Professional Documents
Culture Documents
Clinical Neurophysiology
journal homepage: www.elsevier.com/locate/clinph
a r t i c l e
i n f o
Article history:
Accepted 23 January 2014
Available online 2 February 2014
Keywords:
Parkinsons disease
Event-related synchronization
Event-related desynchronization
Mu-rhythm
Mirror neuron system
h i g h l i g h t s
EEG alpha and beta band desynchronization (i.e. mu-rhythm ERD) during movement observation was
a b s t r a c t
Objective: Patients with Parkinsons disease often experience difculties in adapting movements and
learning alternative movements to compensate for symptoms. Since observation of movement has been
demonstrated to lead to the formation of a lasting specic motor memory that resembled that elicited by
physical training we hypothesize that mu-rhythm desynchronization in response to movement observation is impaired in Parkinsons disease.
Method: In a pilot study with nine patients with Parkinsons disease at a Hoehn and Yahr stage of I or II
and eleven age-matched controls, we tested this hypothesis by comparing the event related desynchronization (ERD) patterns from the EEG recorded during the observation of hand action and baseline videos.
Results: Healthy subjects showed normal bilateral ERD of the mu-rhythm. In patients with Parkinsons
disease this distinct ERD pattern was lacking.
Conclusion: The results of this study suggest that event-related mu-rhythm desynchronization is
impaired in Parkinsons disease, even at early stages of the disease.
Signicance: Studying event-related mu-rhythm desynchronization dysfunction in Parkinsons disease
patients may enhance our understanding of symptoms as impaired motor learning.
2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights
reserved
1. Introduction
Idiopathic Parkinsons disease (PD) is a progressive neurodegenerative disease resulting in multiple motor symptoms which differ
in type and severity depending on disease progression, and show a
large variability between patients. PD symptoms are associated
Corresponding author. Address: University of Twente, Biomedical Signals &
Systems, Zuidhorst 214, P.O. Box 217, 7500 AE Enschede, The Netherlands. Tel.: +31
53 4892759; fax: +31 53 4892287.
E-mail address: t.heida@utwente.nl (T. Heida).
with basal ganglia dysfunction caused by the loss of dopamineproducing cells in the substantia nigra pars compacta. The basal
ganglia are involved in enabling practiced motor acts and in gating
the initiation of voluntary movements by modulating motor
programs stored in the motor cortex and elsewhere in the motor
hierarchy (Marsden, 1982; Mink, 1996; OReilly and Frank, 2006).
Parkinsons patients often experience difculties in adapting
movements and learning alternative movements to compensate
for symptoms. Observation of movement has been demonstrated
to lead to the formation of a lasting specic motor memory that
resembled that elicited by physical training (Stefan et al., 2005).
http://dx.doi.org/10.1016/j.clinph.2014.01.016
1388-2457/ 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved
1820
2. Methods
All patients fullled the UK Brain Bank criteria for Parkinsons disease (Hughes et al., 1992). Patients with severe tremor, dyskinesias
or those treated with deep brain stimulation were excluded in order to prevent EEG artifacts. Patients with known PD related
dementia according to the clinical diagnostic criteria (Emre et al.,
2007) and those using sedative medication were also excluded. Patients were allowed to take their usual anti-parkinsonian medication on the day of the experiment. All procedures conformed to the
Declaration of Helsinki and were approved by the Medical Ethical
Committee of the Medisch Spectrum Twente in Enschede, the
Netherlands. All subjects gave written informed consent prior to
participation in the study.
2.2. Measurement set up
Subjects sat comfortably in an armchair in an electrically and
sound-shielded room watching hand action and baseline videos.
The EEG was recorded using ASA acquisition software (ANT International BV, the Netherlands); 64 Ag/AgCl electrodes were positioned according the 10/10-system using a waveguard EEG cap.
The signals were amplied, low-pass ltered (digital FIR lter
1350 Hz cut off) and sampled at 5 kHz (Refa, Twente Medical Systems International BV, the Netherlands).
2.3. Experimental protocol
A measurement consisted of six trials. During each trial, subjects watched a video consisting of eight fragments showing hand
movements interspersed with seven baseline fragments. Subjects
could rest between trials whenever desired. Presented hand actions included pinching, grasping, ball grasping, nger tapping,
and hand turning (supination/pronation), randomly executed with
left or right hand. As an example, Fig. 1A shows three frames of the
video showing pinching movements executed with the left hand. A
moving red ball over a black background was used as baseline video (Fig. 1B). It was expected that during the observation of this
baseline video the brain returned to a resting state in which mirror
neurons are deactivated. Patients were carefully watched to not
perform any voluntary movements.
Each trial had a duration of approximately two and a half minutes with the hand action and baseline fragments having a length
varying from 8 to 12 s. The various types of hand actions were randomly distributed over the six trials to prevent predictability. Each
hand action appeared several times in a number of trials up to a
maximum of six times.
Table 1
Patient details (time in years).
Patient
Sex
Age
Disease duration
Antiparkinsonian medication
1
2
3
4
5
6
7
8
9
F
M
F
M
M
M
M
M
M
49
62
54
72
72
68
75
73
74
5
4
10
2
8
7
12
7
4
I
II
II
I
II
II
II
II
III
Tremor
Bradykinesia
Tremor, bradykinesia
Tremor
Tremor, rigidity
Tremor, freezing, bradykinesia
Freezing, bradykinesia
Rigidity, bradykinesia, akinesia
Tremor, bradykinesia
None
Ropinirole
Levodopa/benserazide, pergolide, amantadine
None
Levodopa/benserazide, pramipexole
Pramipexole
Levodopa/benserazide, pramipexole
Levodopa/benserazide
Levodopa/carbidopa, entacapone
1821
Fig. 1. (A) Video frames of the pinching movement performed with the left hand; the movement was repeated several times during the video. (B) Three frames of the baseline
video; the red ball moved across the screen in a random fashion with variable speed. (For interpretation of the references to color in this gure legend, the reader is referred to
the web version of this article.)
2.4.1. Preprocessing
The EEG data was down-sampled to 500 Hz and ltered using a
band-pass lter of 170 Hz (4th order non-causal Butterworth lter) to remove drift and noise. In addition, a 50 Hz notch-lter was
used to remove power line noise. To be able to analyze local cortical activity and to limit volume conduction effects, source derivation was applied (Defebvre et al., 1998): for each electrode the
average of the signals of the neighboring electrodes was extracted.
In a few occasions electrodes were malfunctioning during part of
the experiment. These electrodes were excluded from the data
set and the source derivation was adapted accordingly.
The EEG signals were split into ve components by band-pass
ltering the data (4th order non-causal Butterworth lter) according to the EEG frequency bands: delta (14 Hz), theta (48 Hz), alpha (813 Hz), beta (1330 Hz), and gamma (3070 Hz) band. For
artifact detection and removal the signal components of each video
fragment were split into epochs of 500 ms, with 50% overlap. The
mean signal power of each epoch was calculated by taking the
mean of the squared voltage. For each electrode and for each signal
component, a threshold level equal to the mean power of the total
fragment plus two times the standard deviation was used to remove epochs having a mean power exceeding this threshold. The
remaining data from all repetitions of a single hand action video
or baseline video were combined for further analyses.
3. Results
3.1. Mu-rhythm ERD in healthy subjects
Fig. 2 shows the grand average ERD/ERS topoplots of the control
group that resulted from the observation of the ball grasping
movement performed by the right hand. The alpha and beta band
show relatively similar ERD patterns in the area surrounding the
1822
Theta
4-8 Hz
Alpha
8-13 Hz
Beta
14-30 Hz
synchronization
Delta
<4 Hz
desynchronization
Gamma
>30 Hz
Fig. 2. Grand average ERD topoplots of the control group (n = 11) resulting from the observation of the ball grasping movement performed by the right hand. Bilateral alpha
and beta band desynchronization (around C3 and C4) indicate mu-rhythm desynchronization. The high levels of desynchronization in the frontal areas in the delta, theta and,
to some extent, the alpha band may be ascribed to eye blinking and eye movements.
electrodes C3 and C4 (i.e. C3, FC3, C5, CP3, C1, and C4, FC4, C2, CP4,
C6, resp.), corresponding to the mu-rhythm desynchronization patterns. No statistical differences between the observation of the action performed by the left and right hand, or between the left and
right hemisphere, were found. Furthermore, no age and gender related effects were found.
Besides the induced ERD patterns all observed hand actions resulted in a distinct alpha and beta band synchronization (ERS) area
around electrodes Pz and POz, the central part of the parietaloccipital area. The level of synchronization in this area seemed to
be related to the level of desynchronization in the area around
C3 and C4: a high ERD level around C3/C4 was accompanied by a
high ERS level around Pz/POz.
3.2. Mu-rhythm ERD in Parkinsons patients
Fig. 3 shows the grand average ERD/ERS topoplots of the Parkinson group resulting from the observation of the ball grasping
movement performed by the right hand. In contrast to the grand
average topoplots of the control group the Parkinson group did
not show distinctive desynchronization spots around C3 and C4
in the alpha and beta band. Some types of hand actions did evoke
ERD around C3/C4 (e.g. left hand pinching and right hand grasping)
in the alpha band, but the effect was much smaller compared to
the control group. No statistically signicant difference was found
between the observation of left and right hand actions for all types
of hand actions (p > 0.05). In addition, no statistically signicant
difference was found between the left and right hemisphere, and
1823
Theta
4-8 Hz
Alpha
8-13 Hz
Beta
14-30 Hz
synchronization
Delta
<4 Hz
desynchronization
Gamma
>30 Hz
Fig. 3. Grand average ERD/ERS topoplots of the Parkinson group (n = 9) resulting from the observation of the ball grasping movement performed by the right hand. The alpha
and beta band ERD levels around C3/C4 are much lower compared to the control group (p < 0.05).
4. Discussion
The main nding of our study was that patients with PD did not
show event-related mu-rhythm desynchronization in the area
around electrodes C3 and C4 in response to movement observation. Our experimental protocol was able to detect normal ERD
(and also ERS) patterns in our healthy controls very similar to
the patterns described in previous studies (Iacoboni and Dapretto,
2006; Keysers and Gazzola, 2006; Rizzolatti et al., 2009), both with
respect to magnitude and topographic distribution over the scalp.
The ERD levels of the healthy control and PD group showed signicant differences for the alpha and beta band for all observed
movements (except one in the alpha band, see Fig. 4), while the
other frequency bands did not show these clear differences. It
should be noted that for all frequency bands ERD levels within
the same area were compared, while we only expected to see clear
mu-rhythm ERD, and thus desynchronization in the alpha and beta
band. The level of ERD (or even ERS) and the spatial distribution
may differ for the other frequency bands, and may be related to different functions.
The elevated overall baseline power and increased alpha and
beta synchronization levels in the baseline EEG in PD may express
overactivity of motor cortical areas and reduced inhibition of the
motor cortex as a result of dopamine depletion (Seiss and Praamstra,
2004; Stoffers et al., 2008). It has been found that increased resting
state cortico-cortical coupling in the alpha range (810 Hz, i.e. alpha1) is a feature of PD from the earliest clinical stages onward
(Stoffers et al., 2008). With increasing disease duration and severity of PD symptoms, neighboring frequency bands (i.e. theta
(48 Hz), alpha2 (1013 Hz) and beta band (1330 Hz)) become
increasingly involved in the off state (Stoffers et al., 2008;
Silberstein et al., 2005). EEG-EEG coherence over 1035 Hz has
been found to correlate with severity of disease symptoms in untreated PD (Silberstein et al., 2005). Dopaminergic therapy reduces
the coupling in parallel with motor improvement. The reduced motor cortex inhibition in PD is proposed to reect the loss of spatial
and temporal selectivity in basal ganglia functioning that inuence
the selection and suppression of competing responses (Seiss and
Praamstra, 2004). The desynchronization of these idling rhythms
is impaired as well as the local synchronization at 3050 Hz (the
gamma band), required for the execution of motor programs (Devos et al., 2003; Salenius et al., 2002). This failure of high frequency
synchronization is reected downstream as the inability to generate fused muscle contraction. Impairments in the degree of suppression of mu-power during movement have been shown to
correlate with bradykinesia (Defebvre et al., 1993; Silberstein
et al., 2005).
A high baseline level may mean that it is easier to detect desynchronization. However, average levels of ERS/ERD were comparable for the PD group and the group of healthy subjects. Alegre
et al. (2010) found a similar effect in STN activity: In STN recordings in PD patients in which abnormal beta oscillatory activity is
detected, movement-related decrease in beta activity is still present with the same relative intensity under medication on and off
conditions, while after medication intake the oscillatory activity
in the beta band is greatly attenuated (Alegre et al., 2010).
In the central part of the parietal-occipital areas (around electrodes Pz and POz) synchronization was observed in the alpha
1824
Fig. 4. Visualization of the comparison of the ERD level in the area around electrodes C3 and C4, of the PD group with the control group for each of the observed hand actions,
and for all frequency bands. The bars represent the mean SEM (standard error of the mean). Hand actions: FT = nger tapping; PC = pinching; HT = hand turning;
GP = grasping; and BG = ball grasping. Indicates a statistically signicant difference (p < 0.05).
and beta band for most of the hand action videos in the control
group. A similar simultaneous contralateral localized mu rhythm
ERD and an occipital localized alpha rhythm ERS was observed during the performance of self-paced hand movements in a group
study of nine healthy subjects as described by Pfurtscheller and
Lopes da Silva (1999). They explained this contrasting behavior
as the expression of the activation of those cortical areas involved
in a single motor program while areas that are not involved in the
execution of this program are deactivated.
ERS around the parietal-occipital electrodes Pz and POz may
also be related to motion processing, since these electrodes are located in the region including the visual area V3A. Pfurtscheller and
Lopes da Silva (1999) pose that gamma band oscillations (around
40 Hz) may reect a stage of active information processing since
they appear appropriate to establish rapid coupling or synchronization between spatially separated cell assemblies (Pfurtscheller
and Lopes da Silva, 1999). A prerequisite for gamma band ERS
may be the desynchronization in alpha and beta band rhythms
(Pfurtscheller and Lopes da Silva, 1999). In almost all of the topoplots from the control group alpha and beta band ERS in the parietal-occipital area was accompanied by gamma band ERD (see
Fig. 2). The ERS around electrodes POz and Pz as observed in the
control group was also present in the Parkinson group, at an even
higher level. In most cases this ERS was not accompanied by a gamma band ERD pattern, which may indicate a dysfunction of visuomotor integration as is often suggested to occur in PD (Devos et al.,
2003; Labyt et al., 2003; Tropini et al., 2011).
Another explanation for this elevated ERS level can be found in
the choice for the baseline movies. The moving ball movie could
1825
Frey SH, Gerry VE. Modulation of neural activity during observational learning of
actions and their sequential orders. J Neurosci 2006;26:13194201.
Hughes AJ, Daniel SE, Kilford L, Lees AJ. Accuracy of clinical diagnosis of idiopathic
parkinsons disease: a clinico-pathological study of 100 cases. J Neurol
Neurosurg Psych 1992;55:1814.
Iacoboni M, Dapretto M. The mirror neuron system and the consequences of its
dysfunction. Nat Rev Neurosci 2006;7:94251.
Kessler K, Biermann-Ruben K, Jonas M, Siebner HR, Bumer T, Mnchau A, et al.
Investigating the human mirror neuron system by means of cortical
synchronization during the imitation of biological movements. Neuroimage
2006;33:22738.
Keysers C, Gazzola V. Towards a unifying neural theory of social cognition. Prog
Brain Res 2006;156:379401.
Khn AA, Williams D, Kupsch A, Limousin P, Hariz M, Schneider G-H, et al. Eventrelated beta desynchronization in human subthalamic nucleus correlates with
motor performance. Brain 2004;127:73546.
Labyt E, Devos D, Bourriez JL, Cassim F, Deste A, Guieu JD, et al. Motor preparation
is more impaired in Parkinsons disease when sensorimotor integration is
involved. Clin Neurophysiol 2003;114:242333.
Marceglia S, Fiorio M, Foffani G, Mrakic-Sposta S, Tiriticco M, Locatelli M, et al.
Modulation of beta oscillations in the subthalamic area during action
observation in Parkinsons disease. Neuroscience 2009;161:102736.
Marsden CD. The mysterious motor function of the basal ganglia: the Robert
Wartenberg lecture. Neurology 1982;32:51439.
Mink JW. The basal ganglia: focused selection and inhibition of competing motor
programs. Prog Neurobiol 1996;50:381425.
Nishitani N, Hari R. Temporal dynamics of cortical representation for action. PNAS
2000;97:9138.
OReilly RC, Frank MJ. Making working memory work: a computational model of
learning in the prefrontal cortex and basal ganglia. Neural Comp
2006;18:283328.
Pfurtscheller G. Lopes da Silva FH. Event-related EEG/MEG synchronization and
desynchronization: basic principles. Clin Neurophysiol 1999;110:184257.
Rizzolatti G, Craighero L. The mirror-neuron system. Annu Rev Neurosci
2004;27:16992.
Rizzolatti G, Fabbri-Destro M, Cattaneo L. Mirror neurons and their clinical
relevance. Neurology 2009;5:2434.
Salenius S, Avikainen S, Kaakkola D, Har R, Brown P. Defective cortical drive to
muscle in Parkinsons disease and its improvement with levodopa. Brain
2002;125:491500.
Seiss E, Praamstra P. The basal ganglia and inhibitory mechanisms in response
selection: evidence from subliminal priming of motor responses in Parkinsons
disease. Brain 2004;127:3309.
Silberstein P, Pogosyan A, Khn AA, Hotton G, Tisch S, Kupsch A, et al. Corticocortical coupling in Parkinsons disease and its modulation by therapy. Brain
2005;128:127791.
Stefan K, Cohen LG, Duque J, Mazzocchio R, Celnik P, Sawaki L, et al. Formation of a
motor memory by action observation. J Neurosci 2005;25:933946.
Stoffers D, Bosboom JLW, Deijen JB, Wolters ECh, Stam CJ, Berendse HW. Increased
cortico-cortical functional connectivity in early-stage Parkinsons disease: an
MEG study. Neuroimage 2008;41:21222.
Tangwiriyasakul C, Verhagen R, Van Putten MAJM, Rutten WLC. Importance of
baseline in eventrelated desynchronization during a combination task of motor
imagery and motor observation. J Neural Eng 2013;10. http://dx.doi.org/
10.1088/1741-2560/10/026009.
Tropini G, Chiang J, Wang ZJ, Ty E, McKeown MJ. Altered directional connectivity in
Parkinsons disease during performance of a visually guided task. Neuroimage
2011;56:214456.
Zaidel A, Spivak A, Grieb B, Bergman H, Israel Z. Subthalamic span of oscillations
predicts deep brain stimulation efcacy for patients with Parkinsons disease.
Brain 2010;133:200721.