). A method for removing 10.1016/0013- 4694(89)90260-5
imaging artifact from continuous EEG recorded during functional MRI. Brefczynski-Lewis, J. A., Lutz, A., Schaefer, H. S., Levinson, D. B., and
Neuroimage 12, 230–239. doi: 10.1006/nimg.2000.0599 Davidson,
Allen, P. J., Polizzi, G., Krakow, K., Fish, D. R., and Lemieux, L. R. J. (2007). Neural correlates of attentional expertise in long-term meditation
(1998). Identification of EEG events in the MR scanner: the problem of practitioners. Proc. Natl. Acad. Sci. U S A 104, 11483–11488. doi:
pulse artifact and a method for its subtraction. Neuroimage 8, 229–239. 10.1073/pnas. 0606552104
doi: 10.1006/nimg. 1998.0361 Brewer, J. A., Worhunsky, P. D., Gray, J. R., Tang, Y.-Y., Weber, J., and
Baerentsen, K. B., Stødkilde-Jørgensen, H., Sommerlund, B., Hartmann, Kober,
T., Damsgaard-Madsen, J., Fosnaes, M., et al. (2010). An H. (2011). Meditation experience is associated with differences in default
investigation of brain processes supporting meditation. Cogn. Process. mode network activity and connectivity. Proc. Natl. Acad. Sci. U S A
11, 57–84. doi: 10. 1007/s10339-009-0342-3 108, 20254–20259. doi: 10.1073/pnas.1112029108
Baijal, S., and Srinivasan, N. (2010). Theta activity and meditative states: Britz, J., Van De Ville, D., and Michel, C. M. (2010). BOLD correlates of
spectral changes during concentrative meditation. Cogn. Process. 11, 31– EEG topography reveal rapid resting-state network dynamics. Neuroimage
38. doi: 10. 1007/s10339-009-0272-0 52, 1162–1170. doi: 10.1016/j.neuroimage.2010.02.052
Baker, A. P., Brookes, M. J., Rezek, I. A., Smith, S. M., Behrens, T., Smith, P. J. Brown, D. P. (1977). A model for the levels of concentrative
P., et al. (2014). Fast transient networks in spontaneous human brain meditation. Int. J. Clin. Exp. Hypn. 25, 236–273. doi:
activity. Elife 3:e01867. doi: 10.7554/elife.01867 10.1080/002071477084 15984
Bar, M., Tootell, R. B. H., Schacter, D. L., Greve, D. N., Fischl, B., Mendola, J. Buckner, R. L., Andrews-Hanna, J. R., and Schacter, D. L. (2008). The
D., et al. (2001). Cortical mechanisms specific to explicit visual object brain’s default network: anatomy, function, and relevance to disease. Ann. N
recognition. Neuron 29, 529–535. doi: 10.1016/s0896-6273(01)00224-0 Y Acad. Sci. 1124, 1–38. doi: 10.1196/annals.1440.011
Baron Short, E., Kose, S., Mu, Q., Borckardt, J., Newberg, A., George, M. S., et Cahn, B. R., Delorme, A., and Polich, J. (2010). Occipital gamma activation
al. (2010). Regional brain activation during meditation shows time and during Vipassana meditation. Cogn. Process. 11, 39–56. doi:
practice effects: an exploratory FMRI study. Evid Based Complement. 10.1007/s10339-009- 0352-1
Alternat. Med. 7, 121–127. doi: 10.1093/ecam/nem163 Cahn, B. R., and Polich, J. (2006). Meditation states and traits: EEG, ERP
Beckmann, C. F., DeLuca, M., Devlin, J. T., and Smith, S. M. (2005). and neuroimaging studies. Psychol. Bull. 132, 180–211. doi: 10.1037/0033-
Investigations into resting-state connectivity using independent 2909.132. 2.180
component analysis. Philos. Trans. R. Soc. Lond. B Biol. Sci. 360, Carter, O. L., Presti, D. E., Callistemon, C., Ungerer, Y., Liu, G. B.,
1001–1013. doi: 10.1098/rstb. 2005.1634 and Pettigrew, J. D. (2005). Meditation alters perceptual rivalry in
Beckmann, C. F., Mackay, C. E., Filippini, N., and Smith, S. M. Tibetan Buddhist monks. Curr. Biol. 15, R412–R413. doi:
(2009). Group comparison of resting-state fMRI data using multi- 10.1016/j.cub.2005. 05.043
subject ICA and dual regression. Neuroimage 47:S148. doi: Chang, C., and Glover, G. H. (2010). Time-frequency dynamics of resting-
10.1016/s1053-8119(09) 71511-3 state brain connectivity measured with fMRI. Neuroimage 50, 81–98. doi:
10.1016/j. neuroimage.2009.12.011

Frontiers in Human Neuroscience | www.frontiersin.org 10 July 2016 | Volume 10 | Article 372

Panda et al. Temporal Dynamics of the DMN and Meditation

Chiesa, A., and Serretti, A. (2010). A systematic review of neurobiological and interpretations. Neuroimage 80, 360–378. doi:
and clinical features of mindfulness meditations. Psychol. Med. 40, 10.1016/j.neuroimage.2013. 05.079
1239–1252. doi: 10.1017/s0033291709991747 Jack, A. I., and Shallice, T. (2001). Introspective physicalism as an
Chiesa, A., and Serretti, A. (2011). Mindfulness based cognitive therapy approach to the science of consciousness. Cognition 79, 161–196. doi:
for psychiatric disorders: a systematic review and meta-analysis. 10.1016/s0010- 0277(00)00128-1
Psychiatry Res. 187, 441–453. doi: 10.1016/j.psychres.2010.08.011 Jack, A. I., and Roepstorff, A. (2002). Introspection and cognitive brain
Damoiseaux, J. S., Rombouts, S. A. R. B., Barkhof, F., Scheltens, P., Stam, C. mapping: from stimulus-response to script-report. Trends Cogn. Sci. 6,
J., Smith, S. M., et al. (2006). Consistent resting-state networks across 333–339. doi: 10. 1016/s1364-6613(02)01941-1
healthy subjects. Proc. Natl. Acad. Sci. U S A 103, 13848–13853. doi: Jenkinson, M., Bannister, P., Brady, M., and Smith, S. (2002).
10.1073/pnas. 0601417103 Improved optimization for the robust and accurate linear registration
Deco, G., Jirsa, V. K., and McIntosh, A. R. (2013). Resting brains never and motion correction of brain images. Neuroimage 17, 825–841. doi:
rest: computational insights into potential cognitive architectures. Trends 10.1006/nimg.2002. 1132
Neurosci. 36, 268–274. doi: 10.1016/j.tins.2013.03.001 Karahanoglu, F. I., and Van De Ville, D. (2015). Transient brain
Dierks, T., Jelic, V., Julin, P., Maurer, K., Wahlund, L. O., Almkvist, O., et activity disentangles fMRI resting-state dynamics in terms of spatially and
al. (1997). EEG-microstates in mild memory impairment and Alzheimer’s temporally overlapping networks. Nat. Commun. 6:7751. doi:
disease: possible association with disturbed information processing. J. 10.1038/ncomms8751
Neural. Transm. (Vienna) 104, 483–495. doi: 10.1007/bf01277666 Khanna, A., Pascual-Leone, A., Michel, C. M., and Farzan, F. (2015).
Faber, P. L., Lehmann, D., Barendregt, H., Kaelin, M., and Gianotti, L. R. R. Microstates in resting-state EEG: current status and future directions.
(2005). Increased duration of EEG microstates during meditation. Neurosci. Biobehav. Rev. 49, 105–113. doi:
Brain Topogr. 18, 131. 10.1016/j.neubiorev.2014.12.010
Farb, N. A. S., Segal, Z. V., Mayberg, H., Bean, J., Mckeon, D., Fatima, Kikuchi, M., Koenig, T., Munesue, T., Hanaoka, A., Strik, W., Dierks, T., et
Z., et al. (2007). Attending to the present: mindfulness meditation reveals al. (2011). EEG microstate analysis in drug-naive patients with panic
distinct neural modes of self-reference. Soc. Cogn. Affect. Neurosci. 2, 313– disorder. PLoS One 6:e22912. doi: 10.1371/journal.pone.0022912
322. doi: 10. 1093/scan/nsm030 Kiviniemi, V., Vire, T., Remes, J., Elseoud, A. A., Starck, T., Tervonen, O., et
Fernández-Espejo, D., Soddu, A., Cruse, D., Palacios, E. M., Junque, al. (2011). A sliding time-window ICA reveals spatial variability of the
C., Vanhaudenhuyse, A., et al. (2012). A role for the default mode network default mode network in time. Brain Connect. 1, 339–347. doi:
in the bases of disorders of consciousness. Ann. Neurol. 72, 335–343. doi: 10.1089/brain.2011. 0036
10.1002/ana. 23635 Klein, A., Andersson, J., Ardekani, B. A., Ashburner, J., Avants, B.,
Filippini, N., MacIntosh, B. J., Hough, M. G., Goodwin, G. M., Frisoni, G. Chiang,
B., Smith, S. M., et al. (2009). Distinct patterns of brain activity in young M. C., et al. (2009). Evaluation of 14 nonlinear deformation algorithms applied
carriers of the APOE-E4 allele. Proc. Natl. Acad. Sci. U S A 106, 7209– to human brain MRI registration. Neuroimage 46, 786–802. doi:
7214. doi: 10. 1016/s1053-8119(09)71381-3 10.1016/j. neuroimage.2008.12.037
Fukunaga, M., Horovitz, S. G., van Gelderen, P., de Zwart, J. A., Koenig, T., Prichep, L., Lehmann, D., Sosa, P. V., Braeker, E., Kleinlogel, H., et
Jansma, al. (2002). Millisecond by millisecond, year by year: normative EEG
J. M., Ikonomidou, V. N., et al. (2006). Large-amplitude, spatially correlated microstates and developmental stages. Neuroimage 16, 41–48. doi:
fluctuations in BOLD fMRI signals during extended rest and early sleep 10.1006/nimg.2002.1070 Kouider, S., de Gardelle, V., Sackur, J., and
stages. Magn. Reson. Imaging 24, 979–992. doi: 10.1016/j.mri.2006.04.018 Dupoux, E. (2010). How rich is consciousness? The partial awareness
Garrison, K. A., Santoyo, J. F., Davis, J. H., Thornhill, T. A., Kerr, C. E., and hypothesis. Trends Cogn. Sci. 14,
Brewer, 301–307. doi: 10.1016/j.tics.2010.04.006
J. A. (2013). Effortless awareness: using real time neurofeedback to investigate Lagopoulos, J., Xu, J., Rasmussen, I., Vik, A., Malhi, G. S., Eliassen, C. F., et
correlates of posterior cingulate cortex activity in meditators’ self-report. al. (2009). Increased theta and alpha EEG activity during nondirective
Front. Hum. Neurosci. 7:440. doi: 10.3389/fnhum.2013.00440 meditation.
Goldman, R. I., Stern, J. M., Engel, J., and Cohen, M. S. (2000). J. Altern. Complement. Med. 15, 1187–1192. doi: 10.1089/acm.2009.0113
Acquiring simultaneous EEG and functional MRI. Clin. Neurophysiol. Lazar, S. W., Kerr, C. E., Wasserman, R. H., Gray, J. R., Greve, D. N.,
111, 1974–1980. doi: 10.1016/s1388-2457(00)00456-9 Treadway,
Grill-Spektor, K., Kushnir, T., Hendler, T., and Malach, R. (2000). The M. T., et al. (2005). Meditation experience is associated with increased
dynamics of object-selective activation correlate with recogntion cortical thickness. Neuroreport 16, 1893–1897. doi:
performance in human. Nat. Neurosci. 3, 837–843. doi: 10.1038/77754 10.1097/01.wnr.0000186598.
Guldenmund, J. P., Vanhaudenhuyse, A., Boly, M., Laureys, S., and Soddu, 66243.19
A. (2012). A default mode of brain function in altered states of Lehmann, D., and Skrandies, W. (1980). Reference-free identification
consciousness. Arch. Ital. Biol. 150, 107–121. doi: 10.4449/aib.v150i2.1373 of components of checkerboard-evoked multichannel potential
Handwerker, D. A., Roopchansingh, V., Gonzalez-Castillo, J., and Bandettini, P. fields. Electroencephalogr. Clin. Neurophysiol. 48, 609–621. doi:
A. (2012). Periodic changes in fMRI connectivity. Neuroimage 63, 1712– 10.1016/0013- 4694(80)90419-8
1719. doi: 10.1016/j.neuroimage.2012.06.078 Lehmann, D., Faber, P. L., Galderisi, S., Herrmann, W. M., Kinoshita, T.,
Hasenkamp, W., and Barsalou, L. W. (2012). Effects of meditation experience Koukkou, M., et al. (2005). EEG microstate duration and syntax in
on functional connectivity of distributed brain networks. Front. Hum. acute, medication- naïve, first-episode schizophrenia: a multi-center
Neurosci. 6:38. doi: 10.3389/fnhum.2012.00038 study. Psychiatry Res. 138, 141–156. doi:
Hasenkamp, W., Wilson-Mendenhall, C. D., Duncan, E., and Barsalou, L. 10.1016/j.pscychresns.2004.05.007
W. (2012). Mind wandering and attention during focused meditation: Lehmann, D., Ozaki, H., and Pal, I. (1987). EEG alpha map series:
a fine- grained temporal analysis of fluctuating cognitive states. brain micro-states by space-oriented adaptive segmentation.
Neuroimage 59, 750–760. doi: 10.1016/j.neuroimage.2011.07.008 Electroencephalogr. Clin. Neurophysiol. 67, 271–288. doi: 10.1016/0013-
Hofmann, S. G., Sawyer, A. T., Witt, A. A., and Oh, D. (2010). 4694(87)90025-3
The effect of mindfulness-based therapy on anxiety and depression: a Lehmann, D., Pascual-Marqui, R. D., Strik, W. K., and Koenig, T. (2010).
meta- analytic review. J. Consult. Clin. Psychol. 78, 169–183. doi: Core networks for visual-concrete and abstract thought content: a brain
10.1037/a00 18555 electric microstate analysis. Neuroimage 49, 1073–1079. doi:
Horovitz, S. G., Fukunaga, M., de Zwart, J. A., van Gelderen, P., Fulton, S. 10.1016/j.neuroimage. 2009.07.054
C., Balkin, T. J., et al. (2008). Low frequency BOLD fluctuations during Leonardi, N., Richiardi, J., Gschwind, M., Simioni, S., Annoni, J. M., Schluep,
resting wakefulness and light sleep: a simultaneous EEG-fMRI study. M., et al. (2013). Principal components of functional connectivity: a new
Hum. Brain Mapp. 29, 671–682. doi: 10.1002/hbm.20428 approach to study dynamic brain connectivity during rest. Neuroimage
Hutchison, R. M., Womelsdorf, T., Allen, E. A., Bandettini, P. A., Calhoun, V. 83, 937–950. doi: 10.1016/j.neuroimage.2013.07.019
D., Corbetta, M., et al. (2013). Dynamic functional connectivity: promise, Liu, X., and Duyn, J. H. (2013). Time-varying functional network
issues information extracted from brief instances of spontaneous brain activity.
Frontiers in Human Neuroscience | www.frontiersin.org 11 Proc. Natl. Acad. Sci. U SJulyA2016110, | Volume 10 | Article 372
4392–4397. doi:
10.1073/pnas.1216856110
Lo, P. C., and Zhu, Q. (2009). ‘‘Microstate analysis of alpha-event brain
topography during chan meditation,’’ in Proceedings of the 2009
Panda et al. Temporal Dynamics of the DMN and Meditation

ongoing conscious states during a simple visual task. Proc. Natl. Acad. Sci. U image analysis and implementation as FSL. Neuroimage 23, S208–S219. doi:
S A 99, 1586–1591. doi: 10.1073/pnas.032658199 10. 1016/j.neuroimage.2004.07.051
Lutz, A., Slagter, H. A., Dunne, J. D., and Davidson, R. J. (2008). Smith, S. M., Miller, K. L., Moeller, S., Xu, J., Auerbach, E. J., Woolrich, M.
Attention regulation and monitoring in meditation. Trends Cogn. Sci. W., et al. (2012). Temporally-independent functional modes of
12, 163–169. doi: 10.1016/j.tics.2008.01.005 spontaneous brain activity. Proc. Natl. Acad. Sci. U S A 109, 3131–
Mantini, D., Perrucci, M. G., Del Gratta, C., Romani, G. L., and Corbetta, 3136. doi: 10.1073/pnas. 1121329109
M. (2007). Electrophysiological signatures of resting state networks in the Srinivasan, N., and Baijal, S. (2007). Concentrative meditation
human brain. Proc. Natl. Acad. Sci. U S A 104, 13170–13175. doi: enhances preattentive processing: a mismatch negativity study.
10.1073/pnas. 0700668104 Neuroreport 18, 1709–1712. doi: 10.1097/wnr.0b013e3282f0d2d8
Michel, C. M., Murray, M. M., Lantz, G., Gonzalez, S., Spinelli, L., and Grave Takahashi, T., Murata, T., Hamada, T., Omori, M., Kosaka, H., Kikuchi, M., et
de Peralta, R. (2004). EEG source imaging. Clin. Neurophysiol. 115, al. (2005). Changes in EEG and autonomic nervous activity during
2195–2222. doi: 10.1016/j.clinph.2004.06.001 meditation and their association with personality traits. Int. J.
Musso, F., Brinkmeyer, J., Mobascher, A., Warbrick, T., and Winterer, G. Psychophysiol. 55, 199–207. doi: 10.1016/j.ijpsycho.2004.07.004
(2010). Spontaneous brain activity and EEG microstates. A novel Tei, S., Faber, P. L., Lehmann, D., Tsujiuchi, T., Kumano, H., Pascual-
EEG/fMRI analysis approach to explore resting-state networks. Neuroimage Marqui,
52, 1149–1161. doi: 10. 1016/j.neuroimage.2010.01.093 R. D., et al. (2009). Meditators and non-meditators: EEG source imaging during
Nichols, T. E., and Holmes, A. P. (2002). Nonparametric permutation tests resting. Brain Topogr. 22, 158–165. doi: 10.1007/s10548-009-0107-4
for functional neuroimaging: a primer with examples. Hum. Brain Mapp. 15, Vincent, J. L., Patel, G. H., Fox, M. D., Snyder, A. Z., Baker, J. T., Van Essen, D.
1–25. doi: 10.1002/hbm.1058 C., et al. (2007). Intrinsic functional architecture in the anaesthetized
Nishida, K., Morishima, Y., Yoshimura, M., Isotani, T., Irisawa, S., Jann, monkey brain. Nature 447, 83–86. doi: 10.1038/nature05758
K., et al. (2013). EEG microstates associated with salience and Wallace, B. A. (1999). The buddhist tradition of samatha: methods for refining
frontoparietal networks in frontotemporal dementia, schizophrenia and examining consciousness. J. Conscious. Stud. 6, 175–187.
and Alzheimer’s disease. Clin. Neurophysiol. 124, 1106–1114. doi: Xu, J., Vik, A., Groote, I. R., Lagopoulos, J., Holen, A., Ellingsen, O., et al.
10.1016/j.clinph.2013. 01.005 (2014). Nondirective meditation activates default mode network and areas
Pascual-Marqui, R. D., Michel, C. M., and Lehmann, D. (1995). Segmentation associated with memory retrieval and emotional processing. Front. Hum.
of brain electrical activity into microstates: model estimation and Neurosci. 8:86. doi: 10.3389/fnhum.2014.00086
validation. IEEE Trans. Biomed. Eng. 42, 658–665. doi: 10.1109/10.391164 Yuan, H., Zotev, V., Phillips, R., Drevets, W. C., and Bodurka, J.
Picchioni, D., Fukunaga, M., Carr, W. S., Braun, A. R., Balkin, T. J., Duyn, J. (2012). Spatiotemporal dynamics of the brain at rest–exploring EEG
H., et al. (2008). fMRI differences between early and late stage-1 sleep. microstates as electrophysiological signatures of BOLD resting state
Neurosci. Lett. 441, 81–85. doi: 10.1016/j.neulet.2008.06.010 networks. Neuroimage 60, 2062–2072. doi:
Raichle, M. E., MacLeod, A. M., Snyder, A. Z., Powers, W. J., Gusnard, D. A., 10.1016/j.neuroimage.2012.02.031
and Shulman, G. L. (2001). A default mode of brain function. Proc. Natl. Zalesky, A., Fornito, A., Cocchi, L., Gollo, L. L., and Breakspear, M.
Acad. Sci. U S A 98, 676–682. doi: 10.1073/pnas.98.2.676 (2014). Time-resolved resting-state brain networks. Proc. Natl. Acad. Sci.
Schlegel, F., Lehmann, D., Faber, P. L., Milz, P., and Gianotti, L. R. R. (2012). U S A of
Conflict 111,Interest
10341–10346. doi: 10.1073/pnas.1400181111
Statement: The authors declare that the research
EEG microstates during resting represent personality differences. Brain was conducted in the absence of any commercial or financial relationships
Topogr. 25, 20–26. doi: 10.1007/s10548-011-0189-7 that could be construed as a potential conflict of interest.
Smith, S. M. (2002). Fast robust automated brain extraction. Hum. Brain
Mapp. Copyright © 2016 Panda, Bharath, Upadhyay, Mangalore, Chennu and Rao.
17, 143–155. doi: 10.1002/hbm.10062 This is an open-access article distributed under the terms of the Creative
Smith, S. M., Fox, P. T., Miller, K. L., Glahn, D. C., Fox, P. M., Mackay, C. E., et Commons Attribution License (CC BY). The use, distribution and reproduction
al. (2009). Correspondence of the brain’s functional architecture during in other forums is permitted, provided the original author(s) or licensor are
activation and rest. Proc. Natl. Acad. Sci. U S A 106, 13040–13045. doi: credited and that the original publication in this journal is cited, in accordance
10.1073/pnas. 0905267106 with accepted academic practice. No use, distribution or reproduction is
Smith, S. M., Jenkinson, M., Woolrich, M. W., Beckmann, C. F., Behrens, T. E. permitted which does not comply with these terms.
J., Johansen-Berg, H., et al. (2004). Advances in functional and
structural MR

Frontiers in Human Neuroscience | www.frontiersin.org 12 July 2016 | Volume 10 | Article 372