Occipital gamma activation during Vipassana meditation

Abstract

Long-term Vipassana meditators sat in meditation vs. a control rest (mind-wandering) state for 21 min in a counterbalanced design with spontaneous EEG recorded. Meditation state dynamics were measured with spectral decomposition of the last 6 min of the eyes-closed silent meditation compared to control state. Meditation was associated with a decrease in frontal delta (1-4 Hz) power, especially pronounced in those participants not reporting drowsiness during meditation. Relative increase in frontal theta (4-8 Hz) power was observed during meditation, as well as significantly increased parieto-occipital gamma (35-45 Hz) power, but no other state effects were found for the theta (4-8 Hz), alpha (8-12 Hz), or beta (12-25 Hz) bands. Alpha power was sensitive to condition order, and more experienced meditators exhibited no tendency toward enhanced alpha during meditation relative to the control task. All participants tended to exhibit decreased alpha in association with reported drowsiness. Cross-experimental session occipital gamma power was the greatest in meditators with a daily practice of 10+ years, and the meditation-related gamma power increase was similarly the strongest in such advanced practitioners. The findings suggest that long-term Vipassana meditation contributes to increased occipital gamma power related to long-term meditational expertise and enhanced sensory awareness.

Full text

RESEARCH REPORT
Occipital gamma activation during Vipassana meditation
B. Rael Cahn
•
Arnaud Delorme
•
John Polich
Received: 27 October 2009 / Accepted: 26 November 2009 / Published online: 16 December 2009
Ó The Author(s) 2009. This article is published with open access at Springerlink.com
Abstract Long-term Vipassana meditators sat in medi-
tation vs. a control rest (mind-wandering) state for 21 min
in a counterbalanced design with spontaneous EEG recor-
ded. Meditation state dynamics were measured with spec-
tral decomposition of the last 6 min of the eyes-closed
silent meditation compared to control state. Meditation was
associated with a decrease in frontal delta (1–4 Hz) power,
especially pronounced in those participants not reporting
drowsiness during meditation. Relative increase in frontal
theta (4–8 Hz) power was observed during meditation, as
well as significantly increased parieto-occipital gamma
(35–45 Hz) power, but no other state effects were found for
the theta (4–8 Hz), alpha (8–12 Hz), or beta (12–25 Hz)
bands. Alpha power was sensitive to condition order, and
more experienced meditators exhibited no tendency toward
enhanced alpha during meditation relative to the control
task. All participants tended to exhibit decreased alpha in
association with reported drowsiness. Cross-experimental
session occipital gamma power was the greatest in medi-
tators with a daily practice of 10? years, and the medita-
tion-related gamma power increase was similarly the
strongest in such advanced practitioners. The findings
suggest that long-term Vipassana meditation contributes to
increased occipital gamma power related to long-term
meditational expertise and enhanced sensory awareness.
Keywords Meditation  Electroencephalography (EEG) 
Gamma  Mental state  Altered state of consciousness
(ASC)  Vipassana
Introduction
The term meditation refers a set of diverse and specific
methods of distinct attentional engagement, and recent
reports have begun to focus specifically on state and trait
measures of the same. Although the general effects of
meditation on neuroelectric brain measures are still being
characterized, consensus has emerged that coherence and/
or power for lower frequency spontaneous electroenceph-
alographic (EEG) activity is enhanced as both a trait and a
state effect in many forms of meditative practice (Cahn and
Polich 2006).
Early studies of meditators implicated alpha (8–12 Hz)
power increases as both a state and trait effect of Yogic,
Zen, and Transcendental Meditation practice (Anand et al.
1961a; Kasamatsu and Hirai 1966; Kasamatsu et al. 1957;
Wallace 1970; Wenger and Bagchi 1961). Later studies
have failed to replicate the early findings of increased alpha
in advanced practitioners but have reported increased
alpha coherence, especially in assays of TM practitioners
B. R. Cahn (&)
Division of Geriatric Psychiatry, Department of Psychiatry,
University of California San Diego, 8950 Villa La Jolla Drive,
Suite B-122, La Jolla, CA 92037, USA
e-mail: rael@ucsd.edu
A. Delorme
Institute for Neural Computation,
University of California San Diego,
La Jolla, CA, USA
A. Delorme
CERCO, CNRS, Universite Paul Sabatier,
133 Route de Narbonne, 31062 Toulouse Cedex 9, France
J. Polich
Cognitive Electrophysiology Laboratory,
Molecular and Integrative Neurosciences Department,
The Scripps Research Institute,
10550 North Torrey Pines Road,
La Jolla, CA 92037, USA
e-mail: polich@scripps.edu
123
Cogn Process (2010) 11:39–56
DOI 10.1007/s10339-009-0352-1

(Gaylord et al. 1989; Travis 1991; Travis and Pearson
1999; Travis et al. 2002), theta (4–8 Hz) power, especially
in the assays of concentrative/focused attention practitio-
ners (Aftanas and Golocheikine 2001; Baijal and Sriniva-
san 2009; Hebert and Lehmann 1977; Pan et al. 1994), or
gamma effects (Lehmann et al. 2001; Lutz et al. 2004).
Recent reports using LORETA to analyze EEG from Zen
(Faber et al. 2008) and Qi-Gong (Tei et al. 2009) medita-
tors further suggest that meditation may be associated with
trait increased frontal delta activity, possibly indexing
baseline relative inhibition of cognitive engagement and
greater detachment from ongoing daily experience. Gamma
findings have been reported as either distinguishing
between various meditative states in an advanced practi-
tioner (Lehmann et al. 2001), or as a state and trait marker
for meditation in advanced Tibetan Buddhist practitioners
engaging in compassion meditation (Lutz et al. 2004). One
likely contributing factor to the inconsistency across stud-
ies is the lack of standardization with respect to meditative
style, assessment methodology, and consideration of state
effects for beginning vs. short- vs. long-term meditators
(Cahn and Polich 2006). Toward this end, the present study
was designed to provide fundamental information using
EEG measures during meditation compared to control state
effects in long-term Vipassana meditators.
Vipassana meditation and present study
Vipassana meditation is a Buddhist practice that involves
focusing on present-moment sensory awareness with an
equanimous and non-reactive mental set (Gunaratana 2002;
Hart 1987). This tradition has served as the foundation for
contemporary ‘‘mindfulness’’ meditation techniques such as
the widely practiced mindfulness-based stress reduction
(MBSR) currently used for clinical interventions (Davidson
2003; Grossman et al. 2004; Kabat-Zinn 1982, 2003).
Development of greater awareness of and concomitant non-
reactivity to interoceptive and exteroceptive sensory stimuli
during formal Vipassana/mindfulness meditation is
hypothesized to enhance self-awareness such that selective
adaptive responding is facilitated at the expense of auto-
mated non-adaptive reactions, thereby promoting more
successful management of stressful life situations (Hart
1987; Kabat-Zinn 2003; Lutz et al. 2007; Segal et al. 2002).
Vipassana practitioners of the Theravadan Vipassana
tradition were assayed in the present study, and the
majority had been taught in the tradition of Goenka (Hart
1987). This practice emphasizes attentional absorption in
subtle somatosensory awareness and associated open
monitoring without mental or emotional reactivity to such
sensory experience. The specific Vipassana meditative
technique involves attentional scanning of sensations
throughout the body in an iterative and cyclic fashion,
scanning body sensations from the top of the head to the
toes and back again repeatedly, with the concomitant
adoption of an attitude of detached observation and non-
reactivity to any sensations and thoughts that may arise.
Systematic evaluation of meditation state in comparison
with non-meditative thought conditions requires control
cognitive tasks, and this condition was implemented with the
instruction to let the mind wander freely through non-emo-
tional thoughts and memories. This state was also chosen to
mimic the aspects of the mind-wandering ‘‘default mode’’
state thought to have high ecological validity to a common
mode of cognitive engagement in normal everyday life
(Christoff et al. 2009; Smallwood and Schooler 2006).
Given that Vipassana meditation practice is thought to
enhance the awareness of internal and external stimuli
while reducing automated reactivity, it was hypothesized
that increased frontal theta and alpha would be observed
during meditation—i.e., reflecting increased purposeful
attentional engagement and mental quiescence. Based on
the view that Vipassana practice may promote an
enhancement of sensory awareness through increased
attentional engagement, we hypothesized that increased
gamma activity would also be observed in meditation rel-
ative to the control condition, possibly in bilateral centro-
parietal and/or frontal areas, related to enhanced processing
in frontal and somatosensory cortices. Recording condi-
tions were designed to capture the time period when the
meditation state was deep and stable and therefore most
likely to contrast with the control state.
Methods
Participants
A total of N = 16 Vipassana meditators (F = 5, M = 11)
were assessed (M = 45.5, SD = 9.8, 24–56 years). As a
group, these individuals had been meditating for a con-
siderable period of time (M = 20.0, SD = 12.1, 2.5–
40 years), and all had been meditating daily (7 days/week)
for at least 1 year (M = 13.0, SD = 10.7, 1–30 years),
with at least 0.5? h or more each day (M = 1.3, SD = 0.7,
0.5–3 h). Participants were recruited from a local Vipas-
sana meditation community through word of mouth and
e-mail and compensated $40 for the 3-h study.
Recording conditions
EEG data were collected using a 19-channel ECI electrode
cap from the following locations: Fp1, Fp2, F3, F4, F7, F8,
Fz, C3, C4, T7, T8, Cz, P3, P4, P7, P8, Pz, O1, and O2.
These scalp locations were referenced to linked earlobes,
with the ground at the forehead. Eye-movement (EOG)
40 Cogn Process (2010) 11:39–56
123

activity was assessed with electrodes placed at the outer
canthi and above/below the left eye in line with the pupil
for horizontal and vertical EOG monitoring using bipolar
reference. Impedances were kept below 10 kX. The signals
were recorded with a band pass of 0.01–70 Hz (6 dB
octave/slope) and digitization rate of 256 Hz.
Procedure
The participants were instructed to sit on cushions and
meditate within the Theravadan Vipassana meditation tra-
dition or engage in the control neutral thinking state, with
the order of the tasks counterbalanced across individuals.
Participants were instructed to sit in the same posture for
both the meditation and control task periods of recording
and were fitted with headphones at the outset of the
recording session that they wore throughout the recording.
Pilot testing indicated that some participants found it dif-
ficult to refrain from engaging in their meditative practice
when sitting in the meditative posture with eyes-closed.
Participants were therefore told to think about emotionally
neutral past events if they noticed themselves slipping into
meditative practice state, otherwise to let their mind wan-
der freely through non-emotional neutral thoughts. This
control cognitive engagement was chosen to emulate a
‘‘mind-wandering’’ state with high ecological validity that
can be contrasted with the purposeful attentional engage-
ment of the meditation state (Christoff et al. 2009; Small-
wood and Schooler 2006). Participants were informed that
after 21 min of eyes-closed meditation or control thinking
they would hear a series of tones over the headphones and
that they were to simply continue their meditation or
control cognitive engagement. Both passive and active
presentations of simple tone stimuli were collected, with
those data reported elsewhere (Cahn and Polich 2009).
At the conclusion of the first recording period, the par-
ticipants were given the opportunity to stand and stretch
before taking the same posture and seating position for the
second recording of equal length. Immediately after each
of the two recordings participants completed a short form
indicating whether they experienced drowsiness or sleep
onset during the experimental recording session and rating
the depth of meditative experience on a 1–10 scale, with 1
indicating the normal waking and 10 indicating the deepest
meditative absorption ever experienced.
EEG analysis
We focused the present analysis on the last 6 min of data
from the 21 min recording period so as to assess the
meditative state achieved after allowing an adequate period
of time for the participants to fully absorb themselves into
the meditation state. Given that the spontaneous brain
rhythms generated during prolonged periods of eyes-closed
rest tend to fluctuate and vary with factors such as reduced
arousal and drowsiness, we did not average across the
whole 21 min epoch. Future planned analysis will focus on
the temporal evolution of brain dynamics in the meditation
vs. control state so as to assess the onset vs. maintenance of
meditation state effects and the relative stability of the
spectral power dynamics across the two states. The last
6 min of EEG data from each of the two 21 min recording
periods were first visually inspected, and transient muscle-
and movement-related artifacts were removed. The data
were subsequently high-pass filtered at 0.5 Hz using
FIR filter (Rabiner and Gold 1975). The extended ICA
algorithm was then run on the data using the runica algo-
rithm implemented within EEGLAB running on Matlab
(Delorme and Makeig 2004; Delorme et al. 2007). The
resultant independent components accounting for horizon-
tal and vertical eye movements were then marked and
removed from the data (Jung et al. 2000a, b) as detailed in
the next section.
After removal of eye-movement-related artifact, data
were segmented into non-overlapping 2-s artifact-free
epochs. For the meditation and control states, means of
170.4 ± 13.8 and 164.1 ± 26.6 epochs were obtained. The
Thomson multi-taper spectrum estimator (Matlab PMTM
functions using time-bandwidth product of 4 and FFT
length of 512) was applied to the cleaned continuous data
for spectral decomposition (Thomson 1982). The output
power values in lV
2
were then log-transformed to dB units
using 10 9 log
10
(lV
2
) formula to normalize the power
value distributions.
Statistical analyses of the spectral power data were first
applied to signals from all scalp channels using the boot-
strap method (Wilcox 2005) and using false discovery rate
(FDR) correction for multiple comparisons (Benjamini and
Yekutieli 2001). Statistical assessments were conducted
using analyses of variance with the factors of state (med-
itation vs. control) and electrode in the a priori regions of
interest (midline electrodes for all frequencies, occipital
electrodes for alpha). ANOVAs also were conducted on the
frontal electrodes (F3 and F4) for delta and the occipital
electrodes (O1 and O2) for gamma activity, as these were
regions of statistical significance between states identified
by the initial bootstrap statistical testing using FDR cor-
rection for the scalp data (see Fig. 2). Greenhouse-Geisser
corrections were applied to the degree of freedom (df)to
correct for violations of the sphericity assumption, and
Tukey post hoc means comparisons were used to decom-
pose reliable interactions.
Covariate ANOVAs were conducted using subject
variables related to the order of engaging in meditation
and control states, relative meditative expertise, and self-
reports of drowsiness and meditative depth during
Cogn Process (2010) 11:39–56 41
123

experimental conditions. Bootstrap statistics and scalp
map plotting were performed using the EEGLAB Matlab
software (Delorme and Makeig 2004; Delorme et al.
2007) and custom Matlab scripts. Parametric statistics
were run using the Statistica software and EEGLAB
(Delorme and Makeig 2004; Delorme et al. 2007). The
use of bootstrap statistics was employed as it allows for
more robust statistical inference than standard parametric
statistics since no assumption is made about the proba-
bility distribution at the population level (Wilcox 2005).
However, studying complex patterns of covariate inter-
action is not yet available using bootstrap procedures in
common statistical software, so that parametric statistics
were employed.
Independent component analysis (ICA)
In addition to the statistical assessments, a parallel analysis
using independent component analysis (ICA) of three
classes of independent components from each subject was
conducted: (1) eye-movement artifact, (2) temporal muscle
artifact, and (3) occipital alpha power. For each subject,
vertical eye-movement-related components were selected
based on their characteristic scalp projection and a smooth
exponentially decreasing spectrum (Delorme et al. 2007;
Jung et al. 2000a). This approach was employed by iden-
tifying the EOG artifacts by simultaneously examining the
eye channels (VEOG and HEOG) for activity to verify that
the components were active only during those time periods
when eye channels indicated eye movement. Muscle
components produce component topographies with focal
activities over specific channels, typically located tempo-
rally (T7 and T8), with a characteristic spectral signature
containing strong spectral power over 20–30 Hz and an
erratic spectrum (Jung et al. 2000a). Alpha occipital com-
ponents were based on the 8–12 Hz frequency peak in the
occipital areas, which was also associated with an addi-
tional peak at 20 Hz. These analyses are illustrated in the
following paragraphs.
The spectrum for the activity of these components was
then computed on the independent components during both
the meditation and the control periods with the same
Thomson multi-taper spectrum estimator used on the
channel data (Thomson 1982). Traditional t-tests were
computed with bootstrap statistics with FDR correction for
multiple comparisons, which were conducted between the
power outcomes in the meditation vs. control states for
each of the three classes of independent components at all
frequencies (including gamma). This approach helped to
assess whether eye or muscle gamma activity was greater
in meditation state and whether the occipital alpha inde-
pendent components replicated the occipital scalp channel
gamma effect.
Results
Participants and self-report scales
The ‘‘depth of meditative state’’ from the (1–10) self-report
scale indicated that the mean meditative depth experienced
during the rest state was 1.7 ± 1.4, and meditative depth
experience during the meditative state was 4.5 ± 1.4 (t-
test, df = 15, P = 0.00004). Drowsiness was reported by
7 of the 16 participants during the meditation and 10 of the
16 during the control thought condition. There was no
reliable correlation between the self-reported depth of
meditative state and either the number of years of daily
practice (r=0.24, P = 0.36) or the number of hours of
daily practice (r=-0.02, P = 0.93). A reliable correla-
tion was obtained between the order of experimental ses-
sion and the self-reported experience of drowsiness during
the control state wherein those individuals meditating first
were less likely to experience drowsiness during the con-
trol state (r=0.52, P = 0.041). No similar correlation was
obtained between session order and drowsiness during
meditation (r = 0.38, P = 0.15) (see also Cahn and Polich
2009).
A negative correlation between the number of years of
daily practice and reported drowsiness during the control
state (r=-0.60, P = 0.015) was found, but not between
number of years of daily practice and reported drowsiness
during meditation (r=-0.006, P = 0.98). These findings
imply that individuals with more years of daily meditation
practices were less likely to report drowsiness specifically
during the control cognitive condition. A negative corre-
lation between current number of hours of daily practice
reported and drowsiness during meditative state also was
obtained (r =-0.59, P = 0.016). No association was
observed for the control state drowsiness and number
of hours of daily practice (r = 0.02, P = 0.93), implying
that greater current number of hours of current daily
practice predicted decreased drowsiness specifically during
meditation.
EEG
Figure 1 illustrates the mean amplitude spectral data
averaged across meditation and control thought states with
grand average scalp maps for each of the major bands
represented for delta (1–4 Hz), theta (4–8 Hz), alpha (8–
12 Hz), beta (12–25 Hz), and gamma (35–45 Hz). Pre-
liminary analyses across all electrodes using bootstrap
statistics and FDR correction for multiple comparisons
indicated significant meditation state effects for delta and
gamma log spectral power, but no such effects for theta,
alpha, and beta power. Scalp maps for the two states are
thus only shown separately for delta and gamma bands, and
42 Cogn Process (2010) 11:39–56
123

Fig. 2 shows the spectral power grand average scalp maps
for meditation and control states separately at both fre-
quencies. Also indicated is a scalp map of the statistical
significance for the comparison between the log spectral
power at each electrode across the two states (all red
electrode locations indicate significance at P \ 0.05 via
bootstrap with FDR correction). Delta decreases were in
bilateral frontal regions, and gamma increases were in
parieto-occipital areas only (see Fig. 2).
Delta (1–4Hz). As mentioned above, bootstrap analysis
with FDR corrections for multiple comparisons analysis of
the sensor data indicated decreased delta power during
meditation relative to control period at bilateral frontal
electrodes (F3, F4, F7, F8, C3, and C4, all P \ 0.05).
Parametric repeated-measures ANOVA analysis of the data
was carried out as well, indicating that delta power at
midline electrodes yielded no state (F(1,15) = 1.84,
P = 0.19), electrode (F(2,30) = 1.0, P = 0.37), or inter-
action (F(2,30) = 0.01, P = 0.99) effects, although
somewhat greater delta power in control than meditation
state was observed (P = 0.19). Frontal delta power at
electrodes F3 and F4 was analyzed separately as delta
activity at these electrodes was shown to be most signifi-
cantly decreased in the bootstrap analysis of the scalp data.
ANOVA analysis of the delta power at F3 and F4 dem-
onstrated an effect for meditation state just missing the
P = 0.05 cut-off (F(1,15) = 4.09, P = 0.06), with no
effect for electrode (P = 0.97), and no interaction between
electrode and state was observed (P = 0.67). Analyses of
FP1, F7, FP2, and F8 produced the same pattern of
findings.
Theta (4–8Hz). Bootstrap analysis with FDR correction
for multiple comparisons indicated no main effect of state
at any electrodes for theta power during meditation relative
to control period. Parametric ANOVA assessment of theta
power also yielded no significant effect of state,
(F(1,15) = 0.037, P = 0.85) or electrode location
(F(2,30) = 0.96, P =
0.39). A significant state x electrode
interaction, F(2,30) = 7.75, P = 0.002, indicated with
post hoc evaluation that theta power exhibited relative
increase during meditation state at Fz (P = 0.006), but not
at Cz (P = 0.99) or Pz (P = 0.85).
Alpha (8–12 Hz). Bootstrap analysis with FDR correc-
tion for multiple comparisons indicated no main effect of
state at any electrodes for alpha power during meditation
relative to control period. Parametric ANOVA assessment
of alpha power at midline electrodes likewise produced no
Fig. 1 The power spectrum for electrode Cz is displayed for both
meditation and control states. Grand average scalp maps across states
are displayed for each of the major frequency bands so as to indicate
scalp topography of the various frequencies
Fig. 2 The scalp maps for delta (1–4 Hz) and gamma (35–45 Hz)
bands are displayed with meditation on the left and control task on the
right. Both bands showed a statistically significant difference
comparing meditation and control states and the scalp map indicating
statistical significance is shown below. Statistical significance was
determined using bootstrap statistics with false detection rate (FDR)
correction for multiple comparison with the threshold significance set
at P \ 0.05, indicating bilateral frontal decreases in delta power
(significant at F3, F4, F7, F8, C3, and C4) and parieto-occipital
increases in gamma power (significant at P7, P8, O1, and O2) as
meditation state effects. Similar analyses run on the theta, alpha, and
beta bands indicated no statistical differences between states at any
scalp site and are thus not shown
Cogn Process (2010) 11:39–56 43
123

meditation state effect (F(1,15) = 0.096, P = 0.76),
although a main effect of electrode was found, F(2,30) =
18.8, P = 0.000005, confirming that alpha was the greatest
at more posterior sites as expected. No interaction between
state and electrode was obtained (F(2,30) = 1.1, P =
0.34). Post hoc testing confirmed that alpha power was
greater at Pz than Fz (P = 0.0001) and Cz (P = 0.004),
and greater at Cz than Fz (P = 0.04).
A separate ANOVA analysis of alpha power at occipital
sites (O1 and O2) was conducted to assess alpha power at
the area of maximal power. No state (F(1,15) = 0.055,
P = 0.82), electrode (F(2,30) = 1.5, P = 0.24), or inter-
action was found (F(2,30) = 0.55, P = 0.47). Additional
separate analyses of low (8–10 Hz) and high (10–12 Hz)
alpha indicated the same pattern of findings.
Beta (12–25 Hz). Bootstrap analysis with FDR correction
for multiple comparisons indicated no main effect of state for
beta power during meditation relative to control period.
Analysis of beta power at the midline electrodes indicated no
state (F(1,15) = 0.62, P = 0.44) or interaction (F(2,30) =
1.4, P = 0.26) effects, but a main effect of electrode was
found, F(2,30) = 4.33, P = 0.02, indicating that beta power
was greater at more posterior sites. Post hoc assessment
found beta power at Pz was greater than at Fz (P = 0.03), but
not significantly different from Cz (P = 0.92); beta power at
Cz was trend-level greater than at Fz (
P = 0.07). Additional
separate analyses of low (12–15 Hz), medium (15–18), and
high (18–25 Hz) beta indicated the same pattern of findings
and are not presented separately.
Gamma (35–45 Hz). Bootstrap analysis with FDR cor-
rection for multiple comparisons indicated a main effect of
state with increased gamma power during meditation rel-
ative to control period at P7 and P8 (both P \ 0.05), as
well as O1 and O2 (both P \ 0.01). Parametric ANOVA
analysis of the gamma power at midline electrodes dem-
onstrated no state (F(1,15) = 1.26, P = 0.28), electrode
(F(2,30) = 2.22, P = 0.14), or interaction (F(2,30) =
1.35, P = 0.27) effects. Additional analysis of gamma
power was conducted specifically at occipital electrodes
(O1, O2) because of the greater gamma power found at this
location in the bootstrap/FDR analysis. Analysis of occip-
ital gamma power at O1 and O2 yielded a significant state
effect, F(1,15) = 9.32, P = 0.008, but no electrode
(F(2,30) = 2.02, P = 0.18) or interaction (F(2,30) = 0.59,
P = 0.45) effects. Additional analyses including P7 and P8
in the ANOVA indicated the same pattern of findings, but
as the effect was the greatest at O1 and O2, only these data
are presented.
Independent component analyses (ICA)
Figures 3, 4, 5 illustrate the grand average scalp maps of
the independent components of the three EOG classes or
clusters: Fig. 3 illustrates those accounting for vertical and
horizontal eye movements. Figure 4 illustrates those
associated with the left and right temporalis muscle
activity. Figure 5 illustrates the occipital alpha cluster of
components. Each subject yielded one to two eye-move-
ment-related components and one to two occipital alpha
components for a total of 23 components for both the alpha
and eye-movement clusters. Additionally, one to six mus-
cle components were identified per subject for a total of 37
components in the muscle cluster.
Bootstrap statistics with FDR correction across fre-
quencies 1–55 indicated that no difference in spectral
frequency power was present between states for either the
eye- or the muscle-independent component clusters (see
Figs.
3 and 4). A statistically significant increase in
gamma ([25 Hz) power was found in the occipital alpha
independent components (see Fig. 5, gray area at bottom
indicates the area of statistical significance comparing
between states). Analysis of the specific 35–45 Hz
Fig. 3 Panel A indicates the grand average scalp map for the left and
right muscle-independent components, respectively. Panel B presents
the grand average power spectra for the muscle components for both
meditation and control states. In panel C, the thin, colored lines
indicate difference in power across the range of frequencies for the
muscle components; meditation minus control period for all subjects
(when more than one component was present for a given subject, the
power spectrum for these components was averaged). The bold, black
trace indicates the grand average spectrum difference for the contrast
meditation minus control such that segments above 0 indicate greater
average component activity during meditation and below 0 indicate
greater average component activity during control period
44 Cogn Process (2010) 11:39–56
123

frequency gamma band activity for each independent
component (IC) class comparing meditation to control
state activity indicated that independent components due
to muscle (t-test, df = 15, P = 0.55) and eye (t-test,
df = 15, P = 0.83) demonstrated no difference between
meditation and control states. Occipital alpha IC gamma
activity was significantly greater in meditation relative to
control state using bootstrap statistic with FDR correction
for multiple comparisons (bootstrap, df = 15, P =
0.0075), but only marginally different between states
using parametric statistics (t-test, df = 15, P = 0.07), in
any case mirroring the significant difference observed in
gamma power at the occipital scalp electrodes. Analysis
was applied to the delta power for the eye-movement ICs
as this activity reflects typical slow eye movements. Less
eye-movement activity was found for the meditation
compared to control state, with a marginal parametric
difference obtained (t(1,15) = 3.05, P = 0.10), and a
significant non-parametric (Wilcoxon sign test) outcome
obtained (P = 0.006).
Additional analyses
Correlations were calculated to explore the relationships
between both the eye-movement-related IC delta power
(eye) and scalp-recorded delta power as well as between
gamma power (muscle, eye) and scalp-recorded gamma
power, so as to assess possible non-cortical sources for the
observed state effects of meditation. Specifically, correla-
tions were computed between the gamma power in the
eye- and muscle-independent components and the scalp-
recorded gamma power across midline as well as occipital
electrodes in each experimental state. To assess the pos-
sibility that slow eye-movement differences between
meditation and control state might contribute to the
occipital gamma findings, correlations between eye-
movement independent component delta power and
gamma power at scalp channels were also computed. As
the increased delta activity observed at fronto-lateral
locations might be related to eye-movement activity not
Fig. 4 Panel A indicates the grand average scalp map for vertical and
horizontal eye-movement independent components, respectively.
Panel B presents the grand average power spectra for the eye-
movement components for both meditation and control states. In
panel C, the thin, colored lines indicate the difference in power across
the range of frequencies for the eye-movement components between
meditation and control period for all subjects. The bold, black trace
indicates the grand average spectrum difference for the contrast
meditation minus control such that segments above 0 indicate greater
average component activity during meditation and below 0 indicate
greater average component activity during control period
Fig. 5 Panel A indicates the grand average scalp map for the
occipital alpha independent components. Panel B presents the grand
average power spectra of the occipital alpha components in medita-
tion and control states. In panel C, the thin, colored lines indicate the
difference in power across the range of frequencies for the occipital
alpha components between meditation and control periods for all
subjects. The bold, black trace indicates the grand average spectrum
difference for the contrast meditation minus control. The gray bar at
the bottom of the figure indicates frequencies over which bootstrap
statistics with FDR correction for multiple comparisons indicated
statistically significant greater values in meditation than control state
(25–55 Hz)
Cogn Process (2010) 11:39–56 45
123

fully removed from the data, additional correlations were
run between delta power in the eye-movement independent
components and delta power in the (post eye-movement
independent component removal) midline and frontal
channel data.
There were no correlations between delta activity in the
eye-movement IC’s and midline or frontolateral electrodes
in either meditation or control states: meditation—Fz, r =
0.044, P = 0.87, Cz, r = 0.00, P = 1.00, Pz, r = 0.16,
P = 0.56, F3, r = 0.22, P = 0.42, F4, r = 0.084,
P = 0.76; control—Fz, r =-0.17, P = 0.52, Cz, r =
-0.087, P = 0.75, at Pz, r =-0.051, P = 0.85, F3,
r = 0.094, P = 0.73, F4, r = 0.016, P = 0.95.
In contrast, significant correlations between gamma
power in the muscle IC cluster and gamma power at
midline electrodes were obtained for both state conditions:
meditation—Fz, r = 0.61, P = 0.013, Cz, r = 0.60, P =
0.014, Pz, r = 0.47, P = 0.066; control—Fz, r = 0.55,
P = 0.028, at Cz, r = 0.46, P =
0.073, at Pz, r = 0.32,
P = 0.22. Significant correlations between gamma power
in the eye-movement IC cluster and gamma power at
midline electrodes were also obtained for the control
condition with trends observed also at Fz in meditation:
meditation—Fz, r = 0.50, P = 0.058, Cz, r = 0.41, P =
0.13, Pz, r = 0.081, P = 0.78; control—Fz, r = 0.82,
P = 0.0001, Cz, r = 0.63, P = 0.012, Pz, r = 0.54,
P = 0.039. Importantly, no correlations were found
between gamma power at occipital electrodes and
gamma power in the muscle IC cluster (meditation—O1,
r = 0.11, P = 0.69, O2, r = 0.17, P = 0.53; control—O1,
r =-0.061, P = 0.82, O2, r =-0.093, P = 0.73) or the
eye-movement IC cluster (meditation—O1, r =-0.04,
P = 0.89, O2, r = 0.014, P = 0.96; control—O1,
r = 0.38, P = 0.17, O2, r = 0.19, P = 0.49). The corre-
lations between the occipital alpha IC gamma activity and
scalp-recorded gamma power were of moderate signifi-
cance at midline electrodes: meditation—Fz, r =
0.42,
P = 0.104; Cz, r = 0.38, P = 0.15; Pz, r = 0.47,
P = 0.064; control—Fz, r = 0.50, P = 0.048; Cz,
r = 0.60, P = 0.013; Pz, r = 0.70, P = 0.003. These
correlations were, however, quite significant at occipital
electrodes: (meditation—O1, r = 0.77, P = 0.0001, O2,
r = 0.71, P = 0.002; control—O1, r = 0.79, P = 0.0001,
O2, r = 0.75, P = 0.001.
Covariate analyses
Additional analyses were conducted using covariates to
characterize individual differences underlying the spectral
power findings. The primary covariates were those relating
to order of experimental conditions, reported drowsiness
during the experimental sessions, and intensity of medita-
tion practice (number of years of daily meditation practice,
current number of hours/day of meditation practice). Sig-
nificant interactions for some of these covariates were
obtained for the delta, alpha, and gamma bands, with no
reliable effects obtained for theta or beta power.
Delta
A significant interaction among state, order of experi-
mental sessions (meditation ? control vs. control ?
meditation), and midline electrode location was found,
F(2,28) = 4.70, P = 0.017). This outcome suggests
decreased midline delta power during meditation relative
to rest specifically for those participants doing the control
period prior to the meditation period, but not those
meditating first. Breakdown of this interaction with Tukey
post hoc testing indicated that when the control period
occurred first, midline delta power was decreased in the
subsequent meditation session at Fz (P = 0.0004) and Cz
(P = 0.027) but not Pz (P = 0.30). No differences among
midline electrodes for delta power were found in partic-
ipants meditating first (Fz, P = 1.0; Cz, P = 0.77; Pz
P = 0.15). A second covariate interaction was found for
the state 9 reported drowsiness 9 midline electrode
location during meditation, F(2,28) = 3.39, P = 0.06),
indicating that only those subjects not reporting drowsi-
ness during meditation showed a tendency for decreased
midline delta power in meditation. No significant out-
comes were found for state effect at lateral-frontal
locations.
Alpha
The interaction among order of experimental session x
state effects x midline electrode location was significant,
F(1,14) = 5.31, P = 0.037. For the occipital electrodes the
order 9 state interaction was reliable, F(1,14) = 4.00,
P = 0.065, with both such interactions indicative that
whichever experimental period occurred second in the
experimental order scheme tended to produce more alpha
power (See Fig. 6a). The interaction among state x medi-
tation daily practice was trend-level significant for the
midline, F(1,14) = 3.89, P = 0.068, as well as occipital
electrodes, F(1,14) = 3.99, P = 0.066. These outcomes
suggested that more years of meditative practice tended to
correlate with slight decreases in alpha power during
meditation, whereas fewer meditation practice years cor-
related with trend increases in alpha power during medi-
tation (See Fig. 6b). The state 9 drowsiness 9 electrode
interaction was marginally significant, F(2,28) = 3.06,
P = 0.071, such that participants reporting drowsiness
during the control period tended to produce decreased
alpha in the control relative to meditation period, with the
opposite pattern obtained for those not reporting such
46 Cogn Process (2010) 11:39–56
123

drowsiness. The same but weaker interaction was observed
for occipital alpha power (F(2,28) = 2.44, P = 0.14).
Gamma
A number of covariates produced significant findings for
both the midline and occipital gamma power (reported
drowsiness, number of hours of current daily meditation
practice). Given the fact that the muscle IC cluster gamma
power correlated with the gamma power at midline elec-
trodes and that the same pattern of significant findings
obtained with midline gamma covariate analysis obtained
with muscle IC gamma covariate analysis, these associations
are likely due to muscle rather than brain reactivity and are
not reported.
No reliable outcomes were obtained when the self-report
score for meditative depth between meditation and control
sessions was used as a covariate for gamma power. Cate-
gorization of participants into those with a history of
10? years (n = 10, M = 19.3, SD = 8.6 years) and those
with \10 years (n = 6, M = 2.5, SD = 1.4 years) of daily
meditation practice yielded significant interactions for
occipital gamma. The long-term meditator category yiel-
ded a significant main effect, F(1,14) = 4.87, P = 0.044,
indicating that the sub-group reporting more years of daily
meditation practice exhibited greater gamma power across
task conditions (See Fig. 7a, b). In addition, an interaction
among state 9 daily meditational practice length 9 elec-
trode was found, F(1,14) = 4.53, P = 0.05. Specifically,
gamma power was increased in meditation vs. control state
in long-term meditators (O1, P = 0.0002; O2, P = 0.008)
but only marginally in relatively short-term meditators (O1,
P = 0.76; O2, P = 0.09). The level of drowsiness 9 state
interaction was also reliable, F(1,14) = 5.41, P = 0.035.
Gamma power increases were significant only for those
participants not reporting drowsiness during rest (Tukey
post hoc comparison P =
0.0069) and were absent for the
group who did report drowsiness (Tukey post hoc com-
parison P = 0.57). In contrast, the self-reported drowsiness
during meditation state did not show a significant interac-
tion with gamma.
Discussion
Meditation and neuroelectric measures
Vipassana meditative practice involves the adoption of a
mindful and receptive mental awareness, with attentional
absorption on present-moment sensations in the body and
meta-cognitive reframing of ongoing experience as
impersonal phenomena to be observed but not reacted to
(Gunaratana 2002; Hart 1987; Lutz et al. 2007). EEG
measures were obtained from experienced Vipassana
meditators with conditions that contrasted the meditative
state with a control cognitive condition designed to mimic
‘‘everyday thinking.’’ The pattern of meditation-induced
increase in parieto-occipital gamma activity, concomitant
decrease in frontal delta power, and a shift to a more frontal
distribution of theta activity suggests that sensory pro-
cessing and cognitive processing were altered during
meditation relative to the control state. However, the typ-
ically reported meditation state changes in the theta and
alpha frequencies were not found (Cahn and Polich 2006).
Delta (1–4 Hz) power is known to correspond with
inhibitory functions (Babiloni et al. 2006; de Jongh et al.
Fig. 6 a Experimental order and alpha power. Participants undergo-
ing the control period followed by the meditation period tended to
evince greater alpha power during meditation, whereas those
undergoing the meditation period followed by the control period
tended to evince greater alpha power during the control period.
b Meditators with less than 5 years of daily meditation practice
(n = 6, M = 2.5, SD = 1.4) vs. meditators with 10 and greater years
of daily meditation (n = 10, M = 19.3, SD = 8.5) practice showed
different tendencies with regard to alpha power across experimental
sessions, with shorter-term practitioners showing a trend toward
greater alpha during meditation and longer-term practitioners showing
the opposite pattern
Cogn Process (2010) 11:39–56 47
123

2001; Niedermeyer and Lopes da Silva 1999; Penolazzi
et al. 2008) and has been reported to be associated with
meditation only recently, with a trait increase in frontal
delta power reported for both Zen and Qi-Gong practitio-
ners (Faber et al. 2008; Tei et al. 2009). Theta (4–8 Hz)
power is known to correspond with meditation, although
the cortical sources for these effects are still not fully
understood. Frontal mid-line theta tends to be associated
with concentrative attentional engagement, whereas less
specific topographic theta distribution increases are
observed in periods of drowsiness (Basar et al. 2001a, b;
Mitchell et al. 2008). Alpha (8–12 Hz) power corresponds
with cortical idling, cortical suppression, and relative
deactivation of underlying brain areas (Niedermeyer 1997).
Gamma (35–45 Hz) power is known to correspond with
stimulus representation and feature binding, possibly
coupling tightly with perceptual awareness (Fries et al.
2001, 2008) and selective attention (Fell et al. 2003). The
relative impact of meditation on different frequencies of
brain activity are still not well understood across medita-
tive practices and is likely both practice-specific and rela-
ted to differential effects early vs. late in the learning
process.
Important to the interpretation of the present relative to
past findings is that previous assessments often have not
obtained neuroelectric measures during meditation vs.
cognitive control periods of equal length. Furthermore, a
‘‘resting’’ state is not likely the same for meditators com-
pared to non-meditators, as meditators often report an
inability for non-meditative resting. Indeed, in the present
study, a number of participants reported difficulty in
avoiding engagement in meditative practice with eyes-
closed and the posture used during meditation even with
the instructions to keep the mind engaged with neutral
memories during the control period. Nonetheless, although
a few participants reported approximately the same medi-
tation depth in both periods, the consistent rating of a
greater meditative depth in the meditation period than the
control period likely reflects a different subjective experi-
ence of the two states. Further, both the neuroelectric
measures and the introspective meditative depth differen-
tiated between meditation and the control condition.
Delta effects
Previous assessments of meditation have not often reported
effects on the delta frequency band, but it is unclear
whether it has been systematically analyzed. Recently, two
reports of increased trait frontal delta in long-term Zen and
Qi-Gong meditators suggest that there may be an important
interaction between meditative practice and delta brain
activity (Faber et al. 2008; Tei et al. 2009). Increased
frontal delta in long-term meditators may indicate a func-
tional inhibition of brain appraisal systems in line with a
detachment from analysis, judgment, and expectation (Tei
et al. 2009). In this study, decreased delta activity in the
temporo-parietal junction, secondary motor cortex, and
sensory association cortices appeared indicative of
increased brain activation associated with detection and
integration of internal and external sensory information,
with detachment from the same as indexed by inhibited
activity (increased delta) in prefrontal areas responsible for
analyzing, judging, and expectation (Tei et al. 2009).
The present study found a significant state effect in the
delta frequency band, such that the meditation state was
characterized by a decrease in bilateral frontal delta power,
indicative of an increase in frontal activation during Vip-
assana meditation relative to the control condition. It is
worth noting that this frontal delta decrease was significant
Fig. 7 a Meditators with 10 and greater years of daily meditation
practice in solid circles (n = 10, M = 19.3, SD = 8.5), Meditators
with less than 5 years of daily meditation practice in solid circles
(n = 6, M = 2.5, SD = 1.4), smaller ‘‘?’’ indicates the mean
average occipital gamma in control vs. meditation conditions in
shorter-term meditators, larger ‘‘?’’ indicates the mean average
occipital gamma in control vs. meditation conditions in longer-term
meditators. b Interaction between meditation expertise and occipital
gamma effect
48 Cogn Process (2010) 11:39–56
123

in the bootstrap with FDR correction for multiple com-
parison statistical analysis of the scalp channel data at the
frontal electrodes, but just missed statistical significance by
standard parametric ANOVA testing (P = 0.06). It has
been argued that bootstrap statistics may be of greater
theoretical utility in application to relatively non-Gaussian
measures such as EEG spectral power values (Darvas et al.
2004; Delorme 2006), suggesting that the significant find-
ing here with bootstrap statistics is quite valid, but it should
be noted that the parametric testing result was trend-
level. It is important to note that the delta power in the
eye-movement-related independent components during
meditation relative to control states was also decreased
reflecting less eye movement during meditation. However,
the eye-movement-related delta power did not correspond
with the scalp-recorded delta power, indicating that the
eye-movement artifact rejection was effective and that the
decrease in delta power measured at the frontal scalp
electrodes was separate from meditation’s inhibitory effect
on eye movements. One significant covariate in the anal-
ysis of midline delta activity was self-reported drowsiness
during meditation, and this covariate showed a trend sig-
nificance (P = 0.09) when used in the ANOVA analysis of
lateral frontal delta as well. Those participants reporting
drowsiness during meditation did not appreciably contrib-
ute to the decrease in frontal delta power seen during
meditation, further suggesting that this delta power
decrease is a marker of the more highly aware state seen in
meditation relative to control state.
It is possible that the delta power decrease we report
here as a state effect may be correlated with the increase in
baseline delta as a trait effect in previous reports, consistent
with the notion that through sustained engagement of
frontal circuits during meditative practice, practitioners
may train other baseline frontal circuits associated with
elaborative processing such as judging, analyzing, and
expecting downward. These hypotheses are suggestive but
provide important footholds for theoretical development of
the relationship between meditation and EEG measures.
Theta effects
No absolute difference in theta power between the medita-
tion and control states was observed. A significant interac-
tion between state and electrode location was found that
suggested a more frontal distribution of theta activity was
present during the meditation state. The implications of this
outcome are uncertain but likely reflect the operation of
enhanced attentional mechanisms mediated through anterior
cingulate engagement (Cahn and Polich 2006). It is of note
that a number of recent studies have found strong increases
in frontal theta power during concentrative/focused atten-
tion meditation states (Aftanas and Golocheikine 2001,
2003; Baijal and Srinivasan 2009) and that frontal theta
power is thus a key contributor to meditative neural
dynamics that likely shows differential engagement depen-
dent on the specific technique employed.
Alpha effects
Meditation and alpha power effects in the long-term Vip-
assana practitioners were absent comparing meditation and
control states, which supports the assertion that alpha
increases often reported in early studies of meditation were
related to assessing beginning meditators vs. experts.
While meditation state did not affect alpha power the order
with which participants engaged in the meditation vs.
control tasks was significant—i.e., whichever task was
engaged later tended to have greater associated alpha
power (see Fig. 6a). This interaction between order and
alpha power was significant at midline electrodes and
trend-level at occipital electrodes. A number of early
studies on meditation utilizing a fixed-order design for the
engagement in control task (often the non-specific
instruction to ‘‘rest’’) followed by meditation may have
shown increases in alpha activity actually related to the
passage of time within the study rather than anything
specific regarding meditation. Nonetheless, given the large
number of studies reporting alpha state and trait effects in
the EEG literature, it is also possible that some forms of
meditative practice may be more reliably associated with
alpha state effects, and/or that there are alpha state effects
toward the beginning of regular practice that dissipate with
the development of expertise.
Supporting this latter interpretation, occipital alpha
power in this study was somewhat related to meditation
expertise, as participants with 10? years daily practice
(n = 10, M = 19.3) tended to demonstrate more similar
alpha power levels across states, whereas those subjects
meditating \10 years of daily practice (n = 6, M = 2.5)
tended to demonstrate enhanced alpha power during med-
itation (see Fig.
6b). Longitudinal studies assessing the
impact of meditation over time in large samples are nec-
essary to substantiate the hypothesis of increased state and
trait alpha power resulting from meditative practices at
different points in the learning process. The current find-
ings support that meditative practice may enhance alpha
power in the beginning stages of learning and that with
expertise a trait-level increased alpha power may develop
after which meditation state effect is no longer character-
ized by alpha enhancement (Cahn and Polich 2006). Direct
support for this hypothesis also would require demon-
strating higher trait alpha in the longer-term meditator
participants, which was not observed in the present sample.
An association between intensity of meditative practice and
alpha power was obtained; however, as participants with
Cogn Process (2010) 11:39–56 49
123

?2 h daily practice demonstrated higher baseline alpha
across states for the occipital alpha independent compo-
nents. Further studies with greater numbers and multiple
meditator cohort groups (multiple techniques and experi-
ence levels) are clearly needed to substantiate the alpha
trait and state hypotheses.
Gamma effects
Gamma rhythms (30 Hz and above, often reported as
centered around 40 Hz (Basar-Eroglu et al. 1996)) have
been widely characterized as significantly related to
momentary contents of consciousness (Sauve 1999). Fur-
ther, electrical activity in this high frequency range has
been shown to be a possible candidate for the neurophys-
iologic substrate of the ‘‘binding’’ of multiple aspects of
conscious experience and perceptions into the coherent
subjective state of moment-to-moment awareness (Singer
1993; Varela 1995). Evidence has accumulated supporting
the notion that enhanced gamma synchrony and/or power
in appropriate cortical areas is critically associated with
both perceptual events (Gross et al. 2007; Lachaux et al.
2005; Meador et al. 2002; Rodriguez et al. 1999) and
readiness to perceive periliminal and/or ambiguous stimuli
(Melloni et al. 2007). While we hypothesized finding a
meditation-related gamma power increase in frontal and/or
parietal areas, reflecting increased functional processing in
frontal and/or somatosensory cortices related to body sen-
sation and/or executive control, instead we found increases
in occipital areas. The increase in occipital gamma syn-
chronization found in our current sample may indicate
that this open-awareness meditative state involves a more
sensitive and perceptually clear awareness of moment-
to-moment experience. The reasons for an occipital dis-
tribution are certainly not clear but may be specific to the
Vipassana meditative technique as previous reports have
not found such gamma increases associated with other
meditative techniques. Further, the fact that both baseline
and meditation-related increases in occipital gamma power
and were found to significantly covary with meditational
expertise as indexed by total years of daily practice (see
Fig. 7) suggests that there may be both a state and trait
effect of increased gamma power associated with Vipas-
sana practice. Clearly additional studies employing a non-
meditator control group are needed to further substantiate
this possible trait effect finding.
Of possible relevance to the parieto-occipital increase in
gamma power with respect to Vipassana meditation state, a
recent report indicated increased gamma power in this
approximate distribution during imagined actions (de
Lange et al. 2008). Although Vipassana meditation practice
does not invoke imagined actions, it does involve the
repetitive scanning of body sensations from head to toes,
which may recruit some surreptitious access to either
imagined body action and/or visualization of body parts as
the scanning occurs. Participants were not specifically
asked to report on their visual experiences during the
meditation and control sessions, but no one indicated a
strong visual component on their free-form written sum-
maries of the meditation state. This distribution of gamma
power is therefore not readily explained via known prop-
erties of Vipassana practice, and it may be related to
properties of widespread posterior gamma increases not
currently understood. Finally, it is also noteworthy that
these increases did not correlate with reported depth of
meditation but did correlate with increased number of years
of daily meditation practice and likely meditative expertise.
Early gamma reports in meditation. Previous findings of
widespread gamma increases in meditation are mostly
limited to early studies prior to the development of
sophisticated computerized methods for the quantitative
EEG analysis and the separation of artifact from cortical
signals (Anand et al. 1961b; Banquet 1973; Das and Gas-
taut 1957), although a few more recent meditation gamma
findings have been reported as well (Aftanas and Golo-
sheykin 2005; Lehmann et al. 2001; Lutz et al. 2004). Das
and Gastaut first reported widespread increased high fre-
quency (20–40 Hz) activity in association with meditation,
reporting that after a long period of meditation some of the
more advanced Yogis studied exhibited increased gamma
states associated with periods of subjective deep medita-
tion/samadhi (Das and Gastaut 1957). Anand et al. (Anand
et al. 1961b) reported that ‘‘fast waves’’ were observed in
the EEG recordings from a Yogi meditating in a box for a
period of 2–3 days, but a separate comprehensive report on
the EEG records from this case study participant and others
with similar expertise did not mention this finding, instead
noting the pronounced lack of alpha blocking exhibited
while these participants were in meditation (Anand et al.
1961a). Banquet reported increased 20 and 40 Hz activity
in a subset of TM practitioners who reported experiencing
a deep meditative/transcendent subjective state during the
EEG recording (Banquet 1973), replicating Das and Gas-
taut’s assertion that deep transcendent states of meditative
consciousness may be marked by increases in gamma
activity.
Recent reports of gamma and meditation. In consider-
ation of recent findings relating gamma to conscious
experience and the early suggestive gamma findings sum-
marized earlier, Ott (2001) specifically hypothesized
gamma activations in deep meditation, possibly correlated
with intensive wakefulness and all-encompassing unity. No
such meditation state effects of gamma power were found,
however, and instead the increases in gamma observed in
some subjects were only those associated with movement
artifacts. However, this study used only one electrode (Cz)
50 Cogn Process (2010) 11:39–56
123

so that this study can not be taken as a comprehensive
gamma assessment. A case study of an advanced Tibetan
Buddhist meditation teacher/practitioner indicated that
gamma (35–44 Hz) was the most reliable frequency band
distinguishing between different meditative states in this
single individual (Lehmann et al. 2001). They specifically
reported that relative occipital increases in gamma were
observed in a meditation focused on visualization of the
Buddha relative to other meditative states not incorporating
visualization. Another recent study of meditation and
gamma activity indicated that long-term meditators relative
to controls exhibited decreased negative emotional stimu-
lus-induced gamma power activation, likely related to
decreased emotional reactivity due to such practice (Afta-
nas and Golosheykin 2005).
The one recent previous report of widespread increases
in gamma power assessed expert Tibetan Buddhist medi-
tators engaged in a loving-kindness/compassion meditative
practice and found both trait and state associations between
meditation and gamma activity (Lutz et al. 2004). Midline
frontoparietal gamma power was higher at baseline (trait)
in advanced Tibetan Buddhist practitioners, and upon
engaging in compassion meditation gamma power
increased in magnitude (state) to a very significant degree.
The topography of the meditation state effect was located
bilaterally over the parieto-temporal and mid-frontal elec-
trodes. The outcomes suggested that increased gamma may
be related to a change in the quality of moment-to-moment
awareness, as claimed by the meditation practitioners.
Further, a reliable association between the estimated total
hours in meditative practice and the baseline gamma power
was found that implied attention and affective processes
are flexible skills which can be trained and that gamma
activation may be a marker for these processes.
The gamma state effect of meditation in the present
study is similar but refers to a different meditative tech-
nique with a different topography obtained, although it is
of note that we employed a mastoid reference, whereas the
previous study employed average reference. A striking
similarity in the pattern of results obtained across the
studies is seen in the fact that the previous study found
higher baseline gamma power in the expert meditator
group than the non-meditator group as well as meditation-
induced increases in gamma power; similarly, the medita-
tion expertise covariate analysis of the present gamma
findings indicated greater baseline occipital gamma power
as well as meditation state-induced increases in this activity
for the more experienced participants. The previous
study employed Tibetan Buddhist/Mahayana meditators
practicing a non-referential compassion technique, whereas
long-term practitioners of Vipassana as taught within
the Theravadan/Hinayana tradition were assessed in the
present study. The presently-reported Vipassana technique
involves attentional absorption in moment-to-moment
subtle sensations of the body concomitant with an atten-
tional stance of non-reactive mindful awareness/open
monitoring.
A significant commonality across the states assessed by
these two studies is the specific inclusion of a mindful/
open-monitoring component to the practice. Lutz et al.
(2004) asserted that the assessed state was an objectless
state of mind, involving a dissipation of the object-oriented
aspect of experience. Vipassana practitioners report that
they are able to adopt a wide-open awareness during
practice that is characterized by a subtle yet rich somato-
sensory experience (Gunaratana 2002), but whether that
experience serves as an ‘‘object’’ of attention is likely
experienced differently across such meditators. It may be
that the widening of the attentional spotlight involved in
these meditational practices correlates with the finding of a
gamma state effect rather than an effect on the slower theta
or alpha frequencies more commonly reported in the past.
Gamma effects and artifact
It is important to note the well-known artifact from muscles
of the scalp, head, and neck that can generate high frequency
electrical in the gamma range and to address the possibility
that the gamma increases we recorded at the scalp may be
muscle related. In addition, significant attention has now
been drawn to the fact that microsaccades are significantly
associated with increased gamma power (Yuval-Greenberg
and Deouell 2009). The multiple analyses conducted uti-
lizing ICA methods to isolate non-brain-related activities
such as eye and muscle artifact from the scalp-recorded
data were performed to counter this possibility (Jung et al.
2000a). Given that gamma activity artifact from eye
movements have been correlated with saccades (Reva and
Aftanas 2004; Yuval-Greenberg and Deouell 2009) and
microsaccades (Dimigen et al. 2009; Yuval-Greenberg et al.
2008) and that the present subjects were recorded in an eye-
closed state, the probability of eye movements contributing
to the gamma finding seems unlikely.
Nonetheless, this possibility was quantitatively assessed
by analyzing the activity of the artifactual independent
component clusters that account for the eye-movement
activities across the two states. Figure 4 illustrates the
results, which indicate that there is no difference in the
activities of these independent component clusters between
states at frequencies above *6 Hz including the gamma
band. At frequencies below 6 Hz, a small decrease in eye
movements in the meditation state relative to control was
observed. This outcome implies a tendency toward a
decrease in the slow eye movements often observed in
eyes-closed conditions during meditation relative to the
control task. In sum, eye movements are a very unlikely
Cogn Process (2010) 11:39–56 51
123

cause for the measured increased gamma power during
meditation.
With respect to the more plausible possibility that scalp
muscle activity contributed to the increase in parieto-
occipital gamma observed, we note first that visual
inspection for artifact tended to show decreased muscle-
and movement-related artifact in meditation relative to
control task. As is routine, increases in phasic muscle
tension and movement were observed in a small number of
epochs during both states, but more total epochs were
removed from the control state data than from the medi-
tation state data at the level of visual epoch-rejection pre-
processing (remaining epochs analyzed out of 180 were
170.4 ± 13.8 for the meditation state and 164.1 ± 26.6 for
the control state). Muscle-related independent components
were empirically assessed as an objective check of the
possible contribution of muscle activity to the observed
effects. Independent components that resembled the well-
known characteristics of muscle activity in terms of focal
scalp distribution (often centered at temporal electrodes
overlying the temporalis muscle) and featuring relatively
high-amplitude high frequency signals were assessed
(Fig. 3). No difference in the gamma (or any other fre-
quency) activity of the muscle components was found
between meditation and control state. Correlation analyses
also were performed on the gamma activity in the muscle-
independent components in meditation and rest vs. the
scalp-recorded gamma power data. These analyses indi-
cated significant correlation between the rest and medita-
tion muscle IC gamma activations and the gamma activity
recorded at midline electrodes, but no correlation with the
gamma activity recorded at parieto-occipital electrodes
where the significant increase gamma power during med-
itation was observed.
A last line of evidence further bolstering the claim that
the observed gamma effect is cortical in origin is that the
occipital alpha independent components also exhibited a
gamma state effect. The increase in gamma power of the
occipital alpha reflected by the independent components
during meditation relative to control state suggest that
significant gamma activity may be related to the common
cortical source shared by each band. The likelihood of
muscle-related gamma co-localizing with cortical-gener-
ated alpha activity after ICA decomposition is not signifi-
cant given that the algorithm is a category of blind source
separation known to segregate time series signals accord-
ing to different spatial and causal generators (Comon 1994;
Delorme et al. 2007; Hyva
¨
rinen and Oja 2000). In sum-
mary, we found that the gamma power increases during
meditation in our meditator participants were not correlated
with increases in scalp muscle or eye-movement-related
activity as assessed by independent component analysis
and that instead the occipital alpha independent
components demonstrated a meditation state effect in the
gamma band. Our report is the first to our knowledge that
uses these advanced signal-processing techniques to clearly
demonstrate that the occipital gamma during meditation
effect we observed is not artifactual.
Choice of control task
It is possible that the sort of ‘‘instructed’’ mind-wandering
state experienced by the present meditators was not con-
sistent with the off-task mind-wandering assayed in the
literature on mind-wandering to date (Smallwood et al.
2008, 2007; Smallwood and Schooler 2006). Continued
assessment of the state effects of meditation require careful
consideration of the control task to employ as well as the
psychometric tests to use to assess the actual experiences
encountered in the control vs. meditation task. With respect
to the various ‘‘control tasks’’ that are used to assess
meditation effects, the present introspective data regarding
drowsiness in relation to meditative expertise may be
instructive. Participants with more years of daily medita-
tion practices reported less drowsiness specifically during
the control cognitive condition, and not the meditation
condition, whereas participants with greater current num-
ber of hours of daily practice were specifically less likely to
report drowsiness during meditation but not necessarily
control conditions. This outcome may imply that the long-
term practice of meditation increases the tendency to
maintain alertness even during boring tasks (e.g., instructed
mind-wandering), whereas the intensity of current practice
is more strongly associated with maintaining alertness
during meditation itself. Whatever the choice of control
task, one of the current challenges in meditation research is
to more fully explore the psychometric characterization of
the control state, whether that be the no-instructions
‘‘resting state’’ often assessed or a more controlled cogni-
tive engagement state such as mental calculations or the
instructed mind-wandering assayed here. This is of special
relevance also to the notion that meditative practices
change the ongoing experience of the world in a way that
may significantly affect the neural ‘‘default network’’
activity mirroring the reported decreases in elaborative
and ruminative processing resulting from such practices
(Pagnoni et al. 2008; Tei et al. 2009).
One limitation of the present study is the lack of a
control group of non-meditators, such that possible EEG
trait measures were not directly assessed, although the
significant covariate indicating that greater length of daily
meditation practice was associated with the increased
gamma meditation state effect is suggestive. Although it is
possible that non-meditators might have shown a similar
‘‘meditation’’ state effect reflected by some aspects of
demand characteristics for the two cognitive tasks assayed,
52 Cogn Process (2010) 11:39–56
123

the significant covariate relating meditational expertise to
increased gamma makes this possibility less likely. The
motivation for staying alert may have varied across the two
experimental periods due to some sort of ‘‘performance’’
pressure in the meditation period, thereby leading to higher
levels of arousal and possibly confounding the results. This
outcome also is unlikely given that analysis of the spon-
taneous EEG data indicated no changes in power for theta
and alpha frequencies between the two states, and
increased arousal is generally associated with higher P3
amplitudes, whereas the meditation effects demonstrated
here include decreased P3a amplitude (Cahn and Polich
2009; Polich 2007; Polich and Kok 1995).
Theoretical perspectives
Previous findings with this meditator cohort demonstrated
decreased engagement of the frontal attentional systems of
the brain to auditory distracters during meditation relative
to the control period as indexed by decreased frontal N1
and P3a event-related potential (ERP) component ampli-
tude to distracter and decreased P2 amplitudes to oddball
stimuli. These findings occurred concomitantly with a
marginal increase in N1 amplitudes to the standards and
oddballs, implying that the Vipassana meditation state is
associated with intact/enhanced sensory processing toge-
ther with top–down control of elaborative attentional
engagement with the contents of awareness (Cahn and
Polich 2009). This outcome is consistent with previous
reports of early studies suggesting that meditation may
produce a state of brain processing less susceptible to
stimulus-driven processing as indexed by alpha blocking
(Anand et al. 1961a; Kinoshita 1975; Lehrer et al. 1980). It
is also consonant with a number of recent reports noting the
enhancement of neural signatures of attentional stability
and efficiency due to meditation interventions (Lutz et al.
2009; Slagter et al. 2007, 2009).
The present finding of increased parieto-occipital
gamma activity is similar to our previous report inasmuch
as gamma activity is a marker for enhanced sensory
awareness. The concomitant increased relative frontal theta
power and decreased frontal delta power during meditation
further support that this form of meditation involves
increased baseline frontal activity with top–down control
over frontal attentional capture due to environmental input
and concomitantly enhanced sensory perceptivity. This
view is in contrast to early definitions of meditation as a
form of relaxation or sleep-like state, although it is
important to note that the variety of very different medi-
tative practices do include those with a greater similarity to
sleep states (e.g., Yoga Nidra), which may be marked by
opposite or markedly different findings from those reported
here (Cahn and Polich 2006).
The current findings emphasize that in highly practiced
Vipassana meditation practitioners, the primary effects of
meditation state on brain rhythms are centered in the low
(delta) and high (gamma) frequency ranges, with moderate
relative increase in frontal theta, and gamma effects most
pronounced in more advanced practitioners. Given the
well-known association of increased slow delta activity
during deep sleep and the more recently described decrease
in gamma power during sleep (Cantero and Atienza 2005;
Cantero et al. 2004; Maloney et al. 1997), the overall
picture can be interpreted as a state of enhanced ‘‘awake-
ness.’’ Further, alpha power does not reliably differentiate
between meditation and control state in advanced Vipas-
sana practitioners but instead tends to vary inversely with
drowsiness. Mindfulness/open-monitoring practices that
involve the widening of the attentional spotlight to present-
moment sensory experiences may be characterized by a
mode of frontal engagement mediating enhanced stimulus
representation and clarity of awareness (spontaneous
gamma, evoked N1) and concomitant decreased cognitive
elaboration upon stimuli in the environment (evoked P2,
P3a). In contrast, practices involving the narrowing of the
attentional spotlight such as mantra and breath-focused
attention practices may likely be characterized by greater
frontal midline theta engagement and less enhanced mea-
sures of stimulus representation, a hypothesis requiring
further specific study of these two forms of practice using
the same protocol across groups.
Conclusions
Appreciation for the variety of mental practices subsumed
by the name ‘‘meditation’’ has recently become a salient
research topic, as observation of the various types of
attentional engagement across meditative practices may
promote different neurophysiologic outcomes (Cahn and
Polich 2006; Depraz et al. 2003; Dunn et al. 1999; Leh-
mann et al. 2001; Lutz et al. 2004, 2008). Assessment of
this group of Vipassana meditators has previously dem-
onstrated decreased frontal engagement in the processing
of unexpected and aversive stimuli during meditation in the
setting of a trend toward enhanced sensory representation
of the standard and oddball stimuli as indexed by increased
N1 amplitudes. The present findings are that of enhanced
frontal engagement as indexed by decreased frontal delta
power and relative increase in the frontal component of
theta activity and a broad increase in parieto-occipital
gamma activity during the meditative state prior to the
onset of the tones used to elicit the findings in our earlier
study. No other frequency bands reliably distinguished the
two states.
It is theoretically plausible that the enhanced gamma
activity observed in this dataset is related to the iterative
Cogn Process (2010) 11:39–56 53
123

body scanning aspect of the technique. Alternatively, this
widespread posterior increase in gamma power may be
more generally related to the enhanced perceptual clarity
often reported in open-monitoring meditative states. Fur-
ther studies contrasting this meditative practice with
focused attention practices such as those involving breath
awareness, mantra, and/or visualization would provide
significant insight into the specificity of the delta, theta and
gamma effects seen in these practitioners. An initial
hypothesis would be that the focused attention practices
might engage the frontal theta activity to a greater degree
than open-monitoring practices such as Vipassana and less
engagement of the gamma activity seen here. An alterna-
tive would be to observe more localized gamma increases
to the cortical areas representing the object of attention. Of
importance also for the development of the field is the
relation between these meditation state changes in the brain
and experiential qualities that must be assessed with
improved psychometric analyses.
Acknowledgments This work was supported by NIH grants
DA018262 and AA006420, The Fetzer Institute, the NIH Medical
Scientist Training Grant T32 GM07198, and the NIH Postdoctoral
grant T32 MH019934 grant in part supported BRC, who is also
affiliated with the Laboratory for Psychopharmacology and Brain
Imaging, University of Zurich Hospital of Psychiatry. The help and
guidance of Drs. Mark Geyer and Franz Vollenweider are gratefully
acknowledged. We thank the meditator participants and Mr. John
Beary of Vipassana Research International for assistance in recruiting
meditation participants. This paper is xxxxxx from The Scripps
Research Institute.
Open Access This article is distributed under the terms of the
Creative Commons Attribution Noncommercial License which per-
mits any noncommercial use, distribution, and reproduction in any
medium, provided the original author(s) and source are credited.
References
Aftanas LI, Golocheikine SA (2001) Human anterior and frontal
midline theta and lower alpha reflect emotionally positive state
and internalized attention: high-resolution EEG investigation of
meditation. Neurosci Lett 310:57–60
Aftanas LI, Golocheikine SA (2003) Changes in cortical activity in
altered states of consciousness: the study of meditation by high-
resolution EEG. Hum Physiol 29:143–151
Aftanas L, Golosheykin S (2005) Impact of regular meditation
practice on EEG activity at rest and during evoked negative
emotions. Int J Neurosci 115:893–909
Anand BK, Chhina GS, Singh B (1961a) Some aspects of electro-
encephalographic studies in yogis. Electroencephal Clin Neuro-
physiol 13:452–456
Anand BK, Chhina GS, Singh B (1961b) Studies on Shri Ramanand
yogi during his stay in an air-tight box. Indian J Med Res 49:82–89
Babiloni C, Frisoni G, Steriade M, Bresciani L, Binetti G, Del Percio
C, Geroldi C, Miniussi C, Nobili F, Rodriguez G, Zappasodi F,
Carfagna T, Rossini PM (2006) Frontal white matter volume and
delta EEG sources negatively correlate in awake subjects with
mild cognitive impairment and Alzheimer’s disease. Clin
Neurophysiol 117:1113–1129
Baijal S, Srinivasan N (2009) Theta activity and meditative states:
spectral changes during concentrative meditation. Cogn Process
Banquet JP (1973) Spectral analysis of the EEG in meditation.
Electroencephalogr Clin Neurophysiol 35:143–151
Basar E, Basar-Eroglu C, Karakas S, Schurmann M (2001a) Gamma,
alpha, delta, and theta oscillations govern cognitive processes.
Int J Psychophysiol 39:241–248
Basar E, Schurmann M, Sakowitz O (2001b) The selectively
distributed theta system: functions. Int J Psychophysiol
39:197–212
Basar-Eroglu C, Struber D, Schurmann M, Stadler M, Basar E (1996)
Gamma-band responses in the brain: a short review of psycho-
physiological correlates and functional significance. Int J
Psychophysiol 24:101–112
Benjamini Y, Yekutieli D (2001) The control of the false discovery
rate in multiple testing under dependency. Ann Stat 29:1165–
1188
Cahn BR, Polich J (2006) Meditation states and traits: EEG, ERP, and
neuroimaging studies. Psychol Bull 132:180–211
Cahn BR, Polich J (2009) Meditation (Vipassana) and the P3a event-
related brain potential. Int J Psychophysiol 72:51–60
Cantero JL, Atienza M (2005) The role of neural synchronization in
the emergence of cognition across the wake-sleep cycle. Rev
Neurosci 16:69–83
Cantero JL, Atienza M, Madsen JR, Stickgold R (2004) Gamma EEG
dynamics in neocortex and hippocampus during human wake-
fulness and sleep. Neuroimage 22:1271–1280
Christoff K, Gordon AM, Smallwood J, Smith R, Schooler JW (2009)
Experience sampling during fMRI reveals default network and
executive system contributions to mind wandering. Proc Natl
Acad Sci USA 106:8719–8724
Comon P (1994) Independent component analysis: a new concept?
Signal Process 36:287–314
Darvas F, Pantazis D, Kucukaltun-Yildirim E, Leahy RM (2004)
Mapping human brain function with MEG and EEG: methods
and validation. Neuroimage 23(1):S289–S299
Das NN, Gastaut H (1957) Variations de l’activite electrique du
cerveau, du coeur et des muscles squelettiques au cours de la
meditation et de l’extase yogique. Electroencephalogr Clin
Neurophysiol 6:211–219
Davidson RJ (2003) Affective neuroscience and psychophysiology:
toward a synthesis. Psychophysiology 40:655–665
de Jongh A, de Munck JC, Baayen JC, Jonkman EJ, Heethaar RM,
van Dijk BW (2001) The localization of spontaneous brain
activity: first results in patients with cerebral tumors. Clin
Neurophysiol 112:378–385
de Lange FP, Jensen O, Bauer M, Toni I (2008) Interactions between
posterior gamma and frontal alpha/beta oscillations during
imagined actions. Front Hum Neurosci 2:7
Delorme A (2006) Statistical methods. In: Webster JG (ed) Encyclo-
pedia of medical device and instrumentation, vol 6. Wiley
Interscience, Hoboken, NJ, USA, pp 240–264
Delorme A, Makeig S (2004) EEGLAB: an open source toolbox for
analysis of single-trial EEG dynamics including independent
component analysis. J Neurosci Methods 134:9–21
Delorme A, Sejnowski T, Makeig S (2007) Enhanced detection of
artifacts in EEG data using higher-order statistics and indepen-
dent component analysis. Neuroimage 34:1443–1449
Depraz N, Varela JF, Vermersch P (2003) On becoming aware: a
pragmatics of experiencing. John Benjamins Publishing Com-
pany, Amsterdam
Dimigen O, Valsecchi M, Sommer W, Kliegl R (2009) Human
microsaccade-related visual brain responses. J Neurosci
29:12321–12331
54 Cogn Process (2010) 11:39–56
123

Dunn BR, Hartigan JA, Mikulas WL (1999) Concentration and
mindfulness meditations: unique forms of consciousness? Appl
Psychophysiol Biofeedback 24:147–165
Faber PL, Steiner ME, Lehmann D, Pascual-Marqui RD, Jancke L,
Esslen M, Gianotti LRR (2008) Deactivation of the medial
prefrontal cortex in experienced Zen meditators (Abstract). Brain
Topogr 20:172
Fell J, Fernandez G, Klaver P, Elger CE, Fries P (2003) Is
synchronized neuronal gamma activity relevant for selective
attention? Brain Res Brain Res Rev 42:265–272
Fries P, Neuenschwander S, Engel AK, Goebel R, Singer W (2001)
Rapid feature selective neuronal synchronization through corre-
lated latency shifting. Nat Neurosci 4:194–200
Fries P, Womelsdorf T, Oostenveld R, Desimone R (2008) The
effects of visual stimulation and selective visual attention
on rhythmic neuronal synchronization in macaque area V4.
J Neurosci 28:4823–4835
Gaylord C, Orme-Johnson D, Travis F (1989) The effects of the
transcendental meditation technique and progressive muscle
relaxation on EEG coherence, stress reactivity, and mental health
in black adults. Int J Neurosci 46:77–86
Gross J, Schnitzler A, Timmermann L, Ploner M (2007) Gamma
oscillations in human primary somatosensory cortex reflect pain
perception. PLoS Biol 5:e133
Grossman P, Niemann L, Schmidt S, Walach H (2004) Mindfulness-
based stress reduction and health benefits. A meta-analysis.
J Psychosom Res 57:35–43
Gunaratana H (2002) Mindfulness in plain English. Wisdom Publi-
cations, Boston, MA
Hart W (1987) The art of living: Vipassana meditation as taught by
Goenka SN New York: HarperOne
Hebert R, Lehmann D (1977) Theta bursts: an EEG pattern in normal
subjects practising the transcendental meditation technique.
Electroencephalogr Clin Neurophysiol 42:397–405
Hyva
¨
rinen A, Oja E (2000) Independent component analysis:
algorithms and application. Neural Netw 13:411–430
Jung TP, Makeig S, Humphries C, Lee TW, McKeown MJ, Iragui V,
Sejnowski TJ (2000a) Removing electroencephalographic arti-
facts by blind source separation. Psychophysiology 37:163–178
Jung TP, Makeig S, Westerfield M, Townsend J, Courchesne E,
Sejnowski TJ (2000b) Removal of eye activity artifacts from
visual event-related potentials in normal and clinical subjects.
Clin Neurophysiol 111:1745–1758
Kabat-Zinn J (1982) An outpatient program in behavioral medicine
for chronic pain patients based on the practice of mindfulness
meditation: theoretical considerations and preliminary results.
Gen Hosp Psychiatry 4:33–47
Kabat-Zinn J (2003) Mindfulness-based interventions in context: past,
present, and future. Clin Psychol Sci Pract 10:144–158
Kasamatsu A, Hirai T (1966) An electroencephalographic study on
the Zen meditation (Zazen). Folia Psychiatr Neurol Jpn 20:315–
336
Kasamatsu A, Okuma T, Takenaka S, Koga E, Ikada K, Sugiyama H
(1957) The EEG of ‘Zen’ and ‘Yoga’ practitioners. Electroen-
cephalogr Clin Neurophysiol 9:51–52
Kinoshita K (1975) A study on response of EEG during Zen
meditation–alpha-blocking to name calling (in Japanese). Sei-
shin Shinkeigaku Zasshi 77:623–658
Lachaux JP, George N, Tallon-Baudry C, Martinerie J, Hugueville L,
Minotti L, Kahane P, Renault B (2005) The many faces of the
gamma band response to complex visual stimuli. Neuroimage
25:491–501
Lehmann D, Faber PL, Achermann P, Jeanmonod D, Gianotti LR,
Pizzagalli D (2001) Brain sourcesof EEG gamma frequency during
volitionally meditation-induced, altered states of consciousness,
and experience of the self. Psychiatry Res 108:111–121
Lehrer PM, Schoicket S, Carrington P, Woolfolk RL (1980)
Psychophysiological and cognitive responses to stressful stimuli
in subjects practicing progressive relaxation and clinically
standardized meditation. Behav Res Ther 18:293–303
Lutz A, Greischar LL, Rawlings NB, Ricard M, Davidson RJ (2004)
Long-term meditators self-induce high-amplitude gamma syn-
chrony during mental practice. Proc Natl Acad Sci USA
101:16369–16373
Lutz A, Dunne JD, Davidson RJ (2007) Meditation and the
neuroscience of consciousness. In: Zelazo PD, Moscovitch M,
Thompson E (eds) The Cambridge handbook of consciousness.
Cambridge University Press, Cambridge, UK
Lutz A, Slagter HA, Dunne JD, Davidson RJ (2008) Attention
regulation and monitoring in meditation. Trends Cogn Sci
12:163–169
Lutz A, Slagter HA, Rawlings NB, Francis AD, Greischar LL,
Davidson RJ (2009) Mental training enhances attentional stabil-
ity: neural and behavioral evidence. J Neurosci 29:13418–13427
Maloney KJ, Cape EG, Gotman J, Jones BE (1997) High-frequency
gamma electroencephalogram activity in association with sleep-
wake states and spontaneous behaviors in the rat. Neuroscience
76:541–555
Meador KJ, Ray PG, Echauz JR, Loring DW, Vachtsevanos GJ
(2002) Gamma coherence and conscious perception. Neurology
59:847–854
Melloni L, Molina C, Pena M, Torres D, Singer W, Rodriguez E
(2007) Synchronization of neural activity across cortical areas
correlates with conscious perception. J Neurosci 27:2858–2865
Mitchell DJ, McNaughton N, Flanagan D, Kirk IJ (2008) Frontal-
midline theta from the perspective of hippocampal ‘‘theta’’. Prog
Neurobiol 86:156–185
Niedermeyer E (1997) Alpha rhythms as physiological and abnormal
phenomena. Int J Psychophysiol 26:31–49
Niedermeyer E, Lopes da Silva FH (1999) Electroencephalography:
basic principles, clinical applications and related fields. Balti-
more: Williams and Wilkins
Ott U (2001) The EEG and the depth of meditation. J Med Med Res
1:55–68
Pagnoni G, Cekic M, Guo Y (2008) ‘‘Thinking about not-thinking’’:
neural correlates of conceptual processing during Zen medita-
tion. PLoS One 3:e3083
Pan W, Zhang L, Xia Y (1994) The difference in EEG theta waves
between concentrative and non-concentrative qigong states–a
power spectrum and topographic mapping study. J Tradit Chin
Med 14:212–218
Penolazzi B, Spironelli C, Angrilli A (2008) Delta EEG activity as a
marker of dysfunctional linguistic processing in developmental
dyslexia. Psychophysiology 45:1025–1033
Polich J (2007) Updating P300: an integrative theory of P3a and P3b.
Clin Neurophysiol 118:2128–2148
Polich J, Kok A (1995) Cognitive and biological determinants of
P300: an integrative review. Biol Psychol 41:103–146
Rabiner LR, Gold B (1975) Theory and application of digital signal
processing. Prentice-Hall, Englewood Cliffs, NJ
Reva NV, Aftanas LI (2004) The coincidence between late non-
phase-locked gamma synchronization response and saccadic eye
movements. Int J Psychophysiol 51:215–222
Rodriguez E, George N, Lachaux JP, Martinerie J, Renault B, Varela
FJ (1999) Perception’s shadow: long-distance synchronization of
human brain activity. Nature 397:430–433
Sauve K (1999) Gamma-band synchronous oscillations: recent
evidence regarding their functional significance. Conscious
Cogn 8:213–224
Segal ZV, Williams JMG, Teasdale JD (2002) Mindfulness-based
cognitive therapy for depression: a new approach to preventing
relapse. Guilford Press, New York
Cogn Process (2010) 11:39–56 55
123

Singer W (1993) Synchronization of cortical activity and its putative
role in information processing and learning. Annu Rev Physiol
55:349–374
Slagter HA, Lutz A, Greischar LL, Francis AD, Nieuwenhuis S, Davis
JM, Davidson RJ (2007) Mental training affects distribution of
limited brain resources. PLoS Biol 5:1228–1235
Slagter HA, Lutz A, Greischar LL, Nieuwenhuis S, Davidson RJ
(2009) Theta phase synchrony and conscious target perception:
impact of intensive mental training. J Cogn Neurosci 21:1536–
1549
Smallwood J, Schooler JW (2006) The restless mind. Psychol Bull
132:946–958
Smallwood J, McSpadden M, Schooler JW (2007) The lights are on
but no one’s home: meta-awareness and the decoupling of
attention when the mind wanders. Psychon Bull Rev 14:527–533
Smallwood J, Beach E, Schooler JW, Handy TC (2008) Going
AWOL in the brain: mind wandering reduces cortical analysis of
external events. J Cogn Neurosci 20:458–469
Tei S, Faber PL, Lehmann D, Tsujiuchi T, Kumano H, Pascual-
Marqui RD, Gianotti LR, Kochi K (2009) Meditators and non-
meditators: EEG source imaging during resting. Brain Topogr
22:158–165
Thomson DJ (1982) Spectrum estimation and harmonic analysis.
Proceed IEEE 70:1055–1096
Travis F (1991) Eyes open and TM EEG patterns after one and eight
years of TM practice. Psychophysiology 28:58
Travis F, Pearson C (1999) Pure consciousness: distinct phenomeno-
logical and physiological correlates of ‘‘consciousness itself’’.
Int J Neurosci 100:77–89
Travis F, Tecce J, Arenander A, Wallace RK (2002) Patterns of EEG
coherence, power, and contingent negative variation characterize
the integration of transcendental and waking states. Biol Psychol
61:293–319
Varela FJ (1995) Resonant cell assemblies: a new approach to
cognitive functions and neuronal synchrony. Biol Res 28:81–95
Wallace RK (1970) Physiological effects of transcendental medita-
tion. Science 167:1751–1754
Wenger MA, Bagchi BK (1961) Studies of autonomic functions in
practitioners of yoga in India. Behav Sci 6:312–323
Wilcox RR (2005) Introduction to Robust estimation and hypothesis
testing, 2nd edn. Elsevier Academic Press, San Diego, CA, USA
Yuval-Greenberg S, Deouell LY (2009) The broadband-transient
induced gamma-band response in scalp EEG reflects the
execution of saccades. Brain Topogr 22:3–6
Yuval-Greenberg S, Tomer O, Keren AS, Nelken I, Deouell LY
(2008) Transient induced gamma-band response in EEG as a
manifestation of miniature saccades. Neuron 58:429–441
56 Cogn Process (2010) 11:39–56
123