Neuroscience Research 156 (2020) 117–129

Contents lists available at ScienceDirect

Neuroscience Research
journal homepage: www.elsevier.com/locate/neures

Review article

Neuromagnetic oscillations in the human sensory systems: A mini

review of our series and literature
Shozo Tobimatsu
Department of Clinical Neurophysiology, Neurological Institute, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku,
Fukuoka, 812-8582, Japan

a r t i c l e i n f o a b s t r a c t

Article history: Oscillatory neuronal (electrical) activity in defined frequency ranges supports synchronous interactions
Received 8 July 2019 between anatomically distinct regions of the human brain during cognitive tasks. Here, the author reviews
Received in revised form 6 November 2019 our previous studies that focused on the neuromagnetic oscillations in the sensory systems in response
Accepted 27 November 2019
to the external stimuli in normal healthy subjects and neurological disorders. A magnetoencephalog-
Available online 24 December 2019
raphy was applied to evaluate the neuromagnetic oscillations in humans. We have demonstrated that
the oscillatory gamma synchronization binds the primary and secondary somatosensory areas (S1 and
Keywords:
S2) in humans. This functional coupling is modulated by aging. In people who stutter, functional and
Magnetoencephalography
Time-frequency analysis structural reorganization of the right auditory cortex appears to be a compensatory mechanism for
Sensory systems impaired left auditory cortex function. This may be partly caused by increased right hemispheric local
Neurological disorders phase synchronization and increased inter-hemispheric phase synchronization. We have also found that
Oscillology the hippocampus modulates auditory processing differently under normal conditions and in epileptic
patients with hippocampal sclerosis. This indicates that altered neural synchronization may provide
useful information about possible functional deterioration in patients with unilateral mesial tempo-
ral lobe epilepsy. Finally, supraspinal (cortical) mechanism is responsible for pain perception and pain
relief via neural oscillations. Together, neuronal synchronization plays an important role in distributed
cortico-cortical processing.
© 2019 Elsevier B.V. and Japan Neuroscience Society. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
2. Principal approach to studying event-related neuronal oscillations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
2.1. Time-frequency analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
2.2. Continuous wavelet transforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
3. Evoked vs. induced oscillations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
3.1. Evoked power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
3.2. Induced power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
3.3. Event-related phase locking across trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
4. Roles of the oscillatory gamma synchronization in the primary (SI) and secondary somatosensory (SII) areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
4.1. Induced gamma oscillations between SI and SII . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
4.2. Aging effects on the induced oscillations in SII . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
5. Mechanisms of altered auditory system in stuttering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
5.1. Abnormal auditory system in stuttering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
5.2. Abnormal auditory synchronization in stuttering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
6. Hippocampal modulation of auditory processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
6.1. Effects of hippocampal sclerosis on central auditory processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
6.2. Effects of HS on auditory neural synchronization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

E-mail address: tobi@neurophy.med.kyushu-u.ac.jp

https://doi.org/10.1016/j.neures.2019.12.007
0168-0102/© 2019 Elsevier B.V. and Japan Neuroscience Society. All rights reserved.
118 S. Tobimatsu / Neuroscience Research 156 (2020) 117–129

6.3. Effects of HS on auditory gamma band oscillations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

7. Pain perception and neural oscillations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
7.1. Cortical pain relief by tactile stimulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
7.2. Frequency-dependent changes in sensorimotor and pain affective systems induced by empathy for pain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
8. Conclusion and future perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128

1. Introduction Recently, neural oscillations and coordination dynamics have

been paid attention to understand the normal brain function as well
The aim of the systems neuroscience is to understand how as the neuropsychiatric disorders (Uhlhaas and Singer, 2015). It is
billions of neurons in the human brain work synchronously to well recognized that rhythmic synchronization of neural discharges
proceed the real-world information (Knight, 2007). Oscillatory neu- may provide the necessary spatial and temporal links that bind
ronal (electrical) activity in defined frequency ranges supports together the processing in different brain areas to build a coherent
synchronous interactions between anatomically distinct regions percept (Tallon-Baudry and Bertrand, 1999). To visualize such oscil-
of the human brain during cognitive tasks that require voluntary lations, event-related EEG/MEG time frequency analysis provide a
or automatic attention, memory, or visual processing (Tallon- view of the serial or sequential events in the brain’s information
Baudry and Bertrand, 1999). It is now widely accepted that defining processing stream.
network interactions is key to understanding normal cognition.
In our laboratory, we have been studying the neuromagnetic
2.1. Time-frequency analysis
oscillations in the sensory systems in response to the external stim-
uli in normal healthy subjects and neurological disorders using
Neural oscillations appear with a jitter in latency from one trial
magnetoencephalography (MEG). The magnetic field passes unaf-
to another, centered around a given latency (Figs. 1 and 2). There-
fected through brain tissue and the skull. Although the magnetic
fore, it tends to cancel out completely in the averaged evoked
field is extremely small, it can be detected by superconducting
potential. Specific methods must be used in order to character-
quantum interference detectors that are based on superconduc-
ize induced activities. The time–frequency power of the signals by
tivity. By analyzing the spatial distributions of magnetic fields,
means of a complex Morlet’s wavelet transform which provides a
it is possible, by using a model such as a single equivalent
better compromise between time and frequency resolutions (Roach
current dipole, to estimate the intracranial localization of the
and Mathalon, 2008). When this analysis is applied to the aver-
generator source and superimpose it on an MRI (Tobimatsu and
aged response, the phase-locked activity can be clearly identified.
Kakigi, 2016). Thus, MEG has extremely high temporal and spa-
When this method is applied to each single trial, followed by an
tial resolution. Other functional modalities such as functional MRI,
averaging of the powers across trials, it becomes possible to iden-
except invasive EEG, have either poor temporal or spatial resolu-
tify non-phase-locked activity as long as the signal-to-noise ratio
tion.
is high enough and the jitter does not exceed the wavelet duration
Overall, MEG is nowadays a well-established, non-invasive tech-
(Fig. 2).
nique, which can be used for basic science research and clinical
As the above mentioned, event-related EEG/MEG signals are
applications (Hari et al., 2018). This review focuses on the neu-
decomposed into magnitude and phase information for each fre-
rophysiological implications of the neuromagnetic oscillations in
quency present in the EEG/EMG (Roach and Mathalon, 2008). This
the human brain. For this purpose, the author introduces our find-
approach characterizes their changes over time on a millisecond
ings on the induced oscillations produced by the sensory system in
time scale with respect to task events (Fig. 1). Broadly speaking,
response to the external stimuli. Then, the author will discuss the
this approach is referred to as “time-frequency analysis”. Time-
roles of the neural oscillation in the sensory system in the normal
frequency analyses of EEG/MEG provide additional information
human brain as well as neurological disorders.
about neural synchrony not apparent in the ongoing EEG/MEG.
They can tell us which frequencies have the most power at specific
points in time and space and how their phase angles synchronize
2. Principal approach to studying event-related neuronal
across time and space.
oscillations

Electroencephalographic (EEG)/MEG recordings have been used 2.2. Continuous wavelet transforms
to study the sensory and cognitive information processing. When
a large number of parallel-oriented cortical neurons receive the Wavelets are waveforms of limited duration that have an aver-
same repetitive synaptic input and/or generate the same repetitive age value of zero. The waveforms contained in the wavelet must
sequence of outputs, their synchronous activity produces extracel- provide a biologically reasonable fit to the signal being modeled.
lular rhythmic field potentials. These open rhythmic field potentials One common type of biologically plausible wavelet, the Morlet
can be recorded as EEG from the scalp if they are strong enough wavelet, is a Gaussian-windowed sinusoidal wave segment com-
and have the right orientation (i.e., perpendicular or radially ori- prising several cycles (Roach and Mathalon, 2008; Tallon-Baudry
ented fields with respect to the scalp surface produce stronger scalp et al., 1999). Wavelet analyses utilize a different time window
potentials than parallel or tangentially oriented fields) (Nunez et al., length for each frequency, with the longest windows applied to the
2019). In contrast, MEG preferentially picks up the tangentially ori- lowest frequencies and the shortest windows applied to the high-
ented fields. Thus, if the neural activity recorded by EEG/MEG were est frequencies. For example, assuming a wavelet family contains
not already synchronized and not already powerful, it would not be 6 cycles of a sinusoidal oscillation, the wavelet for the 10-Hz fre-
evident at the scalp. Therefore, even before it is spectrally decom- quency spans a time window of 600 ms, whereas the wavelet for the
posed, EEG/MEG at the scalp is clear evidence of neural synchrony 40 Hz frequency spans a time window of 150 ms. This variation in
of cortical activity. the wavelet from coarser to finer temporal resolution with increas-
 S. Tobimatsu / Neuroscience Research 156 (2020) 117–129 119

Fig. 1. An individual EEG trial (A) is convolved with a complex Morlet wavelet (B), containing both real (solid line) and imaginary (dotted line) wave components, to produce
a single, complex time-frequency data point (C) consisting of both real (x-axis, denoted r) and imaginary (y-axis, denoted i) parts. This operation is known as “convolution”
that can be defined as the multiplication of one series or “vector” of numbers by another (*). By sliding this windowed function across the EEG time series one point at a time,
a complex number at the window’s center point is estimated for each time point in the EEG (C). The length of the vector from the origin (0, 0) to the complex data point (real,
imaginary) depicts the signal magnitude, and the angle this vector forms with the x-axis depicts the phase angle (␪), for the 100-ms time point. The wavelet’s center time
point (hatched vertical line, B) is overlaid on the 100-ms time point in the EEG epoch (hatched vertical line, A) to perform this convolution. Single subject’s complex data
points from 5 trials (each shown in different color) for the 40-Hz frequency and 100-ms poststimulus time point, plotted on the real (x) and imaginary (y) axes (D(a)). After
removing the phase angles from the complex numbers, remaining magnitude values are squared and then averaged (black line), providing an estimate of total power at 40 Hz
and 100 ms (D(b)). After removing the magnitude values from the complex numbers, remaining equal length vectors, which retain phase angle information, are averaged to
obtain the phase-locking factor (PLF) (length of vector shown in black) (D(c)). Repeating steps D(b) and D(c) for every trial, time, and frequency point in the dataset yields a
time-frequency matrix of total power values (D(d)) and PLF values (D(e)) (with permission, from Roach and Mathalon, 2008, Oxford Press).

ing frequency is accomplished at the cost of diminishing frequency
resolution as frequency increases. The sinusoidal waves contained
in a wavelet are typically shaped by an envelope function (e.g., a
Gaussian function), such that the wavelet has its largest magnitude
at the center time point and tapers off toward the edges of the time
window. Fig. 1 shows that single-trial EEG epochs (Fig. 1A) can be
spectrally decomposed into complex numbers for each EEG time
point, providing estimates of the magnitude and phase angles of the
oscillations (Fig. 1C) at any given frequency. This is accomplished
through multiplication of the EEG with a windowed transforma-
tion function (e.g., Morlet wavelet transform, as shown in Fig. 1B)
centered on a segment of the EEG epoch. When this is done for each
trial, the complex number values for a specific time point relative
to an event’s onset (e.g., stimulus onset) are collected across trials
(Fig. 1D).
3. Evoked vs. induced oscillations
across trials (Fig. 2A–C). The phase-synchronized oscillations in the
EEG/MEG across trials are isolated by first time domain averag-
ing the event-locked EEG/MEG epochs to derive the event-related
potentials (ERPs)/event-related magnetic fields (ERFs). Frequencies
that are phase synchronized with respect to stimulus onset across
repeated trials survive the averaging process and can be seen in
the average ERP/ERF. This is not the case for oscillations that are
out of phase with respect to stimulus onset across trials, which can-
cel out toward zero during the averaging used to generate ERP/ERF.
Accordingly, evoked power is calculated by spectral decomposition
of an individual’s ERP/ERF, squaring the magnitude values associ-
ated with each time and frequency point in the time-frequency
matrix. Evoked power in specific frequencies, such as the gamma
band, have been linked to sensory registration (Tallon-Baudry et al.,
1996; Fries et al., 2001) as well as to top-down cognitive processing
of stimulus events (Spencer et al., 2007; Debener et al., 2003) and
generally occur within the first 200 ms following stimulus onset.

3.1. Evoked power 3.2. Induced power

Evoked power refers to event-related changes in EEG/MEG Induced power refers to event-related changes in EEG/MEG
power that are phase-locked with respect to the event onset power that are time-locked, but not phase-locked, with respect to
120 S. Tobimatsu / Neuroscience Research 156 (2020) 117–129

Fig. 2. Algorithm for visualization of the induced gamma response. (A) Successive EEG trials (simulated data) with a small amplitude gamma response phase-locked to stimulus
onset (blue boxes) and a gamma burst jittering in latency (green boxes). (B) Averaging across single trials leads to the conventional evoked potential. (C) Time–frequency
power representation of the evoked gamma response. The abscissa represents time, and the ordinate, frequency. The color scale codes the variations of power (positive or
negative) with respect to a pre-stimulus baseline. The non-phase-locked activity is cancelled out. When the time frequency power is computed for each singe trial (D), and
then averaged across trials (E), the induced gamma response is revealed (with permission, from Tallon-Baudry and Bertrand, 1999, Elsevier Science).

the event onset across trials (Fig. 2D, E). Induced power, also known
as “asynchronous power” or “phase-invariant power”, is contained
within, and is sometimes confused with, total power because the
latter is calculated from time-frequency decomposition of single
trial EEG/MEG epochs using only the squared magnitude informa-
tion without regard to the phase of the signal. It is important to
recognize that evoked power is explicitly removed from measures
of total power because the latter actually contain both phase-locked
and non–phase-locked power. Thus, to isolate pure induced power,
evoked power must be removed from the single trial–based total
power estimate.
3.3. Event-related phase locking across trials
Event-related phase consistency, or phase locking with respect
to an event’s onset, across trials can be calculated within one elec-
trode, complementing the total power measure described above.
To this end, the phase information shown in Fig. 1D(c) in which
magnitude information has been unit normalized (i.e., transformed
to 1) is used. By averaging these normalized complex numbers
across trials for each time point and frequency bin, a 2-dimensional
matrix of time-frequency values describing the consistency of the
phase angles with respect to an event’s onset is obtained. Specif-
ically, each value in this time-frequency matrix is a real number
between zero and one, with zero reflecting a completely uniform
random distribution of phase angles between trials and with one
reflecting identical, or perfectly synchronized, phase angles across
trials. The measure defined by these values has been called phase-
locking factor (PLF) (Tallon-Baudry et al., 1997) and it represents
one minus the circular variance of phases (i.e., phase variance) for
each time-frequency point examined. Event-related phase locking
is an important complement to total power because the complex
number magnitude values on which power calculations depend
have no influence on the phase angles used to calculate phase lock-
ing. Importantly, the term “PLF” is not analogous to one of the terms
commonly used to describe the consistency of phase differences
 S. Tobimatsu / Neuroscience Research 156 (2020) 117–129 121

between two electrodes across trials, “phase-locking value (PLV)” cSII and between cSII and iSII occurred in the early post-stimulus
(Lachaux et al., 1999). stage. Therefore, the oscillatory gamma synchronization binds the
early-stage tactile information processing within the somatosen-
sory cortical network in normal subjects. We also demonstrated
4. Roles of the oscillatory gamma synchronization in the
the decrease in PLVs between SI and SII in MS patients compared
primary (SI) and secondary somatosensory (SII) areas
with healthy controls (HCs). Because subcortical U-fiber lesions and
periventricular white matter lesions in MS patients may frequently
In the somatosensory system, SI and SII comprise the early
involve the intra- and interhemispheric associative pathways, the
stage of the hierarchical organization. Electrophysiological studies
lesion load might significantly contribute to the decreased neu-
of evoked responses have often reported differences in activation
ronal synchronization within the somatosensory cortical network
timing between SI and SII, with the SI response occurring earlier
(Tecchio et al., 2008). Overall, it is notable that this is the first MEG
than the SII responses. The typical characteristics of SII include
study to demonstrate impaired functional connectivity between SI
sensitivity towards higher order functions, such as sensorimotor
and SII in MS.
integrations, attention, unitary body image, and integration of noci-
ceptive and non-nociceptive information (for a review, see Lin and
Forss, 2002). Anatomically, it has been shown that the neurons in 4.2. Aging effects on the induced oscillations in SII
SII have larger and more complex receptive fields than those in SI
(for a review, see Iwamura, 1998). Therefore, SII is often considered Age-related changes are well documented in SI in which the
to be a hierarchically higher cortical area than SI. amplitude of N20 of somatosensory evoked potentials, which is
the initial cortical response following electrical stimulation of
the median nerve or fingers. The age-related electrophysiological
4.1. Induced gamma oscillations between SI and SII
changes in SI were also confirmed by MEG studies, which demon-
strated that the N20 m component, an MEG counterpart of the N20
Neuronal synchronization in the gamma-frequency band has
component, exhibited increased amplitude in the old (Huttunen
received increasing attention as the salient mechanism for cortico-
et al., 1999). Although the precise physiological mechanism has not
cortical information processing (Tallon-Baudry and Bertrand, 1999;
been fully elucidated, the increased amplitude is probably caused
Roach and Mathalon, 2008). Recently, the temporal binding
by cortical disinhibition in the old (Huttunen et al., 1999; Stephen
mechanismin various gamma-frequency bands has been well rec-
et al., 2006). In addition to the amplitude difference, the N20 latency
ognized in several studies of top–down and bottom–up information
is also prolonged in the old because of the slowing of conduction
processing among anatomically distributed cortical areas (Tallon-
velocity in the peripheral nerves and spinal cord. So far, less is
Baudry et al., 1997; Engel et al., 2001). However, the significance
known about age-related change in SII. Therefore, we quantified
of temporal binding theory in the functional relationship between
age-related changes across SI and SII mainly based on oscillatory
SI and SII has not yet been established. To elucidate the neural
activity indices measured with MEG (Hagiwara et al., 2014).
synchronization in the early-stage somatosensory processing, we
We recorded SEFs to right median nerve stimulation in healthy
studied the functional connectivity between SI and SII in healthy
young and old subjects and assessed major SEF components. Fif-
subjects using MEG (Hagiwara et al., 2008).
teen young volunteers (29.0 ± 4.1 years) and fifteen old volunteers
Twenty-three healthy volunteers (18 women and 5 men, mean
(63.7 ± 3.7 years) participated in this study. We evaluated the PLF
age 37.3 ± 10.6 years) and 23 patients with clinically definite
for local field synchrony on neural oscillations and the weighted
multiple sclerosis (MS) patinets (18 women and 5 men, mean
phase-lag index (wPLI) for cortico-cortical synchrony between SI
age 38.8 ± 8.1 years) according to the revised McDonald criteria
and SII. wPLI measures the consistency of the phase relationship
(Polman et al., 2005) participated in this study. Somatosensory
between signals in separate areas while diminishing the effect
evoked magnetic fields (SEFs) to right and left median nerve stim-
of volume conduction (Vinck et al., 2011). PLF was significantly
ulation were recorded to determine the locations of each cortical
increased in SI along with the increased amplitude of N20 m in the
activity. As a result, five major deflections of the SEF waveforms in
old subjects. PLF was also increased in SII associated with a short-
the contralateral SI (cSI) and bilateral SII areas (cSII and iSII) were
ened peak latency of SEFs. wPLI analysis revealed the increased
identified. Short-latency deflections, N20 m, P35 m, and P60 m,
coherent activity between SI and SII in the old subjects (Fig. 4).
which were recorded over the cSI whereas middle-latency deflec-
The most novel finding of this study is that the age-related
tions were observed over the cSII and iSII. The peak latency of cSII
changes occur outside SI, namely in SII. PLF measures local field
was earlier than that of iSII. All cSI deflections were clearly identi-
synchrony, while wPLI reflects functional connectivity between
fied in all normal subjects, whereas N20 m and P35 m deflections
the two cortical areas. Thus, age-related changes are profound
were not always recorded in MS patients. Then, we analyzed the
across the somatosensory areas with interaction between each
PLVs of the induced gamma activity to assess neural synchrony
other. Our results suggest that the functional coupling between
within the somatosensory cortical network. We also assessed PLVs
SI and SII is influenced by the cortical disinhibition due to normal
in patients with MS to validate our PLV analysis in evaluating the
aging. Therefore, we provide the first electrophysiological evidence
inter-areal functional connectivity, which can often be impaired in
for age-related changes in oscillatory neural activities across the
MS. The PLVs of the induced gamma activity were calculated for
somatosensory areas.
each pair of unaveraged MEG signals that represented the activi-
ties of the contralateral SI and bilateral SII areas. Analysis of PLVs
between the SI and SII areas showed significantly increased PLVs 5. Mechanisms of altered auditory system in stuttering
for gamma-band activities, starting at an early post-stimulus stage
in NCs, whereas this increase in PLVs was apparently diminished Stuttering is a developmental disorder that affects speech flu-
in MS (Fig. 3), especially in high gamma band (50−70 Hz). ency. The mechanism of stuttering is still a matter of debate. People
The temporal binding mechanism in various gamma-frequency who stutter (PWS) can reduce their stuttering rates under mask-
bands among anatomically distributed cortical areas is responsible ing noise and altered auditory feedback; such a response can be
for top–down and bottom–up information processing (for a review, attributed to altered auditory input (for a review, see Lincoln et al.,
see Engel et al., 2001). To our knowledge, we first demonstrated 2006). Postma and Kolk (1992) proposed “auditory feedback defect
that synchronized gamma-band activities in both between cSI and theories in PWS,” in which PWS have deviant error monitoring of
122 S. Tobimatsu / Neuroscience Research 156 (2020) 117–129

Fig. 3. Time-frequency analysis with the continuous wavelet transform in a representative subject. (A) Induced gamma-band activities (30–70 Hz) recorded by the sensors
showing the SI and SII activities. (B) Topography of the induced gamma activities around the SI and SII areas in the same subject. (C) The grand average of PLVs in normal
subjects shows increased PLVs in the entire gamma-frequency band (30–70 Hz) (upper row). The high PLVs in the high-frequency gamma-band (50–70 Hz) are observed in
the earlier latencies (mostly in the post-stimulus period of 20–50 ms), whereas those in the low-frequency gamma-band (30–40 Hz) last up to 100 ms. There are also some
phase-locking activities of various frequencies during the later post-stimulus period (> 100 ms). In MS patients, decreased PLVs are noted in the entire frequency range (lower
row) (with permission, from Hagiwara et al., 2010, Elsevier Science).

speech production, namely, PWS detect errors more than people
who do not stutter. This suggests that auditory input processing
could be different in PWS compared with non-PWS.
5.1. Abnormal auditory system in stuttering
The details of abnormal processing of basic auditory information
remain unclear in PWS. In order to characterize such abnormal-
ities, we examined the functional and structural changes in the
auditory cortices of PWS by using a 306-channel MEG system
to assess auditory sensory gating (P50 m suppression) and tono-
topic organization (Kikuchi et al., 2011). Additionally, we employed
voxel-based morphometry to compare cortical gray matter (GM)
volumes on structural MR images.
Two successive clicks evoked a well-defined P50 m response in
the hemisphere contralateral to the stimulated ear in PWS (n = 18)
and HCs (n = 18). The amplitude of S2-P50 m was apparently smaller
than that of S1-P50 m in the control subjects. However, no such
amplitude difference between S1-P50 m and S2-P50 m was evident
in the left hemisphere in PWS. This suggests that PWS has impaired
left auditory sensory gating. The tonotopic organization in the right
hemisphere of PWS (n = 16) is expanded compared with that of
the HCs (n = 16). Furthermore, PWS (n = 15) showed a significant
increase in the GM volume of the right superior temporal gyrus
compared with NSs (n = 15), consistent with the right tonotopic
expansion. Taken together, we suggest that PWS have impaired left
auditory sensory gating during basic auditory input processing and
that some error signals in the auditory cortex could result in abnor-
mal speech processing. Functional and structural reorganization of
the right auditory cortex appears to be a compensatory mechanism
for impaired left auditory cortex function in PWS.
5.2. Abnormal auditory synchronization in stuttering
We demonstrated both functional and structural reorganization
of the right auditory cortex and impaired left auditory cortex func-
tion in PWS (Kikuchi et al., 2011). It was interesting to explore the
underlying mechanism(s) of the functional alteration of the audi-
tory cortices in PDS. Therefore, we reevaluated the same dataset to
further investigate how the right and left auditory cortices interact
to compensate for stuttering (Kikuchi et al., 2017). We evaluated
bilateral N100 m latencies as well as indices of local and inter-
hemispheric phase synchronization of the auditory cortices.
The left N100 m latency was significantly prolonged relative to
the right N100 m latency in PWS, while healthy control partici-
pants did not show any inter-hemispheric differences in latency.
A PLF analysis, which indicates the degree of local phase syn-
chronization, demonstrated enhanced alpha-band synchrony in
the right auditory area of PWS (Fig. 5A). A PLV analysis of inter-
hemispheric synchronization demonstrated significant elevations
 S. Tobimatsu / Neuroscience Research 156 (2020) 117–129 123

Fig. 4. Grand-means for wPLIs in young (left column) and old (middle column) subjects, and point-by-point t-test results for wPLI (right column). Since the significant points
were distributed in a contiguous manner ranging from alpha to gamma band in both areas (right column), ROIs were defined as follows: 8–120 Hz and 0–200 ms for SI, and
8–60 Hz and 0–200 ms for cSII (red rectangles). Then, the group comparison using mean values of a contiguous region within the ROIs was performed. Only statistically
significant points (p < 0.05) are displayed in the t-maps. The old subjects show significantly elevated wPLIs in each comparison (with permission, from Hagiwara et al., 2,
Elsevier Science).

in the beta band between the right and left auditory cortices in
PWS (Fig. 5B). In addition, right PLF and PLVs were positively cor-
related with stuttering frequency in PWS. Taken together, our data
suggest that increased right hemispheric local phase synchroniza-
tion and increased inter-hemispheric phase synchronization are
electrophysiological correlates of a compensatory mechanism for
impaired left auditory processing in PWS.
6. Hippocampal modulation of auditory processing
The hippocampus is an integral component of the medial tempo-
ral lobe memory system and recently was thought to be involved in
sensory (particularly visual) perception (Graham et al., 2010). With
respect to auditory perception, the hippocampus receives afferent
inputs from the auditory association cortices, and in turn, it projects
back to the primary auditory cortex and auditory association areas
(Mehta et al., 2009). Within this reciprocal network, the hippocam-
pus has been shown to inhibit redundant auditory inputs, as well
as detect novel auditory information (Kraus and Canlon, 2012).
6.1. Effects of hippocampal sclerosis on central auditory
processing
It is likely that hippocampus plays a critical role in relaying
auditory information to later perceptual and cognitive processes.
However, little attention has been paid to the electrophysiologi-
cal signatures of hippocampal contribution to auditory processing.
Thus, the electrophysiological process by which unilateral hip-
pocampal lesions, such as hippocampal sclerosis (HS), modulate
the auditory processing remains unknown. Auditory-evoked mag-
netic fields (AEFs) are valuable for evaluating auditory functions,
because M100, a major component of AEFs, originates from audi-
tory areas. HS is the primary pathology of mesial temporal lobe
epilepsy (mTLE). Therefore, AEFs of mTLE (n = 17) with unilateral
HS were compared with HCs (n = 17) and disease controls (n = 9),
thereby determining whether AEFs were indicative of hippocampal
influences on the auditory processing (Chatani et al., 2016).
Monaural tone-burst stimuli were presented for each side,
followed by analysis of M100 and a previously less character-
ized exogenous component (M400: 300–500 ms). The frequency
of acceptable M100 dipoles was significantly decreased in the HS
side. Beam-forming-based source localization analysis also showed
decreased activity of the auditory area, which corresponded to the
inadequately estimated dipoles. M400 was found to be related to
the medial temporal structure on the HS side. Volumetric anal-
ysis was also performed, focusing on the auditory-related areas
(planum temporale, Heschl’s gyrus, and superior temporal gyrus),
as well as the hippocampus. M100 amplitudes positively corre-
lated with hippocampal and planum temporale volumes in the HC
group, whereas they negatively correlated with Heschl’s gyrus vol-
ume in the mTLE group. Interestingly, significantly enhanced M400
component was observed in the HS side of the mTLE patients. In
addition, the M400 component positively correlated with Heschl’s
gyrus volume and tended to positively correlate with disease dura-
124 S. Tobimatsu / Neuroscience Research 156 (2020) 117–129

Fig. 5. (A) PLFs in healthy control participants (HCs) versus people who stutter (PWS) in both hemispheres. Black rectangles indicate alpha band (8–13 Hz) synchronization
between 80 and 160 ms after stimulus onset. Auditory tone-burst stimulation of 1000 Hz was monaurally presented for 300 ms (10-ms and 20-ms rise and fall times,
respectively). Right local phase synchronization was significantly higher than left local phase synchronization in PWS when a repeated measure analysis of variance (ANOVA)
performed (p < 0.05). Right local phase synchronization was also higher than that of right local phase synchronization in HCs. Then, time frequency analysis was applied
to the auditory magnetic fields. (B) PLVs in HCs and PWS in 1000 Hz stimulation. White rectangles indicate beta band (14–30 Hz) synchronization between 80 and 160 ms
after stimulus onset. Synchronization between the right and left auditory cortices in PWS was significantly higher than that in HCs when a repeated measures ANOVA was
performed (p < 0.05) (with permission, from Kikuchi et al., 2017, Elsevier Science).

tion. M400 was markedly diminished after hippocampal resection.
Although volumetric analysis showed decreased hippocampal vol-
ume in the HS side, the planum temporale and Heschl’s gyrus, the
two major sources of M100, were preserved. These results sug-
gested that HS significantly influenced AEFs.
With respect to auditory-related dysfunctions, mTLE patients
exhibit deficits in precise detection of changes in frequency and
duration of auditory stimuli, despite intact primary hearing func-
tion (Han et al., 2011). In auditory-evoked potential studies, right
HS caused left lateralized topographical changes of N100 (Rosburg
et al., 2008), a major peak evoked 100 ms after the onset of audi-
tory stimulation, and reduced N100 amplitudes were also found in
mTLE (Bougeard and Fischer, 2002). Based on the results of previ-
ous works and our AEF findings, we conclud that the hippocampus
modulates auditory processing differently under normal conditions
and in HS.
 S. Tobimatsu / Neuroscience Research 156 (2020) 117–129 125

Fig. 6. (A) Time-frequency plots of PLVs in HCs, patients with left mTLE, and patients with right mTLE as grand-averaged data. White rectangles represent neural synchro-
nization between the right and left Heschl’s gyri in the theta, alpha, and beta bands, respectively. (B) PLVs in both patients with left and right mTLE were substantially reduced
compared with those of HCs in the theta and alpha frequency bands, irrespective of the side of ear stimulation. In the alpha band, PLVs in response to left ear stimulation
exhibited larger values compared with right ear stimulation in all groups (with permission, from Matsubara et al., 2018, Elsevier Science).

6.2. Effects of HS on auditory neural synchronization
Our previous study of monaural AEFs demonstrated that HS sig-
nificantly modulated auditory processing in patients with mTLE
(Chatani et al., 2016). However, the small sample size (n = 17) and
focus on the M100 response were insufficient to elucidate the later-
alization of the epileptic focus. Therefore, we increased the number
of patients with mTLE (n = 39) to examine whether neural syn-
chronization induced by monaural pure tone stimulation provides
useful diagnostic information about epileptic foci in patients with
unilateral mTLE (Matsubara et al., 2018).
Twenty-five patients with left mTLE, 14 patients with right
mTLE, and 32 HCs were recruited. Auditory stimuli of 500-Hz tone
burst were monaurally presented to subjects. The AEF data were
analyzed with source estimation of M100 responses in bilateral
auditory cortices (ACs). Neural synchronization within ACs and
between ACs was evaluated with PLF and PLV, respectively. Linear
discriminant analysis was performed for diagnosis and lateraliza-
tion of epileptic focus. The M100 amplitude revealed that patients
with right mTLE exhibited smaller M100 amplitude than patients
with left mTLE and HCs. Interestingly, PLF was able to differenti-
ate the groups with mTLE, with decreased PLFs in the alpha band
observed in patients with right mTLE compared with those (PLFs)
in patients with left mTLE (data not shown). Right hemispheric pre-
dominance was confirmed in both HCs and patients with left mTLE
while patients with right mTLE showed a lack of right hemispheric
predominance. Functional connectivity between bilateral ACs (PLV)
was reduced in both patients with right and left mTLE compared
with that of HCs (Fig. 6). The accuracy of diagnosis and lateralization
was 80 %–90 %.
We confirmed that auditory processing was disrupted in
patients with mTLE (Chatani et al., 2016). Therefore, the current
results indicated that monaural AEFs provide useful information for
detecting mTLE pathologies. In addition, we found that AC subnor-
mal function was more pronounced in patients with right mTLE.
Neural synchronization highlighted lateralization of the epileptic
focus in patients with unilateral mTLE. Our non-invasive simple
task was useful for localizing the epileptic focus in patients with
mTLE, as well as for diagnosing patients with mTLE. Finally, our
findings suggest that altered neural synchronization is exhibited
by patients with mTLE, providing evidence for altered AC function
in mTLE.
6.3. Effects of HS on auditory gamma band oscillations
Patients with mTLE often exhibit central auditory processing
(CAP) dysfunction. Various studies have reported that some mTLE
patients show decreased performance in irregularity discrimina-
tion of rapid auditory sequences (Ehrle et al., 2001), decreased
temporal processing in the Gaps-In-Noise test (Aravindkumar et al.,
2012) and the duration pattern sequence test (Lavasani et al., 2016).
In accord with these results, a recent MEG study in our laboratory
provided neuromagnetic evidence for hippocampal modulation of
CAP (Chatani et al., 2016; Matsubara et al., 2018). However, the
126 S. Tobimatsu / Neuroscience Research 156 (2020) 117–129

Fig. 7. Induced gamma responses of PLFs in HCs, left mTLE and right mTLE patients obtained from contralateral and ipsilateral ACs. Time-frequency plots are shown in three
groups as grand-averaged data (upper row). 40-Hz ASSR elicits temporal frequency dynamics at approximately 40 Hz in bilateral ACs irrespective of ear stimulation during
sound presentation. Note that more prominent responses in the contralateral AC were evident, compared with those of ipsilateral AC. Time courses of PLFs as the average of
35–45 Hz oscillations are shown in each group by each ear stimulation (middle row). Black lines represent the mean activity of contralateral AC, while gray lines represent
the mean activity of ipsilateral AC. More prominent responses were found in the contralateral AC in HCs, while mTLE patients showed less marked contralaterality. Contra-
laterality index (cLI) scores are presented for each group (lower row). HCs (blue lines) exhibited symmetrical contralaterality in response to left and right ear stimulation
during all time courses, while mTLE patients exhibited different response patterns. Left mTLE patients (red lines) exhibited decreased contralaterality in response to right
ear stimulation, whereas right mTLE patients (green lines) exhibited reduced contralaterality in either right or left ear stimulation, with more prominent reductions in left
ear stimulation. Shaded areas represent the standard error of the mean (with permission, from Matsubara et al., 2019, Elsevier Science).

functional impact of CAP dysfunction on unilateral HS remains to
be fully elucidated. Monaural 40-Hz auditory steady-state mag-
netic responses (ASSRs) were recorded to study whether gamma
band neural synchronization induced by 40-Hz monaural stimula-
tion provides useful information on the pathophysiology of mTLE
(Matsubara et al., 2019). Eighteen left mTLE patients, 11 right mTLE
patients and 16 HCs were examined. Monaural clicks were pre-
sented at a rate of 40 Hz. PLF and power values were analyzed
within bilateral Heschl’s gyri. Monaural 40-Hz ASSR demonstrated
temporal frequency dynamics in both PLF and power data. Symmet-
rical hemispheric contralaterality was revealed in HCs. However,
predominant contralaterality was absent in mTLE patients (Fig. 7).
Specifically, right mTLE patients exhibited a lack of contralaterality
in response to left ear but not right ear stimulation, and vice versa
in left mTLE patients.
This is the first study to use monaural 40-Hz ASSR with unilateral
mTLE patients to clarify the relationship between CAP and epileptic
focus. CAP dysfunction was characterized by a lack of contralater-
ality corresponding to epileptic focus. Monaural 40-Hz ASSR can
provide useful information for localizing epileptic focus in mTLE
patients. These results link neuromagnetic assessment and differ-
ences in clinical findings of cognitive processing (i.e., CAP) between
HCs and mTLE patients, and also between left mTLE patients and
right mTLE patients. Overall, our results indicate that CAP dysfunc-
tion revealed by gamma band oscillations can be used to reveal the
pathophysiology of mTLE.
7. Pain perception and neural oscillations
Nociceptive stimuli induce responses in an extensive cortical
network including mainly SI and SII somatosensory, insular and
anterior cingulate areas. The activity of this network, often referred
to as the “pain matrix”, is thought to reflect the mechanisms by
which a nociceptive input is transformed into a conscious per-
cept of pain (for a review, see Tracey and Mantyh, 2007). The SI
and SII cortices have been variably implicated in acute nociception.
Several neuroimaging studies have demonstrated that the insular-
opercular region (IOR) is consistently involved in pain processing
(Apkarian et al., 2005; Peyron et al., 2002). Some reports have also
suggested that this region encodes quantitative as well as quali-
tative aspects of pain (Oertel et al., 2012). Furthermore, the IOR is
known to be involved in multisensory processing, including tac-
tile sensation, pain, proprioception, temperature, and vestibular
information (Mazzola et al., 2012). The anterior cingulate may be
concerned with the affective and attentional components of pain
experience, and experimental evidence lends some support to this
concept (Rainville et al., 1997).
 S. Tobimatsu / Neuroscience Research 156 (2020) 117–129 127

Fig. 8. t-Values (needle condition vs. cotton swab condition) for the ␥ band. (A) A significant cluster of significant electrodes is outlined by a black oval line. Induced ␥
band activities (30–60 Hz) were increased in the needle condition over the right hemisphere. (B) Power values and t-values of a representative electrode. The degree of
power suppression was decreased between 1000 and 1500 ms after movie onset. (C) A significant cluster was observed over the right centroparietal region, as indicated by
a multiplication sign (×). This was confirmed by a cluster-based permutation test. (D) Source estimation of the time and frequency window on the standard brain indicates
that the decreased ␥ band suppression is centered at the right motor area (with permission, from Motoyama et al., 2017, Dove Press).

7.1. Cortical pain relief by tactile stimulation
The analgesic effect of vibrotactile stimuli has been acknowl-
edged as an important mechanism that can modulate pain
perception (Melzack and Wall, 1965). Practically, people uncon-
sciously rub a painful part of their body to reduce pain. Several
studies have shown the analgesic effect of both tactile and vibra-
tory stimuli in psychophysiological evaluation in healthy subjects
(Mancini et al., 2014) and in chronic pain conditions (Staud et al.,
2011). Because these studies presented nociceptive and non-
nociceptive stimuli simultaneously, cross-modal interaction most
likely occurred in the spinal cord. However, Inui et al. (2006)
demonstrated a supraspinal mechanism in tactile-induced pain
relief via experimental paradigms with concurrent bimodal inputs
at the cortical level. However, the precise spatiotemporal profile for
centrally mediated tactile-induced analgesia is largely unknown.
The IOR integrates multiple sensory inputs, and nociceptive modu-
lation by other sensory inputs occurs in this area. In this study, we
focused on the IOR to characterize the spatiotemporal signature of
tactile-induced analgesia using MEG (Hayamizu et al., 2016).
Eleven healthy right-handed volunteers participated in this
study. A␦ (intra-epidermal electrical stimulation) inputs were
modified by A␤ (mechanical tactile stimulation) selective stimu-
lation (Inui et al., 2002), either independently or concurrently, to
the right forearm. The optimal inter-stimulus interval (ISI) for corti-
cal level modulation was determined after comparing the 40-, 60-,
and 80-ms ISI conditions, and the calculated cortical arrival time
difference between A␦ and A␤ inputs. Subsequently, we adopted
a 60-ms ISI for cortical modulation and a 0-ms ISI for spinal level
modulation. Source localization using minimum norm estimates
demonstrated that pain-related activity was located in the poste-
rior insula, whereas tactile-related activity was estimated in the
parietal operculum. We also found significant inhibition of pain-
related activity in the posterior insula due to cortical modulation.
In contrast, spinal modulation was observed both in the poste-
rior insula and parietal operculum. Subjective pain, as evaluated by
the visual analog scale, also showed significant reduction in both
conditions.
Pain–tactile interaction at the cortical level was investigated
using carefully timed sensory modality-specific stimuli, in com-
bination with cortical source-based neuromagnetic analysis. We
were able to clearly distinguish source location from pain and tac-
tile inputs associated with natural-touch like sensation. Our study
clearly demonstrates that the spatial feature of cortical modulation
by nociceptive input was different compared with spinal modula-
tion. The current study also provides new spatiotemporal evidence
that the posterior insula is a critical area of the cortical mechanisms
underlying tactile-induced analgesia in humans.
7.2. Frequency-dependent changes in sensorimotor and pain
affective systems induced by empathy for pain
Understanding or having empathy for others while observ-
ing their circumstances is crucial for making fundamental
social ties. Witnessing another person in a painful situation
can cause “empathy for pain” in the observer. Empathy for
128 S. Tobimatsu / Neuroscience Research 156 (2020) 117–129
pain involves complex processes that make it possible for the
observer to experience sensory and emotional qualities of vicar-
ious pain (Betti and Aglioti, 2016). In addition, empathy for pain
induces not only affective–motivational representations, but also
sensory–discriminatory representations in the brain (Avenanti
et al., 2005). Thus, we can use empathic pain to explore the pain
system without inducing actual pain.
To better comprehend the processing of empathic pain, we stud-
ied the frequency-dependent modulation of cortical oscillations
induced by watching movies depicting pain using high-density EEG,
MEG, and motor evoked potentials (MEP) (Motoyama et al., 2017).
Experiment 1 (17 normal subjects) was designed to explore the
relationship between neuronal oscillations and subjective inner
pain. Experiment 2 (8 normal subjects) was designed to study
the effect of empathy for pain on M1 excitability contralateral
to the pricked hand. Event-related desynchronization of EEG and
MEG was assessed while participants viewed videos of painful
(needle) or neutral (cotton swab) situations. The amplitudes of
MEPs were also compared between the needle and cotton swab
conditions. The degree of suppression in ␣/␤ band power was sig-
nificantly increased, whereas that of ␥ band power was significantly
decreased, in the needle condition compared with the cotton swab
condition. EEG revealed that significant differences in ␣/␤ band
were distributed in the right frontocentral and left parietooccipital
regions, whereas significant ␥ band differences were distributed
predominantly over the right hemisphere, which were confirmed
by source estimation using MEG (Fig. 8). There was a significant
positive correlation between the difference in ␥ power of the two
conditions and the visual analog scale subjective rating of aver-
sion, but not in the ␣/␤ band. The amplitude of MEPs decreased in
the needle condition, which confirmed the inhibition of the pri-
mary motor cortex. MEP suppression supports that modulation
of cortical oscillations by viewing movies depicting pain involves
sensorimotor processing.
We explored the frequency-dependent changes of empathy for
pain by viewing movies depicting pain using high-density EEG and
MEG with the condition that MEP inhibition was observed. Our
results suggest that ␣ /␤ band suppression is involved in the pro-
cessing of somatic aspects of empathy for pain, which in turn results
in inhibition of M1 excitability. ␥ oscillations most likely reflect the
cognitive aspects of pain assessment (Chen and Herrmann, 2001;
Gross et al., 2007; Hauck et al., 2007). The analyses help advance our
understanding of the processing of empathic pain. Further studies
are needed to study the direct link between ␣ /␤ band suppres-
sion and MEP suppression, and the relationship between ␣ /␤ and
␥ band modulations, to further elucidate the neural substrates of
pain observation.
8. Conclusion and future perspectives
It is apparent that neuronal synchronization plays an impor-
tant role in distributed cortico-cortical processing in the sensory
systems in human. Thus, the induced oscillatory activities are
responsible for temporal binding of spatially distributed informa-
tion processing in the human brain. Future studies will be necessary
to facilitate our understanding of the roles of the neural oscillations
in the network interactions.
Declaration of Competing Interest
The authors declare that they have no competing interests.
Acknowledgement
This review article was supported in part by JSPS KAKENHI Grant
Number 15H05875.
References
Apkarian, A.V., Bushnell, M.C., Treede, R.D., Zubieta, J.K., 2005. Human brain mech-
anisms of pain perception and regulation in health and disease. Eur. J. Pain 9,
463–484.
Aravindkumar, R., Shivashankar, N., Satishchandra, P., Sinha, S., Saini, J., Sub-
bakrishna, D.K., 2012. Temporal resolution deficits in patients with refractory
complex partial seizures and mesial temporal sclerosis (MTS). Epilepsy Behav.
24, 126–130, http://dx.doi.org/10.1016/j.yebeh.2012.03.004.
Avenanti, A., Bueti, D., Galati, G., Aglioti, S.M., 2005. Transcranial magnetic stimu-
lation highlights the sensorimotor side of empathy for pain. Nat. Neurosci. 8,
955–960.
Betti, V., Aglioti, S.M., 2016. Dynamic construction of the neural networks underpin-
ning empathy for pain. Neurosci. Biobehav. Rev. 63, 191–206, http://dx.doi.org/
10.1016/j.neubiorev.2016.02.009.
Bougeard, R., Fischer, C., 2002. The role of temporal pole in auditory processing.
Epileptic Disord. 4, 29–32.
Chatani, H., Hagiwara, K., Hironaga, N., Ogata, K., Shigeto, H., Morioka, T., Sakata,
A., Hashiguchi, K., Murakami, N., Uehara, T., Kira, J.I., Tobimatsu, S., 2016.
Neuromagnetic evidence for hippocampal modulation of auditory process-
ing. NeuroImage 124, 256–266, http://dx.doi.org/10.1016/j.neuroimage.2015.
09.006.
Chen, A.C., Herrmann, C.S., 2001. Perception of pain coincides with the spatial expan-
sion of electroencephalographic dynamics in human subjects. Neurosci. Lett.
297, 183–186, http://dx.doi.org/10.1016/S0304-3940(00)01696-7.
Debener, S., Herrmann, C.S., Kranczioch, C., Gembris, D., Engel, A.K., 2003. Top-down
attentional processing enhances auditory evoked gamma band activity. Neu-
roreport 14, 683–686.
Ehrle, N., Samson, S., Baulac, M., 2001. Processing of rapid auditory information
in epileptic patients with left temporal lobe damage. Neuropsychologia 39,
525–531, http://dx.doi.org/10.1016/S0028-3932(00)00121-4.
Engel, A.K., Fries, P., Singer, W., 2001. Dynamic predictions: oscillations and Syn-
chrony in top-down processing. Nat. Rev. Neurosci. 2, 704–716.
Fries, P., Reynolds, J., Rorie, A., Desimone, R., 2001. Modulation of oscillatory neuronal
synchronization by selective visual attention. Science 291, 1560–1563, http://
dx.doi.org/10.1126/science.1055465.
Graham, K.S., Barense, M.D., Lee, A.C.H., 2010. Going beyond LTM in the MTL:
a synthesis of neuropsychological and neuroimaging findings on the role of
the medial temporal lobe in memory and perception. Neuropsychologia 48,
831–853, http://dx.doi.org/10.1016/j.neuropsychologia.2010.01.001.
Gross, J., Schnitzler, A., Timmermann, L., Ploner, M., 2007. Gamma oscillations in
human primary somatosensory cortex reflect pain perception. PLoS Biol. 5 (5),
e133, http://dx.doi.org/10.1371/journal.pbio.0050133.
Hagiwara, K., Okamoto, T., Shigeto, H., Ogata, K., Somehara, Y., Matsushita, T., Kira,
J.-I., Tobimatsu, S., 2010. Oscillatory gamma synchronization binds the pri-
mary and secondary somatosensory areas in humans. NeuroImage 51, 412–420,
http://dx.doi.org/10.1016/j.neuroimage.2010.02.001.
Hagiwara, K., Ogata, K., Okamoto, T., Uehara, T., Hironaga, N., Shigeto, H., Kira, J.-
I., Tobimatsu, S., 2014. Age-related changes across the primary and secondary
somatosensory areas: an analysis of neuromagnetic oscillatory activities. Clin.
Neurophysiol. 125, 1021–1029, http://dx.doi.org/10.1016/j.clinph.2013.10.005.
Han, M.W., Ahn, J.H., Kang, J.K., Lee, E.M., Lee, J.H., Bae, J.H., Chung, J.W., 2011. Cen-
tral auditory processing impairment in patients with temporal lobe epilepsy.
Epilepsy Behav. 20, 370–374, http://dx.doi.org/10.1016/j.yebeh.2010.12.032.
Hauck, M., Lorenz, J., Engel, A.K., 2007. Attention to painful stimulation enhances
gamma-band activity and synchronization in human sensorimotor cortex. J.
Neurosci. 27, 9270–9277, http://dx.doi.org/10.1523/JNEUROSCI.2283-07.
Hari, R., Baillet, S., Barnes, G., Burgess, R., Forss, N., Gross, J., Hämäläinen, M., Jensen,
O., Kakigi, R., Mauguière, F., Nakasato, N., Puce, A., Romani, G.L., Schnitzler,
A., Taulu, S., 2018. IFCN-endorsed practical guidelines for clinical magnetoen-
cephalography (MEG). Clin. Neurophysiol. 129, 1720–1747, http://dx.doi.org/
10.1016/j.clinph.2018.03.042.
Hayamizu, M., Hagiwara, K., Hironaga, N., Ogata, K., Hoka, S., Tobimatsu, S., 2016.
A spatiotemporal signature of cortical pain relief by tactile stimulation: an
MEG study. NeuroImage 130, 175–183, http://dx.doi.org/10.1016/j.neuroimage.
2016.01.065.
Huttunen, J., Wikstrom, H., Salonen, O., Ilmoniemi, R.J., 1999. Human somatosen-
sory cortical activation strengths: comparison between males and females and
age-related changes. Brain Res. 818, 196–203, http://dx.doi.org/10.1016/S0006-
8993(98)01215-3.
Inui, K., Tran, T.D., Hoshiyama, M., Kakigi, R., 2002. Preferential stimulation of A␦
fibers by intra-epidermal needle electrode in humans. Pain 96, 247–252, http://
dx.doi.org/10.1016/S0304-3959(01)00453-5.
 S. Tobimatsu / Neuroscience Research 156 (2020) 117–129
Inui, K., Tsuji, T., Kakigi, R., 2006. Temporal analysis of cortical mechanisms for pain
relief by tactile stimuli in humans. Cereb. Cortex 16, 355–365, http://dx.doi.org/
10.1016/S0304-3959(01)00453-5.
Iwamura, Y., 1998. Hierarchical somatosensory processing. Curr. Opin. Neurobiol. 8,
522–528, http://dx.doi.org/10.1016/S0959-4388(98)80041-X.
Kikuchi, Y., Ogata, K., Umesaki, T., Yoshiura, T., Kenjo, M., Hirano, Y., Okamoto, T.,
Komune, S., Tobimatsu, S., 2011. Spatiotemporal signatures of an abnormal audi-
tory system in stuttering. NeuroImage 55, 891–899, http://dx.doi.org/10.1016/
j.neuroimage.2010.12.083.
Kikuchi, Y., Okamoto, T., Ogata, K., Hagiwara, K., Umezaki, T., Kenjo, M., Nakagawa,
T., Tobimatsu, S., 2017. Abnormal auditory synchronization in stuttering: a mag-
netoencephalographic study. Hear. Res. 344, 82–89, http://dx.doi.org/10.1016/
j.heares.2016.10.027.
Knight, R.T., 2007. Neural networks debunk phrenology. Science 316, 1578–1579,
http://dx.doi.org/10.1126/science.1144677.
Kraus, K.S., Canlon, B., 2012. Neuronal connectivity and interactions between the
auditory and limbic systems: effects of noise and tinnitus. Hear. Res. 288, 34–46,
http://dx.doi.org/10.1016/j.heares.2012.02.009.
Lachaux, J.P., Rodriguez, E., Martinerie, J., Varela, F.J., 1999. Measuring phase syn-
chrony in brain signals. Hum. Brain Mapp. 8, 194–208, http://dx.doi.org/10.
1002/(SICI)1097-0193(1999)8:4<194::AID-HBM4>3.0.CO;2-C.
Lavasani, A.N., Mohammadkhani, G., Motamedi, M., Karimi, L.J., Jalaei, S., Shojaei,
F., Danesh, A., Azimi, H., 2016. Auditory temporal processing in patients with
temporal lobe epilepsy. Epilepsy Behav. 60, 81–85, https//j.yebeh.2016.04.017.
Lin, Y.Y., Forss, N., 2002. Functional characterization of human second somatosen-
sory cortex by magnetoencephalography. Behav. Brain Res. 135, 141–145, http://
dx.doi.org/10.1016/S0166-4328(02)00143-2.
Lincoln, M., Packman, A., Onslow, M., 2006. Altered auditory feedback and the treat-
ment of stuttering: a review. J. Fluency Disord. 31, 71–89, http://dx.doi.org/10.
1016/j.jfludis.2006.04.001.
Matsubara, T., Ogata, K., Hironaga, N., Kikuchi, Y., Uehara, T., Chatani, H., Mitsudo, T.,
Shigeto, H., Tobimatsu, S., 2018. Altered neural synchronization to pure tone
stimulation in patients with mesial temporal lobe epilepsy: an MEG study.
Epilepsy Behav. 88, 96–105 https://10.1016/j.yebeh.2018.08.036.
Matsubara, T., Ogata, K., Hironaga, N., Uehara, T., Mitsudo, T., Shigeto, H., Maekawa, T.,
Tobimatsu, S., 2019. Monaural 40-Hz auditory steady-state magnetic responses
can be useful for identifying epileptic focus in mesial temporal lobe epilepsy.
Clin. Neurophysiol. 130, 341–351, http://dx.doi.org/10.1016/j.clinph.2018.11.
026.
Mancini, F., Nash, T., Iannetti, G.D., Haggard, P., 2014. Pain relief by touch: a quan-
titative approach. Pain 155, 635–642, http://dx.doi.org/10.1016/j.pain.2013.12.
024.
Mazzola, L., Faillenot, I., Barral, F.G., Mauguière, F., Peyron, R., 2012. Spatial segre-
gation of somatosensory and pain activations in the human operculo-insular
cortex. NeuroImage 60, 409–418, http://dx.doi.org/10.1016/j.neuroimage.2011.
12.072.
Mehta, A.D., Ettinger, A.B., Perrine, K., Dhawan, V., Patil, A., Jain, S.K., Klein, G., Schnei-
der, S.J., Eidelberg, D., 2009. Seizure propagation in a patient with musicogenic
epilepsy. Epilepsy Behav. 14, 421–424, http://dx.doi.org/10.1016/j.yebeh.2008.
11.010.
Melzack, R., Wall, P.D., 1965. Pain mechanisms: a new theory. Science 150, 971–979,
http://dx.doi.org/10.1126/science.150.3699.971.
Motoyama, Y., Ogata, K., Hoka, S., Tobimatsu, S., 2017. Frequency-dependent changes
in sensorimotor and pain affective systems induced by empathy for pain. J. Pain
Res. 29 (May 10), 1317–1326, eCollection 2017 https://10.2147/JPR.S129791.
Nunez, P.L., Nunez, M.D., Srinivasan, R., 2019. Multi-scale neural sources of EEG:
genuine, equivalent, and representative. A tutorial review. Brain Topogr. 32,
193–214, http://dx.doi.org/10.1007/s10548-019-00701-3.
Oertel, B.G., Preibisch, C., Martin, T., Walter, C., Gamer, M., Deichmann, R., Lötsch, J.,
2012. Separating brain processing of pain from that of stimulus intensity. Hum.
Brain Mapp. 33, 883–894, http://dx.doi.org/10.1002/hbm.21256.
129
Peyron, R., Frot, M., Schneider, F., Garcia-Larrea, L., Mertens, P., Barral, F.G., Sin-
dou, M., Laurent, B., Mauguière, F., 2002. Role of operculoinsular cortices in
human pain processing: converging evidence from PET, fMRI, dipole modeling,
and intracerebral recordings of evoked potentials. NeuroImage 17, 1336–1346,
http://dx.doi.org/10.1006/nimg.2002.1315.
Polman, C.H., Reingold, S.C., Edan, G., Filippi, M., Hartung, H.P., Kappos, L., Lublin, F.D.,
Metz, L.M., McFarland, H.F., O’Connor, P.W., Sandberg-Wollheim, M., Thomp-
son, A.J., Weinshenker, B.G., Wolinsky, J.S., 2005. Diagnostic criteria for multiple
sclerosis: 2005 revisions to the “McDonald Criteria”. Ann. Neurol. 58, 840–846,
http://dx.doi.org/10.1002/ana.20703.
Postma, A., Kolk, H., 1992. Error monitoring in people who stutter: evidence against
auditory feedback defect theories. J. Speech Hear. Res. 35, 1024–1032, http://dx.
doi.org/10.1044/jshr.3505.1024.
Rainville, P., Duncan, G.H., Price, D.D., Carrier, B., Bushnell, M.C., 1997. Pain affect
encoded in human anterior cingulate but not somatosensory cortex. Science
277, 968–971, http://dx.doi.org/10.1126/science.277.5328.968.
Roach, B.J., Mathalon, D.H., 2008. Event-related EEG time-frequency analysis: an
overview of measures and an analysis of early gamma band phase locking in
schizophrenia. Schizophr. Bull. 34, 907–926, http://dx.doi.org/10.1093/schbul/
sbn093.
Rosburg, T., Trautner, P., Ludowig, E., Helmstaedter, C., Bien, C.G., Elger, C.E., Boutros,
N.N., 2008. Sensory gating in epilepsy: effects of the lateralization of hippocam-
pal sclerosis. Clin. Neurophysiol. 119, 1310–1319, http://dx.doi.org/10.1016/j.
clinph.2008.02.007.
Spencer, K.M., Niznikiewicz, M.A., Shenton, M.E., McCarley, R.W., 2007. Sensory-
evoked gamma oscillations in chronic schizophrenia. Biol. Psychiatry 63,
744–747, http://dx.doi.org/10.1016/j.biopsych.2007.10.017.
Staud, R., Robinson, M.E., Goldman, C.T., Price, D.D., 2011. Attenuation of exper-
imental pain by vibro-tactile stimulation in patients with chronic local or
widespread musculoskeletal pain. Eur. J. Pain 15, 836–842, http://dx.doi.org/
10.1016/j.ejpain.2011.01.011.
Stephen, J.M., Ranken, D., Best, E., Adair, J., Knoefel, J., Kovacevic, S., 2006. Aging
changes and gender differences in response to median nerve stimulation mea-
sured with MEG. Clin. Neurophysiol. 117, 131–143, http://dx.doi.org/10.1016/j.
clinph.2005.09.003.
Tallon-Baudry, C., Bertrand, O., Delpuech, C., Pernier, J., 1996. Stimulus specificity of
phase-locked and non-phase-locked 40 Hz visual responses in human. J. Neu-
rosci. 16, 4240–4249, http://dx.doi.org/10.1523/JNEUROSCI.16-13-04240.
Tallon-Baudry, C., Bertrand, O., Delpuech, C., Permier, J., 1997. Oscillatory gamma-
band (30-70 Hz) activity induced by a visual search task in humans. J. Neurosci.
17, 722–734.
Tallon-Baudry, C., Bertrand, O., 1999. Oscillatory gamma activity in humans and its
role in object representation. Trends Cogn. Sci. 3, 151–162, http://dx.doi.org/10.
1016/S1364-6613(99)01299-1.
Tecchio, F., Zito, G., Zappasodi, F., Dell’ Acqua, M.L., Landi, D., Nardo, D., Lupoi, D.,
Rossini, P.M., Filippi, M.M., 2008. Intra-cortical connectivity in multiple sclero-
sis: a neurophysiological approach. Brain 131, 1783–1792, http://dx.doi.org/10.
1093/brain/awn087.
Tobimatsu, S., Kakigi, R., 2016. Clinical Applications of Magnetoencephalography.
Springer, Germany.
Tracey, I., Mantyh, P.W., 2007. The cerebral signature for pain perception and its
modulation. Neuron 55, 377–391, http://dx.doi.org/10.1016/j.neuron.2007.07.
012.
Uhlhaas, P.J., Singer, W., 2015. Oscillations and neuronal dynamics in schizophrenia:
the search for basic symptoms and translational opportunities. Biol. Psychiatry
77, 1001–1009, http://dx.doi.org/10.1016/j.biopsych.2014.11.019.
Vinck, M., Oostenveld, R., van Wingerden, M., Battaglia, F., Pennartz, C.M., 2011. An
improved index of phase-synchronization for electrophysiological data in the
presence of volume-conduction, noise and sample-size bias. NeuroImage 55,
1548–1565, http://dx.doi.org/10.1016/j.neuroimage.2011.01.055.