Electroencephalography and clinical Neurophysiology 107 (1998) 287295

Perception of motion and qEEG activity in human adults

S. Cochin a ,*, C. Barthelemy a, B. Lejeune b, S. Roux a, J. Martineau a
a

INSERM U316, Service Universitaire dExplorations Fonctionnelles et Neurophysiologie en Pedopsychiatrie, CHU Bretonneau,
2 Bld. Tonnelle, 37044 Tours cedex, France
b
Laboratoire de Biophysique et Mathematiques, UFR Sciences Pharmaceutiques, 31 avenue Monge, 37200 Tours, France
Accepted for publication: 14 April 1998

Abstract
This study was designed to relate visual perception of motion to cortical activity, by evaluation of the association of quantified
electroencephalogram (qEEG) parameters with a video film projection. The EEG was recorded from 14 sites according to the International
1020 system and a common average reference was used. Forty right-handed volunteers (mean age = 24 years) were examined. The video
film consisted of 20 s sequences showing still shots and moving shots with human movements or object movements. The EEG was then
subjected to spectral analysis; the spectral powers for the theta, alpha and beta bands were calculated for 14 s epochs and compared with
sequences of the video film. All analyses were based on logarithmically transformed absolute spectral power values. The power values of
each frequency band were analysed in a 3-way repeated measure ANOVA (Hemisphere Electrode Sequence). The results were
represented by EEG cartography. Significant decreases in the alpha 1, beta 1 and beta 2 power values of EEG in centro-parietal regions
of both hemispheres were shown during perception of human motion sequences. This suggests participation of the sensorimotor cortex
during visual observation of human motion. 1998 Elsevier Science Ireland Ltd. All rights reserved
Keywords: Quantified EEG; Movement; Alpha rhythm; Mu rhythm; Beta rhythm; Human

1. Introduction
The EEG changes occurring during various mental activities in man have been the subject of many investigations
since the beginning of EEG research. In the visual examination of EEGs, most attention has been directed to the
changes in the alpha and beta ranges.
Most studies related to task-specific correlates in the EEG
have concentrated on the alpha band, the range of frequencies between 8 and 13 Hz (Butler and Glass, 1976; Shaw et
al., 1977; Beaumont et al., 1978). Generally, the magnitude
of the alpha band component of the EEG is diminished in
both cerebral hemispheres with any deviation from relaxed
wakefulness toward mental activity. Meanwhile, frequency
components in the beta band (1330 Hz) are thought to
increase in magnitude with increasing mental effort
(Dolce and Waldeier, 1974; Papanicolaou et al., 1986).

* Corresponding author. Tel.: +33 02 47474747, ext. 3-4305; fax: +33 02

47473846.

Many studies have shown that the same cerebral areas are
activated for both motor planning and motor imagery (Lassen and Ingvar, 1990; Beisteiner et al., 1995). Studies of
brain activity during mentally simulated movement were
first published by Ingvar and Philipson (1977), using regional cerebral blood flow measurements and showing that
there was an increase in frontal lobe activity, whereas
overt execution also led to activation of the rolandic area.
Decety and Ingvar (1990) confirmed this early finding by
positron emission tomography (PET) and they found, in
addition, a bilateral increase in metabolic activity in the
cerebellum during mentally simulated movements.
Finally, an increase in excitability of the motor system
has been described when a subject observes an action performed by another individual. Thus, the motor evoked
potentials (MEPs) significantly increased during the conditions in which subjects observed movements. The MEP
pattern reflected the pattern of muscle activity recorded
when the subjects executed the observed actions (Fadiga
et al., 1995). These authors concluded that in humans
there is a system matching action observation and execu-

0013-4694/98/$ - see front matter 1998 Elsevier Science Ireland Ltd. All rights reserved
PII S0013-4694 (98 )0 0071-6

EEG 97694

288

S. Cochin et al. / Electroencephalography and clinical Neurophysiology 107 (1998) 287295

tion. A similar system was previously described in the

monkey, using recordings of single neurons with microelectrodes (di Pellegrino et al., 1992). These authors demonstrated that the observation of an action automatically
recruits neurons that would normally be active when the
subject executes that action.
In 1954, Cohen-Seat and colleagues studied EEG modifications during film projections (Cohen-Seat et al., 1954).
EEG recordings of 80 normal adults showed that motor
exercises executed by an actor on the screen blocked spectators mu rhythm. They concluded that the spectator identified with the actor and this phenomenon was termed
posturo-motor induction. Mu-blocking associated with
the viewing of human motor exercises on a screen has
been the subject of many communications since this pioneer
study (Cohen-Seat et al., 1956; Heuyer et al., 1957; Lelord,
1957, 1966, 1984). Visual interpretation of modifications of
EEG revealed disappearance of the 10 c/s alpha rhythm,
especially in the central area of the scalp.
The aim of this study was to study relationships between
visual perception of motion and quantified cortical activity,
by evaluation of the association of qEEG parameters (EEG
quantification and EEG cartography) with a video film projection.

2. Subjects and methods

were asked to remain as still as possible during all recording

periods.
After receiving instructions concerning the experimental
tasks, each subjects EEG was recorded at the same time of
day during video viewing. EEG recordings lasting 14 s were
made during each sequence, with intervals of 2 s, and each
experimental session lasted 12 min.
2.3. Apparatus
The EEG was recorded with a common average reference
from 14 Ag/AgCl electrodes (F3, F4, F7, F8, C3, C4, P3, P4,
T3, T4, T5, T6, O1, O2) located according to the International 1020 system (Jasper, 1958) with a B.E.S.T. system
(Fa. Grosseger). The ground electrode was positioned on the
forehead. Vertical EOG was measured with electrodes 2 cm
above and below the middle of the left eye, and horizontal
EOG electrodes were placed as close as possible to the left
and right outer canthi of the eyes. Amplifier bandwidth was
set between 0.5 and 35 Hz. Impedance was kept below 10
kQ in all cases. The EEG was sampled at 256 Hz and stored
on a hard disk for further analysis. The stored raw data were
first analysed visually off-line to eliminate EEG segments
contaminated with ocular or muscular artefacts. Automatic
EOG-correction was then performed on EEG data using
regression analysis (Anderer et al., 1989). Automatic acquisition of the EEG signal for each sequence of the video was
triggered electronically.

2.1. Subjects
2.4. Analysis
The study was performed on 20 male and 20 female
healthy student volunteers between 18 and 33 years of age
(mean age = 24 years). Informed consent was obtained for
each subject before beginning the testing and the paradigm
was approved by the INSERM Scientific Council. Subjects
were all right-handed according to self-report, with no history of neurological disease and with a normal EEG. They
were seated in front of a video monitor in a comfortable
armchair in a darkened sound-proofed room.
2.2. Experimental procedure
Each subject watched a video during EEG recording. The
silent movie consisted of 31 sequences, each lasting 20 s.
The video was filmed by the VTCOM society. Still shots
(e.g. landscape, castle) alternated with moving sequences
consisting of objects in movement, still and moving cartoons, animals in motion and gymnastic movements executed by one person. Five sequences were analysed:
white (W) (TV screen white), waterfall (WF) (falling
water), lake (L) (a body of water surrounded by land),
and two human moving sequences comprising flexion
(F) during which a woman was bending her legs, and rotating (R) during which the same woman performed scissor
movements with her legs while lying on her back. The video
monitor was placed 150 cm from the subject. The subjects

2.4.1. Data processing

Fast Fourier transform (Cooley and Tukey, 1965) was
computed in successive 2 s epochs for each sequence and
a measurement was obtained for each monopolar electrode
as follows: absolute power (AP) in each of the 7 frequency
bands: theta 1 (35.5 Hz), theta 2 (5.57.5 Hz), alpha 1
(7.510.5 Hz), alpha 2 (10.513 Hz), beta 1 (1318 Hz),
beta 2 (1825 Hz) and beta 3 (2535 Hz). The logarithm
transformation for AP values (Ln AP) was obtained, in order
to achieve Gaussian distributions for statistical analysis.
The 5 sequences (W, WF, L, F, R) in each frequency band
were compared using 3-way ANOVA for within-subject
repeated measurements (Hemisphere (2) Electrode
(7) Sequence (5)) of AP transformed Ln values (Statistica
1991, Statsoft, ANOVA/MANOVA procedures). This first
ANOVA result revealed whether there was a main dominant
Hemisphere effect in the evaluation of hemisphere differences. The 5 sequences were then compared in each band
with a 2-way repeated measures ANOVA (Electrode
(14) Sequence (5)), in order to take 14 electrodes into
account. Post-hoc probabilities were subsequently calculated using the Newman-Keuls test for the two main effects
(Electrode and Sequence), when the ANOVA result was
significant (P , 0.05). When the Electrode-Sequence interaction was significant, a one-way repeated measures

289

S. Cochin et al. / Electroencephalography and clinical Neurophysiology 107 (1998) 287295

3. Results

Table 1
Grouping according to Electrode factor
Alpha 1
LnAP
...
0.812
0.670
0.635

3.1. Overall results

T4
C3
C4
A

...

ANOVA was performed electrode by electrode. Post-hoc

tests (Newman-Keuls) were then used for the Sequence
factor.
2.4.2. Post-hoc tests for main Electrode factor effect.
Each electrode was compared with the others for each
frequency band regardless of sequence. Electrodes were
then grouped and classified according to means and statistically significant results. For example, in alpha 1 band,
mean Ln AP (regardless of sequence) for T4 = 0.812,
C3 = 0.670 and C4 = 0.635, because these electrodes can
then be classified in decreasing order as T4-C3-C4 whatever
the sequence. If the statistical analysis performed on these
means shows that C3 is not different from C4, C3 is different from T4 and C4 is different from T4, then the electrodes
can be divided into two groups as shown in Table 1.

A decrease in Ln AP values for all the electrode sites was

observed between still or non-human moving sequences
(W, WF, L) and moving sequences (F, R). P values resulting
from the ANOVA for absolute power are displayed in Table
2. Successive still sequences produced comparable power
across all frequencies, as did successive moving sequences.
No significant tendency towards a unilateral hemispheric
effect was found in any frequency band, whereas there
were significant differences between electrode positions
(P , 0.001). There were significant effects for Sequence
in 3 frequency bands (alpha 1: F(4,156) = 4.71, P =
0.001; beta 1: F(4,156) = 5.52, P = 0.0003; beta 2:
F(4,156) = 3.67, P = 0.007), and the interaction Electrode
Sequence in these 3 bands was very significant (alpha 1:
F(24,936) = 1.72, P = 0.017; beta 1: F(24,936) = 2.88, P =
5 10 6; beta 2: F(24,936) = 2.63, P = 0.000036).
3.2. Analytical results of main effects (all frequency bands)
3.2.1. Electrode factor (Fig. 1)
The post-hoc tests (Newman-Keuls) performed on the
main Electrode effects made it possible to classify electrodes in groups and to differentiate them regardless of the
sequence. The mean of Ln AP centroparietal electrodes
was always lower than that of occipito-temporal and frontal
electrodes. The mean of Ln AP progressively decreased
from the occipital, temporal and frontal to the centroparietal
electrodes.
The statistical tests for theta 1 and theta 2 bands revealed
that C3, C4, P4 and P3 electrodes were different from occipital, frontal and posterior temporal sites. These tests classified electrodes into 8 groups for theta 1, and 5 groups for
theta 2.
For alpha 1 band, 6 groups were obtained in which two
homogeneous groups were very different from the other 4.
These two groups were composed of electrodes C3 and C4
for the first and electrodes O1 and O2 for the second.
Five electrode groups were obtained for alpha 2 band. No

2.4.3. EEG cartography

EEG cartography shows the means of the Ln values of
spectral power for each electrode performed with a linear
interpolation method (4-nearest-neighbour interpolation
algorithm) (Duffy et al., 1979; Buchsbaum et al., 1982).
This method enables the electrical activity recorded on the
scalp to be displayed topographically using a colour plot.
2.4.4. Statistical EEG cartography
Statistical probability maps show the values of the P level
obtained with Newman-Keuls test for each frequency band
and for each electrode. The colour corresponds to the statistical significance obtained: red indicates a 0.001 P level,
yellow indicates a 0.01 P level, green corresponds to a 0.05
P level, and blue means no significant result.
Table 2

Summary of 3-way ANOVA in each frequency band; F values and probabilities (*P , 0.05, **P , 0.01, ***P , 0.001) are presented
Theta 1

H
E
S
HE
HS
ES
HES

Theta 2

Alpha 1

Alpha 2

Beta 1

Beta 2

Beta 3

0.08
38.23
0.94
1.14
1.86
0.91
0.78

0.780
***
0.445
0.339
0.121
0.589
0.766

0.52
38.11
1.71
1.62
1.45
1.14
1.02

0.480
***
0.151
0.141
0.221
0.296
0.437

2.87
33.92
4.71
1.18
2.40
1.72
0.85

0.10
***
**
0.318
0.052
*
0.677

3.63
14.37
0.61
1.90
2.06
1.22
0.93

0.064
***
0.657
0.082
0.088
0.112
0.563

0.19
30.78
5.52
0.73
1.15
2.88
1.28

0.66
***
***
0.625
0.335
***
0.164

0.39
15.72
3.67
1.41
0.76
2.63
0.93

0.537
***
**
0.210
0.554
***
0.560

1.22
27.01
1.31
1.50
0.68
1.46
0.62

0.274
***
0.267
0.178
0.606
0.072
0.921

H, Hemisphere; E, Electrode; S, Sequence.

290

S. Cochin et al. / Electroencephalography and clinical Neurophysiology 107 (1998) 287295

Fig. 1. Electrodes classification in groups in each frequency band after post-hoc test (Newman-Keuls) performed on the main Electrode effect. This
classification was performed with respect to Ln AP values for each electrode averaged over the 5 sequences. In each frequency band, centroparietal
electrodes have a lower spectral power than that of occipito-temporal and frontal electrodes.

particular group could be distinguished but C3 and C4,

which belonged to the same group, were very different
from the occipital and posterior temporal electrodes.
Central electrodes in the beta 1 band formed a homogeneous group and were distinct from all other electrodes,
and frontal sites (F3-F4-F7-F8) were different from O1O2.
Two homogeneous groups were obtained in the beta 2
band (C3-C4-P3-P4 versus all electrodes). Four groups
were obtained in the beta 3 band in which C3-C4-P3-P4
formed a homogeneous group.

3.2.2. Sequence factor (alpha 1, beta 1 and beta 2) (Fig. 2)

In the alpha 1 frequency band, a significant difference
(Newman-Keuls test) was observed between R and W
(P = 0.0086), R and L (P = 0.0066), F and W (P = 0.020)
and F and L (P = 0.020) whatever the electrode site. These
significant differences were due to a decrease in power in
the F and R sequences (Fig. 2a).
Probabilities for post-hoc tests showed that only the R
sequence was significantly different from all others in beta
1 band: R-W (P = 0.0010), R-WF (P = 0.010), R-L
(P = 0.00019), R-F (P = 0.0154) (Fig. 2b).

S. Cochin et al. / Electroencephalography and clinical Neurophysiology 107 (1998) 287295

291

(P , 0.01) and between R and W for F7 and T5 (P , 0.05).

The statistical EEG cartography demonstrates the statistical
results (Fig. 3b). A significant decrease in spectral power is
observed between still or moving non-human sequences (W,
WF, L) and human moving sequences (F, R) in centro-parietal regions.
3.3.2. Beta 1 band
Statistical analysis in this band showed significant differences between F-R and W-WF and L (P , 0.001) for 3
main electrodes (C3, O1 and O2). The F sequence was
also significantly different from L at P4 (P , 0.001). The
R sequence was significantly different from W, WF and L
sequences also at P4 (P , 0.001) and at P3 (P , 0.01) and
from all other sequences (F-W-WF-L) at F3 (P , 0.01) and
at T6 (P , 0.001). EEG cartography (Fig. 3c) and statistical
EEG cartography (Fig. 3d) showed the topography of the
electrical activity and the statistical results. In the beta 1
band, human moving sequences showed a decrease in spectral power with non-human sequences in occipital areas and
left-central areas.

Fig. 2. Mean Ln AP values for each sequence (W, white; WF, waterfall; L,
lake; F, flexion; R, rotating) averaged over the 14 electrodes in alpha 1 (a),
beta 1 (b) and beta 2 (c) bands. Significant difference between two
sequences are given by P level (*P , 0.05; **P , 0.01; ***P ,
0.001). Significant differences are mainly observed between human movement sequences (F, R) and non-human sequences (W, WF, L).

The decrease in power in R sequence in the beta 2 band

was statistically significant when compared with those of W
(P = 0.008) and L (P = 0.006) (Fig. 2c).
3.3. Electrode-Sequence interaction (Fig. 3)
The Electrode Sequence interaction, which was statistically significant for alpha 1, beta 1 and beta 2, provides
evidence that the qEEG pattern is modified according to
sequence type.
3.3.1. Alpha 1 band
EEG cartography (Fig. 3a) showed the topography of
electrical activity. These results showed a decrease in spectral power in central, parietal and frontal regions for F and R
sequences compared with W, WF and L sequences. Statistical analysis (Newman-Keuls test) in this band showed
significant differences between F and R on the one hand
and W, WF and L on the other. A significant decrease in
Ln AP was thus observed between F and R versus W, WF
and L for C3, P3, P4 (P , 0.001), between F and R versus
W and L for F4 (P , 0.01), between R and W, WF, L for F3

3.3.3. Beta 2 band

The F sequence was different from L at C3 (P , 0.01)
and from W at O2 (P , 0.001). The R sequence was statistically significantly different from all sequences at P3-P4
electrodes (P , 0.001) and from L, WF and W at C3 (P ,
0.01). The R sequence was also different from F, L and W at
T6 (P , 0.001) and from L, W at O2 (P , 0.01). EEG
cartography (Fig. 3e) and statistical EEG cartography
(Fig. 3f) showed the topography and statistical results of
the electrical activity related to beta 2 band.

4. Discussion
This study demonstrates that visual perception of human
movement on a video screen modifies the qEEG pattern,
specifically alpha 1, beta 1 and beta 2 rhythms. The modifications are proved statistically for the electrode sites of
motor cortex reactivity.
No hemisphere difference was found in our study, due to
the fact that the video task was symmetrical and thus
involved participation of both hemispheres. This is in agreement with previous studies which reported a reduction in
spectral power over both hemispheres during motor tasks
(De Toffol et al., 1990; De Toffol and Autret, 1991).
Furthermore, we noted that the spectral power of occipital
electrodes was always greater than the spectral power of
centro-parietal electrodes.
In this study, some methodological problems can be
pointed out. Firstly, the choice of a reference in EEG is
always difficult. There is no classical reference but there
are rules to be respected and the choice of a reference is
made according to the problem. Thus cartographic maps,
which are achieved by a linear interpolation method, are

292

S. Cochin et al. / Electroencephalography and clinical Neurophysiology 107 (1998) 287295

S. Cochin et al. / Electroencephalography and clinical Neurophysiology 107 (1998) 287295

isotropic and a common average reference is applied. This

common average reference involves a minor dispersion of
phenomena, but it requires perfect localisation of the activity recorded. This requirement is fulfilled by an accurate
localisation of electrodes according to the International
1020 system. Moreover, by registering EEG data with
anatomical information from MRI, Gevins et al. (1991)
demonstrated that EEGs provide images of superficial cortical electrical activity with spatial detail approaching that
of PET scans. Secondly, our study was based on frequency
band analysis. Authors usually measure spectral power in
well-defined frequency bands (theta: 3.57.5 c/s; alpha:
7.513.5 c/s; beta: 13.530 c/s) that correspond to classical
rhythms in EEGs, whereas spectra analysis allows measurements of power in narrower bands. This notion of frequency
band can thus lead to confusion if the results are interpreted
with classical significance of conventional rhythms. Traditional EEG distinguishes two rhythms in the alpha band,
posterior alpha and central mu rhythm, but these two
rhythms have a different topography and functional significance although they belong to the same alpha frequency
band. The alpha band was therefore divided into two
bands in our study; alpha 1 (7.510.5 Hz) and alpha 2
(10.513 Hz). According to the literature data, Rolandic
(central) mu rhythm is related in frequency and amplitude
to the slower alpha rhythm (710 Hz) and posterior alpha
belongs to the higher frequencies of the alpha band (1013
Hz). The range of mu frequencies obtained is in general
agreement with Chatrian et al. (1959), who reported a
range of 711 c/s and a mean of 9 c/s, and Yamada and
Kooi (1975), who reported a range of 812 c/s and a mean
of 10 c/s. These results suggest that artificial division in
frequency bands corresponds to the 7.510.5 Hz band
with central mu rhythm and between 10.513 Hz band
with posterior alpha rhythm.
It is generally accepted that the alpha rhythms can be considered as a class of resting rhythms, seen when sensory
systems are idling or not receiving or processing sensory
input (Adrian and Matthews, 1934). Mu waves disappear
with sensorimotor stimulation affecting the area of Rolando
(Chatrian et al., 1958), just as alpha waves disappear with
visual stimulation. In this framework, the mu rhythm can be
considered to be a somatosensory alpha rhythm, sensitive to
somatic afferent input. Similarly, occipital alpha activity is a
visual alpha rhythm, sensitive to disruption by visual afferent input (Kuhlman, 1978). The results obtained in this study
are in accordance with many aspects of these propositions.
The prevalence of mu rhythm in routine EEGs has been
estimated to be from 2.9% to 18% (Gastaut et al., 1954;
Klass and Bickford, 1957; Dongier and Dongier, 1958; Cha-

293

trian et al., 1959; Schnell and Klass, 1966; Koshino and

Niedermeyer, 1975). In fact, mu rhythm occurs more
often than previously described by visually evaluating
EEGs. Frequency analysis methods suggest that it can be
present in 50% to 100% of healthy subjects (Schoppenhorst
et al., 1977, 1980; Storm Van Leeuwen et al., 1978; Kuhlman, 1978; Pfurtscheller and Aranibar, 1979). The detection
of mu rhythm in varying percentages of the population may
in part be attributed to their dependence on the state of
vigilance, as well as to the frequent superposition of alpha
waves because of low voltage (Schoppenhorst et al., 1977,
1980). Mu rhythm is dependent on obvious changes in vigilance (Chatrian et al., 1958; Chatrian, 1976). Mental arithmetic (Chatrian et al., 1958), problem solving (Creutzfeldt
et al., 1969) and fatigue suppress the mu rhythm (Chatrian et
al., 1958; Chatrian, 1976; Niedermeyer and Koshino, 1975).
According to Gevins et al. (1980), there is no significant
correlation between cognitive activity and EEG and falsely
significant results of previous studies might be due to
uncontrollable factors. Gevins et al. demonstrated that, in
addition to their cognitive contents, the tasks used in most
previous studies have also differed in stimulus characteristics, motor demands and performance-related factors
which may account for their results. Gevins and Schaffer
(1980) subsequently proposed methodological rules concerning the choice of subjects and EEG analysis. However,
the criteria established by Gevins and Schaffer (1980) were
respected in our study, in particular the effect of attention.
Video sequences were presented alternating movement
sequences and still sequences. Moreover, the results demonstrate that attention was steady during EEG recording since
sequences were presented at different points in the video. In
this way, Rotating, which is in the middle of the video
(sequence no. 16), was not different from Flexion which
was at the beginning (no. 3). Similarly, Waterfall (no. 22)
was not significantly different from Lake (no. 4) and White
(no. 1).
Our study also showed that beta rhythm is modified by
observation of movement, particularly in the bands we have
called beta 1 (1318 Hz) and beta 2 (1825 Hz). This result
agrees with those of others who have observed a central beta
rhythm which can be blocked by motor activity or tactile
stimulation. Beta rhythm has been investigated since the
pioneering work of Berger (1930) and Tonnies (1934),
who first recorded electrocortical beta band activity (16
22 Hz) in the precentral cortex. Jasper and Andrew (1938)
reported blocking reactions of the precentral beta rhythm
which occurred on human scalp recordings after tactile stimulation of the contralateral side of the body and which was
not affected by visual stimulation. Some years later, Jasper

Fig. 3. EEG cartographies in alpha 1 band (a), beta 1 band (c) and beta 2 band (e) for each sequence (W, white; WF, waterfall; L, lake; F, flexion; R, rotating).
EEG cartographies show the mean of the Ln values of spectral power (AP) in mV2. Statistical EEG cartographies in alpha 1 band (b), beta 1 band (d) and beta
2 band (f) show the values of the P level obtained with the Newman-Keuls test for each electrode for human motion sequences (F, R) versus non-human
sequences (W, WF, L) (red (***): P , 0.001; yellow (**): P , 0.01; green (*): P , 0.05; blue (ns): no significant result).

294

S. Cochin et al. / Electroencephalography and clinical Neurophysiology 107 (1998) 287295

and Penfield (1949) demonstrated by human electrocorticograms that voluntary hand movement serves to block precentral beta rhythm. The most characteristic feature of
precentral beta rhythm is its blocking, or desynchronisation,
localised over the motor hand area by fist clenching. The
functional relationship between precentral beta rhythm and
sensorimotor activation was confirmed by Gastaut (1952).
Rougeul et al. (1979) studied fast somato-parietal rhythms
of around 18 Hz in monkeys and found that they were
blocked by the smallest body movement. Beta desynchronisation during voluntary hand movement was quantified
and displayed in the form of event-related desynchronisation (ERD) time courses by Pfurtscheller (1981), beta ERD
occurring in parallel with mu ERD. Movement-related
desynchronisation in the lower beta frequency band
recorded over the sensorimotor hand area has been reported,
not only in a self-paced index finger movement task but also
with foot movement (Toro et al., 1994).
As previously described, both mu and beta rhythms are
modified by movement, whether prepared, executed or, as in
our study, observed. The occipital alpha rhythm is considered to be an idling rhythm of the visual system and the mu
and central beta rhythms to be idling rhythms of the sensorimotor system (Kuhlman, 1978; Pfurtscheller, 1992;
Pfurtscheller and Neuper, 1994; Pfurtscheller et al., 1996).
In 1959, Ciganek reported reactive rhythms over the central
scalp regions in both alpha and beta frequency ranges (Ciganek, 1959). Furthermore, using magnetoencephalographic
recordings in 4 healthy subjects, Tiihonen et al. (1989)
observed two rhythms over the sensorimotor cortex, one
at 21 Hz that reacted to opening and closing of the hands
and another at 10 Hz that reacted to hand movement in only
one patient but was reduced by eye opening in all. More
recently, Arroyo et al. (1993) reported that when mu rhythm
was blocked there was an overall decrease in power density
in all frequency bands. This is consistent with previous
studies of Pfurtscheller et al., who showed a decrease in
mu power spectra in scalp recorded EEG using eventrelated desynchronisation with simultaneous decrease in
low-frequency beta power in response to movement
(Pfurtscheller, 1981; Pfurtscheller et al., 1994). It is therefore clear that reactive somatosensory rhythms occur in two
frequency bands.
Our results and previous conclusions support the hypothesis that mu and beta rhythms are associated with motor
behaviour, and our results are in agreement with Kuhlmans
hypothesis which considered mu rhythm sensitive to
somatic afferent input (Kuhlman, 1978). The mu rhythm
is thus located in central areas during visual perception of
human movement, whereas the beta rhythm appears to be
more posterior since it involves not only centro-parietal
areas but also occipital areas. This reaction to observed
movement is specific to human movement. In order to confirm the specificity of these results, a study is being performed to compare EEG modifications during real motion
and visual motion observation.

Acknowledgements
This study was supported by INSERM U316 The nervous system from the foetus to the child (Prof. L. Pourcelot), INSERM network no. 4R002B, INSERM-France
Telecom 1993 convention, VTCOM society, and the Langlois Foundation. S.C. is a Conseil Regional-Region Centre
grant-holder.

References
Adrian, E.D. and Matthews, B.H.C. The Berger rhythm: potential changes
from the occipital lobes in man. Brain, 1934, 57: 355385.
Anderer, P., Semlitsch, H.V., Saletu, B. and Filz, L. Ein korrekturverfahren zur reduktion okularer artefakte in der EEG- und ERP-topographie.
Forschritte in der Biologischen Psychiatrie, 1989.
Arroyo, S., Lesser, R.P., Gordon, B., Uematsu, S., Jackson, D. and
Webber, R. Functional significance of the mu rhythm of human cortex:
an electrophysiological study with subdural electrodes. Electroenceph.
clin. Neurophysiol., 1993, 87: 7687.
Beaumont, J.G., Mayes, A.R. and Rugg, M.D. Asymmetry in EEG alpha
coherence and power: effects of task and sex. Electroenceph. clin.
Neurophysiol., 1978, 45: 393406.
Beisteiner, R., Hollinger, P., Lindinger, G., Lang, W. and Berthoz, A.
Mental representations of movements. Brain potentials associated with
imagination of hand movements. Electroenceph. clin. Neurophysiol.,
1995, 96: 183193.
ber das Elektrenkephalogramm des Menschen. Zweite MitBerger, H. U
teilung (2nd report). J. Psychol. Neurol. (Leipzig), 1930, 40: 160179.
Buchsbaum, M.S., Rigal, F., Coppola, R., Capelletti, J., King, C. and
Johnson, J. A new system for gray-level surface distribution maps of
electrical activity. Electroenceph. clin. Neurophysiol., 1982, 53: 237
242.
Butler, S. and Glass, A. EEG correlates of cerebral dominance. Adv.
Psychobiol., 1976, 3: 219272.
Chatrian, G.E. The mu rhythms. In: A. Remond (Ed.), Handbook of Electroencephalography and Clinical Neurophysiology. Elsevier, Amsterdam, 1976, pp. 104114.
Chatrian, G.E., Petersen, M.C. and Lazarete, J.A. The blocking of the
rolandic wicket rhythm and some central changes related to
movement. Electroenceph. clin. Neurophysiol., 1958, 10: 771.
(abstract).
Chatrian, G.E., Petersen, M.C. and Lazarete, J.A. The blocking of the
rolandic wicket rhythm and some central changes related to
movement. Electroenceph. clin. Neurophysiol., 1959, 11: 497510.
Ciganek, L. A contribution to the problem of rolandic arceau rhythm.
Electroenceph. clin. Neurophysiol., 1959, 11: 185. (abstract).
Cohen-Seat, G., Gastaut, H., Faure, J. and Heuyer, G. Etudes
experimentales de lactivite nerveuse pendant la projection
cinematographique. Rev. Int. Filmol., 1954, 5: 764.
Cohen-Seat, G., Lelord, G. and Rebeillard, M. Conditions actuelles dutilisation de lEEG dans la recherche filmologique. Principes et premie`res
observations. Rev. Int. Filmol., 1956, 2728: 157176.
Cooley, J.W. and Tukey, J.W. An algorithm for the machine calculation of
complex Fourier series. Math. Comput., 1965, 19: 297301.
Creutzfeldt, O., Grunewald, G., Simonova, O. and Schmitz, H. Changes of
the basic rhythm of the EEG during the performance of mental and
vasomotor task. In: C. Evans and T. Mulholland (Eds.), Attention in
Neurophysiology. Butterworths, London, 1969, pp. 148168.
Decety, J. and Ingvar, D.H. Brain structures participating in mental simulation of motor behavior: a neuropsychogical interpretation. Acta
Psychol., 1990, 73: 1334.
De Toffol, B. and Autret, A. Influence of lateralized neuropsychological

S. Cochin et al. / Electroencephalography and clinical Neurophysiology 107 (1998) 287295

activities with and without sensorimotor components on EEG spectral
power (alpha-rhythm). Int. J. Psychophysiol., 1991, 11: 109114.
De Toffol, B., Autret, A., Markabi, S. and Roux, S. Influence of lateralized
sensorimotor and neuropsychological activities on electroencephalographic spectral power. Electroenceph. clin. Neurophysiol., 1990, 75:
200206.
di Pellegrino, G., Fadiga, L., Fogassi, L., Gallese, V. and Rizzolatti, G.
Understanding motor events: a neurophysiological study. Exp. Brain
Res., 1992, 91: 176180.
Dolce, G. and Waldeier, H. Spectral and multivariate analysis of EEG
changes during mental activity in man. Electroenceph. clin. Neurophysiol., 1974, 36: 577584.
Dongier, M. and Dongier, S. Quelques aspects de lelectroencephalogramme des nevroses. Evol. Psychiat., 1958, 1: 118.
Duffy, F.H., Burchfiel, J.L. and Lombroso, C.T. Brain electrical activity
mapping (BEAM): a new method for extending the clinical utility of
EEG and evoked potential data. Ann. Neurol., 1979, 5: 309321.
Fadiga, L., Fogassi, L., Pavesi, G. and Rizzolatti, G. Motor facilitation
during action observation: a magnetic stimulation study. J. Neurophysiol., 1995, 73 (6): 26082611.
Gastaut, H. Etude electrocorticographique de la reactivite des rythmes
rolandiques. Rev. Neurol., 1952, 87: 176182.
Gastaut, H., Dongier, M. and Courtois, G. On the significance of wicket
rhythms in psychosomatic medicine. Electroenceph. clin. Neurophysiol., 1954, 6: 687.
Gevins, A.S. and Schaffer, R.E. A critical review of electroencephalographic (EEG) correlates of higher cortical functions. CRC Crit. Rev.
Bioeng., 1980, 4 (2): 113162.
Gevins, A.S., Doyle, J.C., Schaffer, R.E., Callaway, E. and Yeager, C.
Lateralized cognitive processes and the electroencephalogram. Science,
1980, 207: 10061007.
Gevins, A.S., Le, J., Brickett, P., Reutter, B. and Desmond, J. Seeing
through the skull: advanced EEGs use MRIs to accurately measure
cortical activity from the scalp. Brain Topogr., 1991, 4 (2): 125131.
Heuyer, G., Cohen-Seat, G., Lelord, G. and Rebeillard, M. Etudes EEG
denfants inadaptes soumis a` la stimulation filmique. Rev. Neuropsychiat. Inf. Hyg. Ment. Enf., 1957, 910: 494511.
Ingvar, D.H. and Philipson, L. Distribution of cerebral blood flow in the
dominant hemisphere during motor ideation and motor performance.
Ann. Neurol., 1977, 2: 230237.
Jasper, H.H. Report of the committee on methods of clinical examination
in electroencephalography. Electroenceph. clin. Neurophysiol., 1958,
10: 370375.
Jasper, H.H. and Andrew, H.L. Electroencephalography. III. Normal differentiations of occipital and precentral regions in man. Arch. Neurol.
Psychiat., 1938, 39: 96115.
Jasper, H.H. and Penfield, W. Electrocorticograms in man: effect of voluntary movement upon the electrical activity of the precentral gyrus. Arch.
Psychiat. Neurol., 1949, 183: 163174.
Klass, D.W. and Bickford, R.G. Observations of the rolandic arceau
rhythm. Electroenceph. clin. Neurophysiol., 1957, 9: 570. (abstract).
Koshino, Y. and Niedermeyer, E. Enhancement of Rolandic mu-rhythm by
pattern vision. Electroenceph. clin. Neurophysiol., 1975, 38: 535538.
Kuhlman, W.N. Functional topography of the human mu rhythm. Electroenceph. clin. Neurophysiol., 1978, 44: 8393.
Lassen, N.A. and Ingvar, D.H. Brain regions involved in voluntary movements as revealed by radioisotopic mapping of CBF or CMR-glucose
changes. Rev. Neurol., 1990, 146 (10): 620625.

295

Lelord, G. Modalites reactionnelles differentes de rythmes moyens et

anterieurs autour de 10 c/s. Rev. Neurol., 1957, 96: 524526.
Lelord, G. Interet des methodes electrophysiologiques dans letude de
laction du film cinematographique chez lenfant et ladolescent.
Pedopsychiat., 1966, 1: 159166.
Lelord, G. Le medecin, lenfant et la television. Rev. Med. Tours, 1984,
18: 10111013.
Niedermeyer, E. and Koshino, Y. Mu-rhythmus: vorkommen und klinische
bedeutung. Z. EEG-EMG, 1975, 6: 6978.
Papanicolaou, A.C., Loring, D.W., Deutsch, G. and Eisenberg, H.M. Taskrelated EEG asymmetries: a comparison of alpha blocking and beta
enhancement. Int. J. Neurosci., 1986, 30: 8185.
Pfurtscheller, G. Central beta rhythm during sensorimotor activities in
man. Electroenceph. clin. Neurophysiol., 1981, 51: 253264.
Pfurtscheller, G. Event-related synchronization (ERS): an electrophysiological correlate of cortical areas at rest. Electroenceph. clin.
Neurophysiol., 1992, 83: 6269.
Pfurtscheller, G. and Aranibar, A. Evaluation of event-related desynchronization (ERD) preceding and following voluntary self-paced
movement. Electroenceph. clin. Neurophysiol., 1979, 46: 138146.
Pfurtscheller, G. and Neuper, C. Event-related synchronization of murhythm in the EEG over the cortical hand area in man. Neurosci.
Lett., 1994, 174: 9396.
Pfurtscheller, G., Pregenzer, M. and Neuper, C. Visualization of sensorimotor areas involved in preparation for hand movement based on classification of mu and central beta rhythms in single EEG trials in man.
Neurosci. Lett., 1994, 181: 4346.
Pfurtscheller, G., Stancak, A. Jr. and Neuper, C. Post-movement beta
synchronization. A correlate of an idling motor area?. Electroenceph.
clin. Neurophysiol., 1996, 98: 281293.
Rougeul, A., Bouyer, J.J., Dedet, L. and Debray, O. Fast somato-parietal
rhythms during combined focal attention and immobility in baboon and
squirrel monkey. Electroenceph. clin. Neurophysiol., 1979, 46: 310
319.
Schnell, R.G. and Klass, D.W. Further observations on the Rolandic arceau
rhythm. Electroenceph. clin. Neurophysiol., 1966, 20: 95.
Schoppenhorst, M., Brauer, F. and Freund, G. Identification of mu-rhythm
by means of spectral analysis and coherence functions. Electroenceph.
clin. Neurophysiol., 1977, 43: 131. (abstract).
Schoppenhorst, M., Brauer, F., Freund, G. and Kubicki, S. The significance of coherence estimates in determining central alpha and mu
activities. Electroenceph. clin. Neurophysiol., 1980, 48: 2533.
Shaw, J.C., OConnor, K.P. and Ongley, C. The EEG as a measure of
cerebral function organization. Br. J. Psychiat., 1977, 130: 260264.
Storm Van Leeuwen, W., Wieneke, G., Spoelstra, P. and Versterg, H. Lack
of bilateral coherence of mu rhythm. Electroenceph. clin. Neurophysiol.,
1978, 44: 140146.
Tiihonen, J., Kajola, M. and Hari, R. Magnetic mu rhythm in man. Neuroscience, 1989, 32: 793800.
Tonnies, J.F. Die unipolare Ableitung electrischer spannungen vom menschlichen Gehirn. Naturwissenschaften, 1934, 22: 411.
Toro, C., Deuschl, G., Thatcher, R., Sato, S., Kufta, C. and Hallett, M.
Event-related desynchronization and movement-related cortical potentials on the ECoG and EEG. Electroenceph. clin. Neurophysiol., 1994,
93: 380389.
Yamada, T. and Kooi, K.A. Level of consciousness and the mu rhythm.
Electroenceph. clin. Neurophysiol., 1975, 6: 8088.