Does a Healing Procedure Referring to Theta Rhythms
Also Generate Theta Rhythms in the Brain?
Thilo Hinterberger, PhD,
Anna von Haugwitz, MD,
and Stefan Schmidt, PhD
Background/Objectives: ThetaHealing (Vianna Stibal, Kalispell, MT) is a spiritual healing method in which
the practitioner and client engage in joint meditations during several healing sessions. It is claimed that these
meditation periods are characterized by a ‘‘theta state’’ in which the presence of theta-waves in the electro-
encephalograph (EEG) frequency spectrum of both the healer and the client is supposed. This study sought to
test this hypothesis as well as the presence of synchronicities in the two EEGs.
Methods: Measurements were obtained with a dual EEG system with 2 · 32 channels, allowing for simulta-
neous EEG measurements of healer and client. Ten healers and 10 clients performed 10 ThetaHealing sessions
while the EEG was measured.
Results: Theta frequency band did not increase in healers or in clients. Rather, the contrary was found, with a
signiﬁcant decrease in theta-2 band during healing in healers. Small correlations were seen between the Fourier
amplitudes of healer and client in the theta-2 band, as well as small phase synchronicities in theta frequencies.
Conclusion: The hypothesis that ThetaHealing is associated with an enhanced generation of theta frequencies in
the brain could not be conﬁrmed. This ﬁnding makes no claim about whether ThetaHealing is beneﬁcial from a
new healing procedure termed ThetaHealing
(Vianna Stibal, Kalispell, MT) is making extraordinary
claims regarding healing effects and healing mechanism. The
method is propagating quickly, but to date no scientiﬁc eval-
uation has been conducted. One of the basic claims of Theta-
Healing is that an experienced theta healer enters a ‘‘theta
state’’ via special meditation. According to ThetaHealing, al-
leged healing effects are taking place in this special state,
which is also induced in the client by the healer. The name
ThetaHealing originates from the claim that this ‘‘theta state’’
is correlated with strong theta rhythmic activity patterns (4–
7 Hz) in the respective electroencephalogram (EEG) of healer
The theta rhythm is an oscillating EEG pattern in the fre-
quency range of 4–7 Hz. This theta band is often additionally
split into theta-1 (4–5 Hz) and theta-2 (5–7 Hz). On human
EEGs, theta is shown during resting state,
early slow-wave sleep.
In the waking state, theta has been
associated with both long-term and working memory tasks.
Theta power has also been found during meditation.
ThetaHealing is a healing procedure developed in the United
States by Vianna Stibal. According to descriptions on various
websites (see, for example, http://www.thetahealing.com),
Stibal discovered by chance that she was able to perform
‘‘instant healing.’’ At that time she had cancer of the right
femur and reports that her leg was instantaneously healed.
After this experience she developed several techniques and
started teaching them to others. On the basis of an EEG
assessment of this method that showed increased activity in
the theta frequency band, she called her technique Theta-
Meanwhile, the practice of ThetaHealing ex-
panded in the Western hemisphere. ThetaHealing courses
and ‘‘healing treatments’’ by ‘‘ThetaHealers’’ are available
in all major cities and countries. Courses follow the basic
design by Vianna Stibal and build on each other.
ThetaHealing claims to heal, more or less instantaneously,
many conditions for which conventional medicine has no
cure or needs much more time. There are several reports, for
example, of the disappearance of bone fractures, but also
claims of success in hepatitis C, Epstein-Barr virus infection,
HIV infection, herpes, tumors, various types of cancers, and
Department of Psychosomatic Medicine, University Medical Center Regensburg, Regensburg, Germany.
Academic Section Evaluation of Complementary Medicine, Department of Psychosomatic Medicine and Psychotherapy, University
Medical Center Freiburg, Freiburg, Germany.
Institute of Transcultural Health Studies, European University Viadrina, Frankfurt (Oder), Germany.
THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE
Volume 22, Number 1, 2016, pp. 66–74
ª Mary Ann Liebert, Inc.
To practice this method, the healer instructed by this
method enters into a unique state by means of a special
meditation. This state, called the theta state, is assumed to
be physiologically indicated by the presence of strong theta
EEG activity. The healer then aims to mentally connect to a
client, who, it is supposed, will also enter into this associ-
ated theta rhythmic brain activity state. According to the
theory of ThetaHealing, a kind of ‘‘reprogramming’’ takes
place by means of energy coming from an external
‘‘source’’: The roles of theta healer and client are intended
only to command and witness the change, which may take
place on several levels: physical, mental, emotional, or
The present study sought to investigate the claim of in-
duction of theta frequency on the EEG of the healer as well
as the client. A growing set of studies have investigated
coherence patterns between the neurophysiologic activity of
two related persons by the means of parallel and synchro-
nized EEG recordings (dual EEG). These studies have been
carried out in relation to interpersonal interaction paradigms
with both persons being in the same room
and also in
regard to anomalous information transfer between separated
In a recent dual EEG study, Ventura and
assessed spontaneous EEG patterns of Reiki
practitioners and clients and found an increase in coherence
in the theta band between the two persons that was not
present in control participants.
The current study used dual EEG assessment in which the
EEG from experienced ThetaHealers and their clients was
simultaneously recorded in parallel. The study hypotheses
were that (1) the healers taught in this method would be able
to elicit voluntary theta rhythmic activity in their EEG, (2)
theta rhythm in the brain activity of the client would in-
crease once the ThetaHealers attempted to connect their
state to that of the client, and (3) there would be signiﬁcantly
greater correlations in spectral EEG changes regarding the
theta bands between the healer’s and the client’s EEG once
healer and client had entered into the so-called theta state
compared to a control condition.
Materials and Methods
This experimental study included pairs of healthy
participants—the experienced healer and the naive partici-
pant (client)—in whom the EEG was recorded simulta-
neously. The experimental session was broken down into
several epochs with different instructions. The hypotheses
were assessed by comparing different phases in an intrain-
dividual design as detailed below.
Ten experienced healers trained in ThetaHealing were
recruited: nine from Germany and one from Canada. Re-
cruitment of healers took place through the informal net-
work of German ThetaHealers. One of the healers brought
his own client with him. For the remaining nine healers, the
investigators recruited clients naive to ThetaHealing. All
participants were required to abstain from caffeine on the
day of the session. Participants who took medication known
to interfere with the EEG spectrum were excluded.
Set-up and materials
The experiment took place in the Neuroscience Labora-
tory at the Institute for Environmental Health Sciences at the
University Medical Center Freiburg, Germany. The labora-
tory is equipped with a sound-attenuating chamber with
electromagnetic shielding, providing a suitable environment
for conducting EEG studies.
The laboratory was equipped with a 72-channel QuickAmp
ampliﬁer (bandpass, 0.01–100 Hz; sampling rate, 512 samples/s;
MES, Munich, Germany) and two 64-channel electrode caps
with active electrodes and active shielding (ActiCap, MES,
Munich). This 64-channel system was redesigned to record
a spontaneous EEG from 32 scalp positions (10–20 system)
with a common reference from two persons simultaneously.
Each of the two electrode sets had its own ground and ref-
erence electrode and its own independent active shielding
mechanism, but both were connected to the same ampliﬁer
(QuickAmp 72). By this procedure, the EEG of both par-
ticipants was recorded identically and simultaneously. This
procedure was tested thoroughly in several pilot studies to
guarantee that the recordings were completely independent
of each other and that no cross-talk between the two EEG
units was taking place. In addition to the EEG, an electro-
oculogram and electrocardiogram were recorded. Electrode
impedances for EEG were kept to less than 5 kO.
Participants arrived in pairs at the laboratory, and both
participants signed informed consent forms. Participants
completed a questionnaire on sociodemographic data, ex-
periences with ThetaHealing and meditation, and a mood
scale (Basler Beﬁndlichkeitsskala
). While the client was
being connected to the respective EEG system, the experi-
menter explained the speciﬁc tasks of the different epochs to
the healer. The healer was then also connected to the EEG
system. Healer and client were seated next to each other,
face to face. Communication with the experimenter took
place via an intercom. Once it was conﬁrmed that all the
equipment was working correctly, data recording started.
The experimental session was broken down into 10 phases
as shown below.
Phases 1 and 2: ﬁrst baseline measurement. Both par-
ticipants sat quietly in the closed chamber of the laboratory.
Two minutes of resting-state EEG was recorded from both
participants with eyes open and another 2 minutes with eyes
Phase 3: theta induction in the healer. The healer was
asked to enter the theta state as usually done in a Theta-
*Wackermann J, Naranjo Murada
s JR, Pu
tz P. Event-related
correlations between brain electrical activities of separated human
subjects: preliminary results of a replication study. The Para-
psychological Association, 47th Annual Convention, August, 5–8,
2004. Proceedings of Presented Papers (pp. 465–468).
THETA HEALING 67
Phases 4–8: theta elicitation in naive participant. In these
phases, the healer was asked to connect to the client and to
conduct four successive tasks from the ThetaHealing pro-
cedures. The start and end of the various phases were
communicated with the experimenter via the intercom, but
they were not named in order so that the client remained
blinded. At the end of phase 8, both healer and client left the
healing state and returned to their normal resting state.
Phases 9 and 10: second baseline measurement. Finally,
a second baseline measurement with eyes closed (2 minutes)
and eyes opened (2 minutes) was conducted.
Post-recording. At the end of the EEG recording, the
participants again completed the questionnaire on their mood
state (Basler Beﬁndlichkeitsskala). All sessions were sched-
uled in the morning in order to avoid tiredness as a confounder
(tiredness is known to be correlated with theta activity).
On the basis of the study hypotheses, predictions for the
different experimental phases were made as outlined in
Table 1. Here, an 11th phase was deﬁned as the composite
of phases 5–8, all healing phases in which theta was supposed
to be increased. For statistical analysis, phases 3–8 were
compared to the baseline phases 2 and 9 as described below.
EEG data were subjected to a quantitative analysis per-
formed using Matlab software, version 7.2 (MathWorks,
Natick, MA) and self-written scripts. Initially, the EEG was
scanned for high-amplitude artifacts, and eye-movement
artifacts were corrected on the EEG using a linear correction
algorithm, similar to that suggested by Gratton et al.
spectral decomposition, a fast Fourier transformation was
calculated each second with a moving window of 2 seconds,
resulting in 0.5-Hz resolution. Spectral amplitudes were
merged into eight frequency bands as follows: delta (0–
3.5 Hz), theta-1 (4.0–5.5 Hz), theta-2 (6–7.5 Hz), alpha-1
(8.0–10.0 Hz), alpha-2 (10.5–12.0 Hz), beta-1 (12.5–
15.0 Hz), beta-2 (15.5–25 Hz), and gamma (25.5–45.0 Hz).
For further statistical analysis, log-transformed band
power (logBP) was used. The data streams were separated
into 10 epochs according to the previously deﬁned phases.
Epochs of phases 5–8 were added, forming the 11th phase.
An epoch of 2 minutes would thus contain 120 time sam-
ples. The phases were averaged over time by calculating the
median rather than the mean of the logBP because the
median is known to be more robust with respect to outliers.
For the same reason, the standard deviation was calculated
through the interquartile range. This resulted in a data ma-
trix of 32 channels · 8 frequency bands · 11 phases for each
of the 10 healers and 10 clients.
Band power comparisons.
For the statistical comparison
of the changes in EEG band power during healing phases
compared with the nonhealing baseline phases, nonpara-
metric Wilcoxon rank-sum tests were carried out. Table 2
lists ﬁve different comparisons of interest together with the
hypotheses. The ﬁrst comparison (3 versus 2) compared
phase 3 (healer entering into ThetaHealing) with phase 2
(baseline, eyes closed). Here, according to our hypothesis,
an increase in theta activity would be expected for the data
of the healer but not for the client. The second comparison
(4 versus 2) compared phase 4 (the client also entered theta
state) with phase 2 (baseline, eyes closed). According to
Tables 1 and 2, an increase in theta activity would be ex-
pected in both client and healer. The same applies to the
comparison (11 versus 2) comparing all healing epochs (i.e.,
phases 5–8) with baseline. The last two comparisons com-
pare the two baselines at the beginning and the end of the
Table 1. Overview of Experimental Tasks and Respective Hypotheses of Changes in Theta Power
Phase Task Hypothesis: healer Hypothesis: client
1 and 2 Resting state baseline Normal theta Normal theta
3 Centering healer Theta increasing Normal theta
4 Centering client Increased theta Theta increasing
5 Exercise 1 (‘‘baby in the womb’’) Increased theta Increased theta
6 Exercise 2 (‘‘healing of the heart’’) Increased theta Increased theta
7 Exercise 3 (‘‘healing of the soul’’) Increased theta Increased theta
8 Exercise 4 (‘‘genetic healing’’) Increased theta Increased theta
9 and 10 Resting state baseline Normal theta Normal theta
Healing (composite of phases 5–8) Increased theta Increased theta
This phase was deﬁned for the analysis as a composite of phases 5–8.
Table 2. Predeﬁned Comparisons and Expected Outcome According to Hypotheses
Comparison Tasks Hypothesis: client Hypothesis: healer
3 vs. 2 Centering healer vs BL No change Theta increase
4 vs. 2 Centering client vs BL Theta increase Theta increase
11 vs. 2 Healing vs BL Theta increase Theta increase
9 vs. 2 BL end-begin (eyes closed) Trend? Trend?
10 vs. 1 BL end-begin (eyes opened) Trend? Trend?
68 HINTERBERGER ET AL.
session (eyes closed: 9 versus 2; eyes open: 10 versus 1).
Here, a potential overall trend in the power spectrum over
the entire session may be seen.
Intersubject correlations. EEG correlations were also
calculated between healer and client. Therefore, the Fourier
spectral amplitudes calculated for each second, each elec-
trode, each frequency band, and each participant served as
the basis for cross-correlating the time series of the healing
period (phase 11) of the healer data with the client data.
Zero-lag correlations were calculated for each electrode
separately to ﬁnd area-speciﬁc correlations. Global corre-
lations calculated from averaged electrodes were analyzed
as well. To test whether such correlations occurred with a
time delay in the clients, a lag/lead analysis was performed.
Therefore, the band-pass ﬁltered time series signals were
correlated after shifting of the client data samplewise up to a
delay of 250 samples. This corresponds to nonzero lag
cross-correlations from 0 to 1 second.
Intersubject phase synchronicity. The phase coherence
of band-pass ﬁltered signals between healers and clients in
various theta frequency bands was calculated to examine
whether brain oscillations at corresponding EEG locations
had signiﬁcant phase synchronicity. Therefore, the signal
was ﬁltered in four bands from 4 to 7 Hz, and each time
series was wavelet transformed using a Morlet wavelet with
the same frequency and a length of six cycles that was
shifted sample by sample through the time series. This re-
sulted in a time series of amplitudes and phase angles for
each electrode. Using the phase angle time series data of
healers and corresponding clients, correlation coefﬁcients
(r) and their p-values between healer and client were cal-
culated. The calculation was done for the series of all
healing conditions and all baseline conditions. This resulted
in a set of r-values and p-values of the dimension 10 par-
ticipants · 32 electrodes · 4 theta bands · 2 conditions. The
set of p-values was then corrected for multiple testing to
account for random signiﬁcances using the false discovery
rate (FDR) adjustment method.
The University Medical Center’s ethic committee ap-
proved the study. All participants gave written informed
consent before the start of the data collection.
The study comprised 20 participants: 10 healers (7 wo-
men and 3 men) and 10 clients (7 women and 3 men).
Participants had an average age of 38.7 years (range 27–54
years). Healers had a mean of 3 years of experience in
ThetaHealing, with an average of 253 healing sessions.
Individual global theta band power contrasts
An overall result was ﬁrst calculated to average the
spectral values of all electrodes. This global band power was
calculated for all ﬁve comparisons and all eight frequency
bands. Before showing statistical results, the analysis
FIG. 1. Individual theta band power differences between healing and baseline tasks in healers (A) and clients (B) .
Negative effect sizes (Cohen d) indicate decreases of theta power during the healing phases and are marked in black.
Positive values are marked in white. BL, baseline.
THETA HEALING 69
remained on the level of single subjects and shows the
standardized differences (Cohen d) of the theta band activ-
ities between the phases of interest. Figure 1A and B il-
lustrate the individual effects for healers and clients of the
ﬁve contrasts according to the comparisons of Table 2.
Figure 1A demonstrates that most healers showed reduced
theta-1 and theta-2 activities during healing. Similarly, most
clients also showed reduced theta activities during healing,
as shown in Figure 1B. Regarding the trend in comparing
baselines at the end (phases 9 and 10) with those at the
beginning (phases 1 and 2), most healers showed decreases
in theta power over the course of the whole session.
Global contrasts for all spectral bands
For statistical comparison of the phases, nonparametric
Wilcoxon rank-sum tests were calculated (n = 10). To ac-
count for false-positive results in the statistical signiﬁcance
test due to multiple electrodes, the surface maps were
corrected using the FDR adjustment.
Tables 3 and 4
provide the z-values; the signiﬁcant values after FDR ad-
justment are noted with a superscript ‘‘a.’’ There was no
signiﬁcant increase in any comparison among clients or
healers. However, there were signiﬁcant decreases in healers
in a broad spectrum, including theta-1 and theta-2, for the
comparison of the healing phases with the initial baseline.
Most comparisons contrasting healing phases with baseline
result in signiﬁcant decreases in or tendencies toward de-
creasing spectral power in healers and clients. The most sig-
niﬁcant effects could be observed in the alpha-1 and the beta-1
or beta-2 band in healers and clients.
Topographical contrasts for all spectral bands
Figures 2 and 3 also show topographical differences in
band power between different phases of the experiment for
all 10 sessions. Generally, most comparisons showed neg-
ative trends, and negative signiﬁcance was predominantly
seen in theta-2, alpha-1, and beta bands, indicating a de-
crease in power during healing. The two graphs clearly in-
dicate that hypotheses 1 and 2 (that theta activity would
increase in the healer and, later, accordingly in the client)
could not be conﬁrmed by the empirical data. None of the
comparisons signiﬁcantly increased for the clients. The
healers showed a signiﬁcant decrease in the theta-2 band
(phase 3) when entering into the so-called theta state. They
furthermore showed a signiﬁcant decrease in theta-1 and
theta-2 when connecting with the client (phase 4).
In the comparison of all healing epochs (phases 5–8) with
the baseline phase 2, there was also a signiﬁcant decrease in
theta activity in both bands. Regarding the overall trend,
which can be seen by comparing the initial and the ﬁnal
baseline of the 30-minute session, a signiﬁcant decrease in
theta-2 activity was seen in healers in the eyes-closed con-
ditions. The baselines, however (with open eyes), did not
lead to signiﬁcant trends in the power spectral densities.
Zero-lag correlation patterns between healer and client
For all participants and in all frequency bands, correlations
of Fourier amplitudes (averaged across all electrodes) between
healers and clients remained below r = 0.1. Highest correla-
tions could be observed in the gamma band. For the two theta
bands, correlations remained in a range of -0.015 < r < 0.013,
which is very small. Individual values for each healer-client
pair are displayed in Figure 4A.
The results from one-sample t tests across n = 10 are de-
picted in Table 5. Signiﬁcant positive correlations could be
observed in the theta-2 band (t[n = 10] = 2.94; p = 0.017) as
well as in the beta-2 band (t[n = 10] = 2.33; p = 0.045). Re-
garding the theta band only, theta-2 yields a signiﬁcant cor-
relation. An FDR adjustment of the p-values across all eight
bands leads to only nonsigniﬁcant results. Figure 4B and C
shows the t-test results of the theta-1 and theta-2 correlations
topographically. Red or dark blue areas would be signiﬁcant
Table 3. z-Values Resulting from Wilcoxon Rank-Sum Tests for Comparisons as per Table 2
in Various Bands of Power Spectrum Densities in All 10 Healers
Healers Delta Theta-1 Theta-2 Alpha-1 Alpha-2 Beta-1 Beta-2 Gamma-1
3 vs. 2 -1.68 -2.29 -2.80
4 vs. 2 -2.09 -2.29 -1.89 -2.50
11 vs. 2 -2.80
9 vs. 2 -1.48 -1.99 -2.60
-2.09 1.38 -1.78 -1.89 -1.17
10 vs. 1 1.78 -1.07 -0.97 -0.56 0.05 -1.07 -1.99 -0.66
p-values less than a = 0.05 after false discovery rate adjustment across all 5 · 8 values.
Table 4. z-Values Resulting from Wilcoxon Rank-Sum Tests for Comparisons as per Table 2
in Various Bands of Power Spectrum Densities in All 10 Clients
Clients Delta Theta-1 Theta-2 Alpha-1 Alpha-2 Beta-1 Beta-2 Gamma-1
3 vs. 2 0.87 0.36 -1.99 -2.80
4 vs. 2 -1.07 -0.66 -1.58 -2.19 -1.89 -1.48 -2.60
11 vs. 2 -1.89 -1.38 -1.58 -2.60
-1.99 -0.66 -2.19 -0.25
9 vs. 2 -0.25 0.25 -0.87 -1.48 -1.48 -0.25 -1.48 -1.17
10 vs. 1 1.27 -0.56 -0.76 0.25 -0.05 -0.05 -0.97 -1.07
p-values less than a = 0.05 after false discovery rate adjustment across all 5 · 8 values.
70 HINTERBERGER ET AL.
FIG. 2. Healers: topographical views of the comparisons of band power values between healing and baseline phases (rows
1–3) for all frequency bands and the changes in the baseline with eyes closed and opened (rows 4 and 5). Data from all
healers were used. False discovery rate–adjusted p-values are shown resulting from Wilcoxon rank-sum tests. The p-values
from positive z-scores are shown in light gray and when signiﬁcant in white, while negative z-scores are shown in dark gray
and when signiﬁcant in black.
FIG. 3. Clients: topographical views of the comparisons of band power values between healing and baseline phases (rows
1–3) for all frequency bands and the changes in the baseline with eyes closed and opened (rows 4 and 5). Data from all
clients were used. False discovery rate–adjusted p-values are shown resulting from Wilcoxon rank-sum tests. The p-values
from positive z-scores are shown in light gray and when signiﬁcant in white, while negative z-scores are shown in dark gray
and when signiﬁcant in black.
THETA HEALING 71
without correction for multiple testing. After FDR adjustment
across 32 electrodes, no signiﬁcant locations were visible.
Therefore, it is concluded that hypothesis 3 was not con-
ﬁrmed when the whole spectrum was considered.
As shown in Table 6, all absolute correlation values (r)at
any lag-time are less than r = 0.03 for theta-1 and less than
r = 0.02 for theta-2, which are both very small correlations.
The localized lag/lead analysis for all 32 electrode sites
revealed correlations less than r = 0.07 for theta-1 and less
than r = 0.06 for theta-2.
Analysis of intersubject theta-phase synchronicities
A further question was whether theta waves were phase
locked between healer and client. Therefore, a calculation of
phase angle correlation using wavelet transformed time series
for each pair of participants, each condition (healing or base-
line), each electrode, and each frequency between 4 and 7 Hz
was performed, resulting in a set of 10 · 2 · 32 · 4 = 2560
values. The FDR-adjusted p-values are displayed in Figure 5.
The corresponding set of correlation coefﬁcients was subject
to a four-way analysis of variance. Signiﬁcant main effects
were found for the factor participants with F(df = 9) = 8.92
( p < 0.001), as well as for the factor tasks with F(df = 1) = 4.18
( p = 0.04), and the factor electrodes with F(df = 31) = 2.64
( p < 0.001). No signiﬁcant main effect could be observed in the
factor frequency. The mean surface map over all sessions re-
vealed only nonsigniﬁcant results. The mean absolute theta
phase coherence [abs(R)] over all sessions and electrodes was
0.0048 in the healing phases and 0.0042 in the baseline phases.
This dual EEG study was performed to investigate one of
the basic claims of a new healing technique called Theta-
Healing. These claims were evaluated by three objectives
hypothesizing (1) a theta increase in the healer, (2) a theta
increase in the client, and (3) an increase in correlations of
EEG patterns between healer and client during healing
phases. The data did not support the hypotheses in any case.
No increase in theta activity occurred during the theta pro-
cedure. Theta activity remained more or less unchanged in
clients and even decreased in the healers. All these ﬁndings
conﬂict with the proposed hypotheses, and the claims for
this method, which are also responsible for the speciﬁc
name of ‘‘Theta’’-Healing, could not be supported.
The decrease of global alpha power, especially in phases 3
and 4 compared with phase 2, can be explained by a fre-
quently observed alpha behavior in which shortly after the
eyes are closed, alpha power increases more than in the sub-
sequent time of continuously closed eyes. A global decrease in
theta and delta power during healing compared with baseline
accompanied by a decrease in central beta power matches the
authors’ ﬁndings in meditation research. A similar pattern was
found in comparing idiosyncratic meditation practices to a
baseline recording or in comparing the intention for reaching a
state of thoughtless emptiness to a baseline condition similar
to that in this study.
This pattern was present only in the
healers and not the clients, suggesting that the healers entered
such a meditative state during their practice. To the authors’
knowledge, ThetaHealing asks for a state of thoughtless
emptiness in the healer so that he or she is open for receiving
relevant information from the client. This would support the
ﬁndings of this study.
A limitation of the current study may be the statistical
power. Ten participants is a rather small sample. However,
because the study found no tendency in the hypothesized
direction, the assumption that the study may have lacked the
statistical power to detect some effect of increased theta ac-
tivity power is unlikely. A positive outcome with respect to
the hypotheses in a larger sample would be possible only if
the additional participants provided data completely different
from those found in this study. On the other hand, in other
similar studies, a sample of 10 pairs was sufﬁcient to detect
relevant effects (e.g., Ventura and Persinger
FIG. 4. (a) Global correla-
tion values of each healer-
client pair for the theta-1
and theta-2 bands for healing
conditions. (b) Shows the
t-test results across partici-
pants of the theta-1 correla-
tions topographically. After
FDR adjustment across elec-
trodes no signiﬁcant loca-
tions were visible. Color
images available online at
Table 5. t-Test Results and Related p-Values of Correlations of 10 Participant Pairs
Variable Delta Theta-1 Theta-2 Alpha-1 Alpha-2 Beta-1 Beta-2 Gamma-1
t (n = 10) 2.32 -0.10 2.94
1.30 0.69 -0.15 2.33
p-Value 0.045 0.920 0.017
0.225 0.505 0.884 0.045
Signiﬁcant ( p < .05) values without correction for multiple testing.
False discovery rate adjustment yields no signiﬁcant values.
72 HINTERBERGER ET AL.
Although positive correlations for the Fourier amplitudes
were found between healers and clients in the theta-2 band
in 8 of 10 sessions, those correlations were very small
(r < 0.015) and therefore may be regarded as irrelevant from
a functional perspective. Furthermore, the result of the to-
pographically resolved correlation map was nonsigniﬁcant
after FDR adjustment. In the non-zero lag correlation, small
but sometimes signiﬁcant correlations were again found for
single sessions. In a third approach, these intersubject cor-
relations were investigated by calculating the intersubject
phase coherence for four theta frequencies. Again, very
small but, after FDR adjustment, still signiﬁcant phase
synchronicities were found. However, averaging those data
across sessions (i.e., across participant pairs) resulted in
nonsigniﬁcant coherences. This shows that even if pairs of
participants had nonrandom phase synchronicities during a
session, the topography and frequency structure of these
synchronicities had no similarities between sessions.
The correlations found are not mysterious but can most
likely be explained by the fact that both subjects were sitting
close to each other in the same measurement chamber
during the sessions. Thus, they were exposed to the same
acoustic surrounding and also shared some visual stimuli
caused by each other. Because it is well known that any
stimulation causes evoked brain responses, which are phase-
locked to the event, it can be assumed that this environ-
mental connectedness causes intersubject correlations. A
closer analysis of this hypothesis would require the time-
series of the acoustic events and perhaps visual perceptions
(if participants did not have their eyes entirely closed). Since
these data were not recorded, such an analysis is not pos-
sible. Another option would have been a sensory separation
of the participants. However, the interconnectedness of
brains beyond a common sensory environment was not the
research question of the present study.
Another limitation might be that the somewhat artiﬁcial
setting, which was used to record simultaneous EEGs from
client and healer. Several constraints had to be introduced; for
example, client and healer were not allowed to talk to each
other during several phases of the experiment. Furthermore,
Table 6. Non-Zero Lag Correlations Averaged Across Lag Times (0–250 Samples)
and 32 Electrodes of 10 Participant Pairs
Participant 1 2 3 4 5 6 7 8 9 10
Theta-1 max(abs(r)) 0.020 0.017 0.023 0.022 0.015 0.018 0.024 0.018 0.024 0.013
Theta-1 mean(abs(r)) 0.016 0.015 0.019 0.015 0.013 0.015 0.019 0.015 0.018 0.009
Theta-2 max(abs(r)) 0.012 0.017 0.016 0.013 0.011 0.017 0.016 0.011 0.017 0.010
Theta-2 mean(abs(r)) 0.009 0.011 0.013 0.009 0.009 0.010 0.011 0.009 0.013 0.008
Mean and maximum correlation values are reported.
FIG. 5. False discovery rate (FDR)–adjusted p-values of the intersubject phase coherence for each participant pair and
condition (healing and baseline) in separate images. Each image contains p-values of 32 electrodes and 4 frequencies from 4
to 7 Hz. Blue p-values represent signiﬁcant differences at a 0.05 level after correction for false discoveries. Color images
available online at www.liebertpub.com/acm
THETA HEALING 73
both the laboratory situation and the time-consuming prepa-
ration of the EEG introduced crucial differences compared to
a normal healing session. Before the start of the healing
session, several tests had to be conducted. Moreover, stimu-
lated by the more technical research context, healer and client
may have felt some pressure to ‘‘prove’’ their abilities. These
atypical circumstances may have resulted in some tension or
relaxation problems in the healer and the client. However,
none of the healers complained about the circumstances.
Overall, the ﬁndings indicate that ThetaHealing most likely
does not generate an increase in theta band power activity.
This does not mean that ThetaHealing is not effective as a
healing method. This study did not investigate any clinical
aspect of this approach. Thus, even in light of the preceding
ﬁndings, further studies should evaluate ThetaHealing.
The authors thank the BIAL Foundation for the support of
this project within grant no. 42/10. They also thank Mi-
chaela Peternell, Claudia Orellana Rios, Jose
and Dennis Centeno for their valuable help in data recording
and Majella Horan for assistance in manuscript preparation.
Author Disclosure Statement
No competing ﬁnancial interests exist.
1. Vianna S. ThetaHealing: Introducing an Extraordinary
Energy Healing Modality. Carlsbad, CA: Hay House, 2010.
2. Nunez PL, Srinivasan R. Electric Fields of the Brain. Ox-
ford: Oxford University Press, 2006.
3. Kandel ER, Schwartz JH, Jesell TM, eds. Principles of
Neural Science. New York: McGraw-Hill, 2000.
ki G. Theta rhythm of navigation: link between path
integration and landmark navigation, episodic and semantic
memory. Hippocampus 2005;15:827–840.
5. Sauseng P. Control mechanisms in working memory: a
possible function of EEG theta oscillations. Neurosci Bio-
behav Rev 2010;34:1015–1022.
6. Aftanas L, Golosheykin S. Impact of regular meditation
practice on EEG activity at rest and during evoked negative
emotions. Int J Neurosci 2005;115:893–909.
7. Cahn BR, Polich J. Meditation states and traits: EEG, ERP,
and neuroimaging studies. Psychol Bull 2006;132:180–211.
8. Stibal V. ThetaHealing: Disease and Disorder. Ammon,
Rolling Thunder Publishing, 2009.
9. Hari R, Himberg T, Nummenmaa L, et al. Synchrony of
brains and bodies during implicit interpersonal interaction.
Trends Cogn Sci. 2013;17:106–106.
10. Tognoli E, Lagarde J, DeGuzman GC, et al. The phi
complex as a neuromarker of human social coordination.
Proc Natl Acad Sci U S A 2007;104:8190–8195.
11. Charman RA. Has direct brain to brain communication
been demonstrated by electroencephalographic monitoring
of paired or group subjects? J Soc Psych Res 2006;70:1–24.
12. Hinterberger T. Searching for neural markers of psi: a
summary of three studies measuring electrophysiology in
distant participants. In: Roe CA, Kramer W, Coly L,
eds. Utrecht II Charting the Future of Parapsychology. Pro-
ceedings of the International Conference. New York: Para-
psychology Foundation, 2008:46–60.
13. Wackermann J, Seiter C, Keibel H, et al. Correlations be-
tween brain electrical activities of two spatially separated
human subjects. Neurosci Lett 2003;336:60–64.
14. Ventura AC, Persinger MA. Enhanced coherence within the
theta band between pairs of brains engaging in experienced
ve Reiki procedures. J Altern Complement Med
15. Hobi V. Basler Beﬁndlichkeits-Skala. Ein self-rating zur
Verlaufsmessung der Beﬁndlichkeit. Weinheim: Beltz Test
16. Gratton G, Coles MG, Donchin E. A new method for off-
line removal of ocular artifact. Electroencephalogr Clin
17. Benjamini Y, Hochberg Y. Controlling the false discovery
rate: a practical and powerful approach to multiple testing.
J R Stat Soc Ser B Methodol 1995;57:289–300.
18. Yekutieli D, Benjamini Y. Resampling-based false dis-
covery rate controlling multiple test procedures for corre-
lated test statistics. J Stat Plan Inference 1999:171–196.
19. Hinterberger T, Schmidt S, Kamei T, et al. Decreased
electrophysiological activity represents the conscious state
of emptiness in meditation. Conscious Res 2014;5:99.
Address correspondence to:
Stefan Schmidt, PhD
Academic Section Evaluation of Complementary Medicine
Department of Psychosomatic Medicine
University Medical Center Freiburg
Hauptstr. 8, 79104 Freiburg
74 HINTERBERGER ET AL.