Content uploaded by Karin Rosenkranz
Author content
All content in this area was uploaded by Karin Rosenkranz on Mar 23, 2016
Content may be subject to copyright.
Available via license: CC BY-NC-SA 4.0
Content may be subject to copyright.
Behavioral/Systems/Cognitive
Motorcortical Excitability and Synaptic Plasticity Is
Enhanced in Professional Musicians
Karin Rosenkranz,
1
Aaron Williamon,
2
and John C. Rothwell
1
1
Sobell Department of Motor Neuroscience and Movement Disorders, Institute of Neurology, London WC1N 3B, United Kingdom, and
2
Royal College of
Music, London SW7 2BS, United Kingdom
Musicians not only have extraordinary motor and sensory skills, but they also have an increased ability to learn new tasks compared with
non-musicians. We examined how these features are expressed in neurophysiological parameters of excitability and plasticity in the
motor system by comparing the results of 11 professional musicians and 8 age-matched non-musicians. Parameters of motor excitability
were assessed using transcranial magnetic stimulation (TMS) to measure motor-evoked potentials (MEPs) together with recruitment of
corticospinal projections [input–output curve (IOcurve)] and of short-latency intracortical inhibition (SICIcurve). Plasticity, here de-
fined as change of synaptic effectiveness, was tested by measuring MEPs and IOcurves after paired associative stimulation (PAS), which
consists of an electric median nerve stimulus repeatedly paired (200 times at 0.25Hz) with a TMS pulse over the hand motor area. Using
an interstimulus interval of 25 ms(PAS25) or 10 ms (PAS10), this leadstolong-term potentiation- or long-term depression-like plasticity,
respectively. Musicians showed steeper recruitment of MEPs and SICI (IOcurve and SICIcurve). Additionally, PAS25 increased and
PAS10 decreased the MEP amplitudes and the slope of the IOcurves significantly more in musicians than in non-musicians. This is
consistent with a wider modification range of synaptic plasticity in musicians. Together with the steeper recruitment of corticospinal
excitatory and intracortical inhibitory projections, this suggests that they regulate plasticity and excitability with a higher gain than
normal. Because some of these changes depend on age at which instrumental playing commenced and on practice intensity, they may
reflect an increase in number and modifiability of synapses within the motor area caused by long-term musical practice.
Key words: motor cortex; excitability; plasticity; musician; transcranial magnetic stimulation; motor training
Introduction
Playing a musical instrument is one of the most complex skills a
human can achieve and is the result of intense sensory and motor
training often started at early age. Compared with non-
musicians, the musicians’ brain shows structural and functional
changes (Elbert et al., 1995; Schlaug et al., 1995; Jancke et al.,
2000; Munte et al., 2002; Gaser and Schlaug, 2003; Haslinger et
al., 2004; Ragert et al., 2004; Bengtsson et al., 2005; Bangert and
Schlaug, 2006). For example, magnetic resonance (MR) imaging
studies have highlighted an increase in gray matter volume in the
sensorimotor cortex in musicians (Gaser and Schlaug, 2003) and
an increased intrasulcal length of the precentral gyrus (Amunts et
al., 1997); magneto-encephalographic studies have shown en-
larged cortical somatosensory representations of fingers (Elbert
et al., 1995). Not only do musicians have impressive motor skills
and better somatosensory discrimination abilities than non-
musicians, they also show enhanced motor and sensory learning
capabilities. Their tactile discrimination skills are improved pro-
portionally more by a tactile costimulation protocol than in non-
musicians (Ragert et al., 2004). Similarly, musicians who started
musical training before 7 years of age learn a timed motor se-
quence task much better than musicians who started later or
non-musicians (Watanabe et al., 2007).
In this study, we focused on the motor system and asked
whether basic neurophysiological measures of motor cortex ex-
citability and synaptic plasticity in musicians are changed in such
a way as to contribute to their enhanced motor skills and learning
abilities. We used transcranial magnetic stimulation (TMS)
applied in single and paired-pulse paradigms to measure basic
excitability parameters: input– output curves (IOcurves) test
stimulus intensity-dependent recruitment of corticospinal pro-
jections to the small hand muscles (Ridding and Rothwell, 1997),
whereas recruitment of short-latency intracortical inhibition
(SICI) is measured in response to different conditioning stimulus
intensities (SICIcurves) (Kujirai et al., 1993; Ziemann et al., 1996;
Orth et al., 2003).
Motor cortical plasticity was assessed with the technique of
paired associative stimulation (PAS), which consists of a median
nerve stimulus followed by a single TMS pulse applied over the
hand area of the contralateral motor cortex. Like associative syn-
aptic conditioning in reduced animal preparations, the effects of
PAS in humans depend on the interval between the sensory input
and the magnetic pulse. Long-term potentiation (LTP)-like ef-
fects on cortical synapses occur with an interval of 25 ms between
Received Oct. 18, 2006; revised March 31, 2007; accepted April 2, 2007.
This work was supported by Action Medical Research UK, the Dystonia Medical Research Foundation, the
Bachmann-Strauss Dystonia and Parkinson Foundation, and the Medical Research Council.
Correspondence should be addressed to Dr. Karin Rosenkranz, Sobell Department of Motor Neuroscience and
Movement Disorders, Institute of Neurology, 8-11 Queen Square, London WC1N 3B, UK. E-mail:
k.rosenkranz@ion.ucl.ac.uk.
DOI:10.1523/JNEUROSCI.0836-07.2007
Copyright © 2007 Society for Neuroscience 0270-6474/07/275200-07$15.00/0
5200 • The Journal of Neuroscience, May 9, 2007 • 27(19):5200 –5206
stimuli (PAS25), and long-term depression (LTD)-like effects
emerge with an interval of 10 ms (PAS10) (Stefan et al., 2002;
Wolters et al., 2003; Classen et al., 2004). Application of the PAS
protocols interacts with short-term motor learning in humans
(Ziemann et al., 2004; Stefan et al., 2006), and therefore the
method seems likely to test circuits involved in natural behaviors.
We hypothesize that, as an expression of their structural and
functional adaptation to musical training, musicians will show
different recruitment of corticospinal and intracortical projec-
tions and, furthermore, that their synaptic plasticity will be
higher than normal and could potentially contribute to their in-
creased learning abilities in motor and sensory tasks.
Materials and Methods
Subjects. Eight healthy right-handed non-musicians (four females) aged
23–35 years and 11 musicians (seven females) aged 20 –35 years gave
informed consent for the study (see Table 1 for details on musicians’
characteristics), which was approved by the joint ethics committee of the
Institute of Neurology and National Hospital for Neurology and Neuro-
surgery, London. All experiments conform to the Declaration of Hel-
sinki. The experiments were all performed in the morning, and none of
the musicians had played their instrument directly beforehand. Subjects
were comfortably seated in an armchair with their right forearm prona-
ted and their hand relaxed on an armrest.
TMS. TMS was performed using two Magstim 200 stimulators (Mag-
stim, Dyfed, UK) connected to a figure-eight-shaped coil with an internal
wing diameter of 7 cm by a Y-cable. The coil was held with the handle
pointing backward and laterally ⬃45° to the interhemispheric line to
evoke anteriorly directed current in the brain and was optimally posi-
tioned to obtain motor-evoked potentials (MEPs) in the abductor polli-
cis brevis muscle (APB). Stimulation intensities are quoted in the text as
a percentage of maximal stimulator output (⫾SE).
Electromyographic recording. Surface electromyographic recordings in
a belly-to-tendon montage were made from the APB, the first dorsal
interosseus, and the abductor digiti minimi muscle of the right hand. The
raw signal was amplified and filtered with a bandpass filter of 30 Hz to 1
kHz (Digitimer, Welwyn Garden City, UK). Signals were digitized at 2
kHz (CED Power1401; Cambridge Electronic Design, Cambridge, UK)
and stored on a laboratory computer for off-line analysis.
Experimental parameters. The MEPs and IOcurves were measured be-
fore and after the PAS protocol (see below). At the beginning of each
experiment, the stimulus intensity that evokes an MEP of ⬃1 mV peak-
to-peak amplitude (SI1mV) was defined. Using SI1mV, 15 MEPs were
recorded before and after each PAS protocol, and the mean amplitude
was calculated for the data obtained before and after PAS in each single
subject. For the IOcurves, the intensities of the single TMS stimuli were
individually adapted according to the predefined SI1mV. Ten MEPs each
were recorded with 50, 70, 80, 90, 100 (⫽SI1mV), 110, 120, 130, and
150% of SI1mV. For each subject, the peak-to-peak amplitudes were
measured on each single trial to calculate the mean amplitude for each
stimulation intensity.
The SICIcurve was measured on the first ex-
perimental session before the PAS protocol. A
subthreshold conditioning stimulus preceded
the suprathreshold test stimulus for 3 ms (Ku-
jirai et al., 1993). The test stimulus intensity was
always set at SI1mV, but the intensity of the
conditioning stimulus was varied. Ten trials
each were recorded in blocks with 70, 80, and
90% of the active motor threshold (aMT). The
aMT was defined as the minimum intensity
needed to evoke an MEP of ⬎200
Vin5of10
trials in the tonically active APB (⬃20% of
maximal contraction as assessed visually on an
oscilloscope). Before, between, and after the
blocks, five single test pulses were given to en-
sure that the unconditioned MEP size was sta-
ble. The peak-to-peak amplitude of the condi-
tioned and test MEPs was measured for each single trial to calculate the
mean amplitude and percentage SICI for the three different conditioning
stimulus intensities.
PAS. PAS consisted of 200 electrical stimuli of the right median nerve
at the wrist paired with a single TMS pulse over the hot spot of the APB
area of the left hemisphere with a rate of 0.25 Hz. Electrical stimulation
was applied through a bipolar electrode (cathode proximal), using
square-wave pulses (duration, 0.2 ms) at an intensity of three times the
perceptual threshold. TMS was delivered through a figure-eight coil (di-
ameter of each wing, 70 mm) connected to a Magstim 200 magnetic
stimulator (Magstim) and was held in the same position as described
above. Stimulation was applied at an intensity adjusted to evoke an MEP
of SI1mV in the relaxed APB. Subjects took part in two experimental
sessions that were separated by at least 1 week. In randomized order, the
effect of PAS given with an interstimulus interval of 25 ms (PAS25) and of
10 ms (PAS10) between peripheral and TMS stimulus was tested. The
former has been shown previously to induce long-lasting MEP increase
(Stefan et al., 2000, 2002), and the latter has previously shown an MEP
decrease (Wolters et al., 2003). Subjects were instructed to look at their
stimulated hand and count the peripheral electrical stimuli they per-
ceived. The MEPs evoked in the APB were displayed on-line during the
intervention to control for the correct coil position and stored for off-line
analysis.
Data analysis and statistics. The comparability of the subject groups for
age was tested by an unpaired t test. The aMT and intensities of condi-
tioning and test stimuli (SI1mV) are given as the percentage of stimulator
output. Comparability between the groups, as well as within the groups
for each experimental part (PAS25 and PAS10), was tested by means of
unpaired and paired two-tailed t tests, respectively. For statistical analy-
sis, repeated-measures ANOVA, with the between-subject factor
“group,” were performed as two-way ANOVAs. Where appropriate, ad-
ditional one-way ANOVAs and post hoc t tests were applied, as outlined
in detail in Results.
To simplify the data set obtained by measuring the IOcurves and the
SICI curves, the slopes of these curves measured as the steepness of the
linear regression line through the given data points [for input–output
slope (IOslope) between 90 and 130% SI1mV; for the SICIcurves for all
three SICI levels measured with 70, 80, and 90% aMT] were calculated.
For the data obtained in musicians, correlation analyses were per-
formed on performance (starting age, practice intensity) and experimen-
tal parameters, and Pearson’s correlation coefficient (r) was calculated
(see Results for details). We set the significance levels for the ANOVAs to
p ⬍ 0.01 to correct for multiple comparisons and for the t tests to p ⬍
0.05. All data are given as means ⫾ SE.
Results
None of the subjects experienced any side effects from TMS dur-
ing the experiments. There was no difference between the groups
in age (non-musicians, 27.6 ⫾ 2.2 years; musicians, 24.7 ⫾ 1.2
years), aMT (non-musicians, 38.8 ⫾ 2.7% stimulator output;
musicians, 36.1 ⫾ 1.8% stimulator output), or intensity of the
Table 1. Characteristics of musicians
Musician
Male (M)/
female (F) Age (years) Instrument
Age when
started playing
Practice intensity
over last year (hours/day)
1 F 25 Piano 6 3
2 M 26 Piano 4 3
3 M 24 Piano 5 4
4 F 27 Piano 7 4
5 M 19 Piano 5 4.5
6 F 18 Piano 4 4
7 F 23 Piano/organ 5 4.5
8 F 24 Recorder 4 4.5
9 M 34 Trumpet 10 0.5
10 F 23 Trombone 11 4
11 F 25 Guitar 12 5
Rosenkranz et al. • Motorcortical Excitability and Plasticity in Musicians J. Neurosci., May 9, 2007 • 27(19):5200 –5206 • 5201
test stimulus needed to evoke an MEP of ⬃1 mV peak-to-peak
amplitude in the APB (non-musicians, 54.3 ⫾ 4.4% stimulator
output; musicians, 54.7 ⫾ 3.5% stimulator output). The number
of peripheral nerve stimuli counted during the PAS protocols was
not different in non-musicians (PAS25, 200 ⫾ 1; PAS10, 201 ⫾ 1)
and musicians (PAS25, 200 ⫾ 1; PAS10, 199 ⫾ 1).
Effect of PAS25 and PAS10 on MEPs
Figure 1 illustrates the main finding with original recordings
from one representative non-musician and one musician before
and after PAS25 and PAS10. In both subjects, MEPs recorded in
the APB are facilitated after PAS25 and suppressed after PAS10.
However, the effect is much larger in the musician than the
non-musician.
Figure 2 shows the mean (⫾SE) data in both groups of sub-
jects for all three intrinsic hand muscles. In each muscle, facilita-
tion of MEPs after PAS25 is expressed as a percentage increase in
the baseline MEP; suppression after PAS10 is expressed as per-
centage reduction in MEP.
In each subject, baseline MEPs were of similar size before
PAS25 and PAS10 (t tests, p ⬎ 0.46); this allowed this simplified
display of the PAS25 and PAS10 effect as double-headed arrows,
indicating the total range of modulation by the PAS interventions
(PAS range).
In both groups of subjects, PAS produced a larger effect on MEPs
in the median nerve-innervated APB than the other two muscles, but
in all cases, the effects were greater in musicians than in non-
musicians, which is most clearly seen by comparing the PAS range.
This was confirmed in the two-way ANOVA on the PAS
range, with the factors group and “muscle.” There was a signifi-
cant main effect of both group (F
(1,4)
⫽ 3.41; p ⫽ 0.0007) and
muscle (F
(2,8)
⫽ 25.60; p ⫽ 0.0003) but no significant interaction,
indicating that the effect of PAS was distributed similarly among
the recorded hand muscles in the two groups, although it was
stronger in the musicians. A subsequent one-way ANOVA for
each group separately showed a significant effect of muscle in
both groups (non-musicians: F
(2,10)
⫽ 18.89, p ⫽ 0.004; musi
-
cians: F
(2,16)
⫽ 15.71, p ⫽ 0.0002), showing that the distribution
of the PAS effect, which was strongest in the APB and weakest in
the ADM, is significant in both groups. Comparing musicians
and non-musicians, PAS25-induced facilitation, PAS10-induced
inhibition, and PAS range were in all recorded muscles signifi-
cantly higher in musicians (t tests, p ⬍ 0.001).
Effects of PAS25 and PAS10 on IOcurves
Figure 3, A and B, shows the effect of PAS25 and PAS10 on the
IOcurves of the APB in both groups of subjects. The slopes of the
curves are increased after PAS25 and decreased after PAS10.
Interestingly the slopes of the baseline IOcurves appear to be
steeper in musicians. To analyze these data, we first noted that in
each group the baseline IOcurves were very similar before PAS25
and PAS10. This was confirmed by a within-group two-way
ANOVA with the factors “stimulus intensity” and “protocol.”
There was a significant main effect of stimulus intensity (F
(1,8)
⫽
72.19; p ⬍ 0.0001), but no effect of protocol, nor a significant
interaction. This allowed us to pool the two sets of baseline data in
each group into a single baseline IOcurve as shown in Figure 3C.
The slope of the IOcurve for musicians was steeper than that for
the non-musicians. Thus, two-way ANOVA with the factors
group and stimulus intensity showed a significant main effect for
both factors (group: F
(1,15)
⫽ 20.61, p ⬍ 0.0001; stimulus inten
-
sity: F
(1,8)
⫽ 107.23, p ⬍ 0.0001), as well as a significant interac
-
tion (F
(8,120)
⫽ 3.82; p ⫽ 0.0004). Post hoc tests at each intensity
revealed that MEPs were larger in musicians at stimulus intensi-
ties of 120 and 130% aMT (t test, p ⬍ 0.025). Given that the slope
of the IOcurve was approximately linear between 90 and 130%
aMT (Fig. 3A–C, gray box), we conducted the remaining analysis
on the slopes of the curves in each condition.
Figure 3D shows the mean slope of the IOcurves before and
after PAS25 or PAS10 for non-musicians and musicians. As
noted above using the ANOVA analysis, the slope of the baseline
IOcurve in musicians was significantly steeper than in non-
musicians (t test, p ⬍ 0.01). After PAS25, the slope increased in
both groups, although this was only significant in musicians (t
test, p ⫽ 0.0001); PAS10 decreased the slope significantly in both
groups (t tests; non-musicians, p ⫽ 0.0125; musicians, p ⫽
0.0002). Figure 3E shows the percentage change in slope relative
to baseline after PAS. In musicians, PAS changed the slope more
than in non-musicians. A two-way ANOVA with the factors
group and protocol showed no significant main effect of group
(F
(1,5)
⫽ 0.19; p ⫽ 0.678) but a significant main effect of protocol
(F
(1,5)
⫽ 80.46; p ⫽ 0.0003) and a significant group ⫻ protocol
interaction (F
(1,5)
⫽ 31.62; p ⫽ 0.0025). This was because the
PAS25 protocol produced a stronger increase (t test, p ⫽ 0.03)
and the PAS10 protocol produced a stronger decrease in slope of
the IOcurve in musicians compared with non-musicians (t test,
p ⫽ 0.01).
Figure 1. Average MEP recordings from the APB of one representative non-musician and
one musician. The TMS intensity was adjusted to evoke a baseline MEP with peak-to-peak
amplitude of 1 mV in the APB asthe main target muscle. After PAS25, the MEP amplitudes were
enhanced, and after PAS10, they were reduced in both subjects. However, both effects were
much larger in the musician than the non-musician.
non-musicians
50%
75%
100%
125%
150%
175%
200%
PAS 10 PAS 25
APB FDI ADM
54%
40%
23%
percentage of baseline
value
musicians
50%
75%
100%
125%
150%
175%
200%
PAS 10 PAS 25
APB FDI ADM
130%
100%
88%
percentage of baseline
value
Figure 2. Mean MEP (⫾SE) in non-musicians and musicians for all three intrinsic hand
muscles. Changes after PAS25 (gray) and PAS10 (black) are expressed as a percentage increase/
decrease in the baseline MEP. The double-headed arrows show the mean total range of modu-
lation by the PAS interventions. In both groups of subjects, PAS produced the largest effects on
MEPs in the median nerve-innervated APB than the other two muscles. But comparing both
groups, the effects were greater in musicians than in non-musicians.
5202 • J. Neurosci., May 9, 2007 • 27(19):5200 –5206 Rosenkranz et al. • Motorcortical Excitability and Plasticity in Musicians
SICIcurves
The amplitude of the MEP evoked in APB by the test pulse alone
(SI1mV) was the same in non-musicians and musicians. This
allows to display the SICI data obtained with a conditioning stim-
ulus intensity of 70, 80, and 90% aMT as percentage of control
values obtained with the test stimulus alone (Fig. 4). Increasing
the intensity of the conditioning stimulus increases the amount
of inhibition in both groups, but the slope of the relationship is
steeper in musicians. This was confirmed
by a two-way ANOVA with the factors
group and “conditioning pulse intensity”.
There was a significant main effect of condi-
tioning pulse intensity (F
(2,10)
⫽ 821.18; p ⬍
0.0001), as well as a significant interaction
(ANOVA; F
(2,10)
⫽ 4.76; p ⫽ 0.035). The fact
that there was no significant main effect of
group is likely to be attributable to the fact
that musicians had significantly less inhibi-
tion at 70% aMT (t test, p ⫽ 0.004) and sig-
nificantly more inhibition at 90% aMT (t
test, p ⫽ 0.015) with no difference between
groups at 80% aMT (t test, p ⫽ 0.14). We
also calculated the slope of the SICIcurve and
found that it was steeper in musicians than in
non-musicians (t test, p ⫽ 0.0056).
Correlations with performance
parameters in musicians
Figure 5, A and B, shows the correlation
between the age at which instrumental
playing was commenced with the experi-
mental parameters SICIslope and IOslope
before and after PAS. There were signifi-
cant correlations between starting age and
the slope of the SICIcurve (r ⫽ 0.86; p ⫽
0.0008) and between starting age and the
normalized slope of the IOcurve after
PAS25 (r ⫽⫺0.63; p ⫽ 0.036): the earlier
playing was started, the higher was the in-
crease in slope of the IOcurve after PAS25 and the steeper was the
slope of the SICIcurve.
However, the seven pianists (Fig. 5A,B, filled symbols) started
playing earlier than the other four instrumentalists (Fig. 5A,B,
open symbols); thus, an instrument-specific effect cannot be
completely ruled out, although the correlation between the start-
ing age and the IOslope change after PAS was significant for
non-pianists alone (r ⫽⫺0.97; p ⫽ 0.034).
Figure 5C shows the correlation between the actual amount of
playing within the last 5 years with the PAS range: the more
intensive the musician played, the wider was the range of MEP
changes induced by the PAS protocols (r ⫽ 0.62; p ⫽ 0.034). This
correlation was independent of the instrument and was signifi-
cant for pianists (r ⫽ 0.95; p ⫽ 0.0008) and non-pianists (r ⫽
0.73; p ⫽ 0.041).
Discussion
There were two main findings in the present data. First, the
IOcurve and the SICIcurve were steeper in musicians compared
with non-musicians. Second, PAS25- and PAS10-induced effects
on MEP amplitudes and slope of the IOcurve were stronger in
musicians. These features are compatible with the idea that both
motorcortical excitability and plasticity are heightened in musi-
cians. Because some of these changes depend on the age at which
instrumental playing commenced, we hypothesize that they re-
flect adaptation to long-term musical practice.
Differences of baseline excitability in non-musicians
and musicians
The slope of the IOcurve depends on the distribution of excitabil-
ity in populations of motoneurons and interneurons at spinal
and cortical sites and is therefore influenced by many factors
non-musicians
50% 70% 80% 90% SI1mV110% 120% 130% 150%
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
PAS 10 after
PAS 10 before
PAS 25 after
PAS 25 before
stimulus intensity
MEP (mV)
musicians
50% 70% 80% 90% SI1mV110% 120% 130% 150%
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
PAS 10 after
PAS 10 before
PAS 25 after
PAS 25 before
stimulus intensity
MEP (mV)
IOcurve baseline
50% 70% 80% 90% SI 1mV110% 120% 130% 150%
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
non-musicians
musicians
*
*
stimulus intensity
MEP (mV)
slope of IOcurve
0.00
0.25
0.50
0.75
1.00
ns
**
**
**
*
*
bef aft bef aft bef aft bef aft
PAS 25 PAS10 PAS25 PAS10
non-musicians musicians
slope
normalised slope of IOcurve
50%
100%
150%
p=0.01
p=0.03
PAS 25 PAS 10 PAS 25 PAS 10
non-musicians musicians
percentage of baseline
A
B
C
D
E
slope
0.0
0.2
0.4
0.6
0.8
non-musicians
musicians
p=0.01
Figure 3. IOcurves in both groups before and after PAS. A–C, The mean MEP amplitude (⫾SE) as given on the y-axis against
the stimulus intensity given on the x-axis (in percentage of SI1mV). A, B, IOcurves measured before and after PAS25 and PAS10,
respectively, for non-musicians (A) and for musicians (B). Musicians showed a steeper increase in both of the IOcurves measured
beforePAS,aswellasafterPAS25.Inboth groups, the IOcurves measuredbeforethePAS were not significantly different.C,Pooled
IOcurves before PAS for each group separately. The slope of the curve has been calculated for the approximately linear part
between 90 and 130% SI1mV and is shown in the inserted column graph for non-musicians (gray) and musicians (black). The
mean MEP amplitudes were significantly higher in musicians at stimulus intensities of 110 –130% SI1mV (t test, *p ⱕ 0.05), as
was the slope of the pooled IOcurve (t test, p ⫽ 0.01). D, The slopes for all IOcurves as shown in A and B for non-musicians and
musicians. Statistical results of direct comparisons of the slopes before and after PAS within the groups as well as of direct
comparisons between the groups are shown (t test; *p ⱕ 0.05; **p ⱕ 0.01). E, The slopes after PAS as a percentage of the slope
measured before. Compared with non-musicians, the musicians showed a significantly stronger slope increase after PAS25 and
decrease after PAS10 (t test; p values shown).
SICIcurve
70% aMT 80% aMT 90% aMT
25%
50%
75%
100%
non-musicians
musicians
*
*
conditioning pulse intensity
inhibition of test
response
slope
-0.5
-0.4
-0.3
-0.2
-0.1
-0.0
non-musicians
musicians
p = 0.0056
Figure 4. The SICI obtained with a conditioning pulse intensity of 70, 80, and 90% aMT is
shown as the percentage of MEP evoked by the test pulse alone (⫾SE). With a conditioning
stimulus of 70% aMT, musicians showed less, with 90% aMT more SICI than non-musicians,
whereas at 80% aMT the level of SICI was similar in both groups (t test, *p ⱕ 0.015). The slope
calculated for the SICIcurve described when displaying the amount of SICI against the condi-
tioning stimulus intensity was significantly steeper in musicians (inserted column graph; t test,
p ⫽ 0.0056).
Rosenkranz et al. • Motorcortical Excitability and Plasticity in Musicians J. Neurosci., May 9, 2007 • 27(19):5200 –5206 • 5203
(Ridding and Rothwell, 1997; Boroojerdi
et al., 2001; Jensen et al., 2005). There are
no detailed studies of spinal excitability in
musicians, so it remains possible that their
steeper IOcurves are attributable to
changes in the excitability of spinal mo-
toneurons such that a larger response is
evoked by a given corticospinal volley. Al-
though subthreshold differences in the
distribution of excitability in the spinal
motoneuron pool between musicians and
non-musicians cannot be excluded, we
tried to control for this by expressing stim-
ulus intensities relative to the SI1mV in
APB, which was similar in both groups.
Therefore, it seems likely that some of the
increased slope of the IOcurve in musi-
cians relates to changes in motor cortical
organization.
Effects on the cortex are consistent with
animal studies showing that the late phase of motor skill learning
is associated with increases in the number of synapses in the
corresponding area of the motor cortex (Kleim et al., 2002, 2004).
The increased gray matter density found in MR images of musi-
cians’ brains suggests that a similar process occurs in them. In-
deed, changes in gray matter are larger in musicians who started
earlier (Gaser and Schlaug, 2003), and they reflect the choice of
instrument (Bangert and Schlaug, 2006), consistent with a causal
connection between presumed synaptic growth and the duration
and pattern of training. If so, then it may be that increased syn-
aptic connectivity is one factor that leads to increased recruit-
ment in musicians.
SICI tests the excitability of local GABA
A
ergic inhibitory cir
-
cuits in the motor cortex (Hanajima et al., 1998; Ilic et al., 2002).
The amount of SICI that is evoked depends on the stimulus in-
tensity of the conditioning pulse, describing a “U-shaped” curve
(Kujirai et al., 1993; Ziemann et al., 1996; Ilic et al., 2002; Orth et
al., 2003) of which the present experiments explored the initial
increase in SICI toward its peak level. Recruitment of SICI repre-
sents a tradeoff between recruitment of inhibitory and excitatory
elements as the conditioning intensity rises (Ilic et al., 2002). In
our study, we expressed the intensity of the conditioning stimu-
lus relative to aMT, which had the same absolute value in musi-
cians and non-musicians. This yielded an SICIcurve that was
steeper in musicians: lower conditioning pulse intensities evoked
less SICI than in non-musicians, whereas higher intensities
evoked more SICI. This difference would not have been noted if
we had compared the groups at only median conditioning stim-
ulus intensity, thus reinforcing the importance of measuring the
SICI with different conditioning pulse intensities.
The difference between groups may reflect differences in the
threshold and/or distribution of excitability in populations of
cortical interneurons. If so, we favor the possibility that, like the
IOcurves, increasing the conditioning stimulus intensity recruits
more intracortical inhibitory connections because of increased
synpatic density in the cortex. However, the finding that at lowest
conditioning stimulus intensity the SICI in musicians is not as
strong as it is in non-musicians shows that the threshold above
which these intracortical inhibitory projections get activated is
higher in musicians.
In summary, the combination of an increased slope for both
the corticospinal (IOcurve) and intracortical (SICIcurve) input–
output relationships suggests that regulation of cortical output is
more sensitive, or has a higher gain, in musicians than in non-
musicians: that is, small changes in TMS intensity, either as a
single or conditioning pulse, have a proportionally stronger effect
on cortical output in musicians. This may favor rapid recruit-
ment of corticospinal output during recruitment of the motor
cortex in volitional movement. The fact that SICI is recruited at
higher thresholds, but with increased gain, might complement
this quick recruitment with a powerful brake to prevent un-
wanted spread of activation, consistent with the fact that musi-
cians show more focused motorcortical activation during move-
ment (Jancke et al., 2000; Krings et al., 2000; Lotze et al., 2003;
Meister et al., 2005)
Differences in the effect of PAS in musicians
and non-musicians
To our knowledge, the PAS effect in humans has only been mea-
sured as amplitude change of MEPs evoked by a standard test
pulse (Stefan et al., 2000, 2002, 2006; Wolters et al., 2003; Zi-
emann et al., 2004). However, the interpretation of any differ-
ences in PAS between groups of individuals depends on whether
the slopes of the IOcurves are equal in the groups. For example,
using the known effect of increasing TMS intensity on the MEP
size as a model for the PAS effect, then as illustrated in Figure 6, an
increase in the test pulse by 20% would increase MEP by 50% in
groups with “standard” IOslope. However, if the IOslope were
twice as steep, then the increase in MEP would be 100%.
Because in musicians and non-musicians the slopes of the
baseline IOcurve were different, it was important to test the effect
of PAS over the entire range of input intensities. We found that in
musicians, PAS increased the slope of the IOcurve significantly
more and therefore that PAS had a greater effect than in non-
musicians. If the steeper baseline IOcurve in musicians reflects an
increased number of interneuronal connections caused by adap-
tation to long-term musical training, a proportionally stronger
PAS effect suggests that these synapses also have a higher propen-
sity to undergo changes in synaptic efficacy.
The neural circuits involved in PAS and short-term behavioral
motor learning overlap. Previous short-term motor learning oc-
cludes or reverses the ability of PAS to induce LTP-like plasticity,
whereas it may have no effect or even increase LTD-like plasticity
(Ziemann et al., 2004; Stefan et al., 2006). The musicians tested
here started their musical training at an average age of ⬃6.5 years,
and by the age of 20 years, Ericsson et al. (1993) estimated they
2 3 4 5 6 7 8 9 10 11 12 13 14
-0.50
-0.45
-0.40
-0.35
-0.30
-0.25
-0.20
-0.15
-0.10
r = 0.86
start age (years)
SICI slope
2 3 4 5 6 7 8 9 10 11 12 13 14
0.3
0.5
0.7
0.9
1.1
1.3
1.5
1.7
1.9
2.1
baseline
slope change after PAS 25
slope change after PAS10
r =-0.63
start age (years)
IO slope
0 1 2 3 4 5 6
1.00
1.05
1.10
1.15
1.20
1.25
1.30
1.35
1.40
1.45
1.50
r = 0.62
practice intensity (hours/day)
PAS range
A
B
C
Figure 5. Correlation between musicians’ performance parameter and measures of motor excitability and plasticity. Filled
symbols represent pianists, and open symbols represent other instrumentalists (two brass players, one guitarist, one recorder
player). A, B, The correlation between the age at which instrumental playing was commenced (x-axis) and the SICIslope (A) and
the IOslope (B) before and after PAS. Musicians who started at a younger age had a steeper SICIslope and also a stronger IOslope
increase after PAS25. However, because almost all early starters were pianists, instrument-specific effects cannot be completely
excluded. C, The correlation between the practice intensity (hours/day) (x-axis) and the MEP range, showing that musicians who
practice more intensively had a significantly higher MEP range, an effect that was unrelated to the instrument played. Pearson’s
r and the unbroken lines are given for significant correlations.
5204 • J. Neurosci., May 9, 2007 • 27(19):5200 –5206 Rosenkranz et al. • Motorcortical Excitability and Plasticity in Musicians
will have had an average of 10,000 h of practice. If occlusion
between periods of learning were to occur as in non-musicians,
then one would have to conclude that many of these hours of
practice failed to achieve any effective outcome. However, be-
cause the number of hours of practice is well known to relate to
musical skill, this seems unlikely. It may be that increased suscep-
tibility for synaptic potentiation/depotentiation in musicians
represents an adaptation to these learning demands and prevents
occlusion from occurring. This might enable them to adapt mo-
tor performance quickly while preserving a high level of profi-
ciency (e.g., when playing on an unfamiliar instrument).
Nature versus nuture?
Our experiments do not directly assess whether increased synap-
tic plasticity is produced by musical training, as a complex motor
task and also a multimodal-sensory and emotional experience, or
whether it is a genetic trait of successful musicians. However,
some parameters correlated significantly with the age at which
playing commenced as well as with practice intensity over the last
5 years.
Similarly, structural and functional changes in the brain of
musicians are linked to performance parameters (Elbert et al.,
1995; Gaser and Schlaug, 2003; Bangert and Schlaug, 2006). Al-
though this does not completely rule out any preselectional bias,
it seems likely that some of the changes of excitability and plas-
ticity are a consequence of long-term musical practice.
Beneficial versus maladaptive?
Increased LTP-/LTD-like effects of PAS have been described in
focal hand dystonia patients, where it is proposed that they may
contribute to development of symptoms (Quartarone et al., 2006;
Weise et al., 2006) and has even been interpreted as endopheno-
typic trait for focal dystonia (Quartarone et al., 2006).
However, the present data show increased plasticity in highly
skilled musicians, implying that it can be associated with either
maladaptive or beneficial changes. If so, then one tentative con-
clusion must be that other factors determine which predomi-
nates. One factor that may influence the behavioral effectiveness
of enhanced PAS is the corresponding gain of inhibitory mecha-
nisms. The steep slope of the SICIcurve in musicians may allow
more effective control of enhanced plasticity and benefit perfor-
mance. This might be not the case in focal dystonia, because some
studies have described reduced SICI at rest (Ridding et al., 1995;
Gilio et al., 2003), although this finding is inconsistent (Stinear
and Byblow, 2004a,b; Butefisch et al., 2005; Rosenkranz et al.,
2005).
Alternatively, aberrant plasticity might not have an early and
primary pathogenic role in focal dystonia but evolve secondarily,
a consequence rather than a cause of the disease (Weise et al.,
2006).
Conclusion
In musicians, motorcortical excitability operates with a higher
gain than normal. Furthermore, they show a higher sensitivity
toward induction of plasticity using the PAS protocol. These
changes may represent a beneficial adaptation in response to
long-term musical training and support their excellent move-
ment skills.
References
Amunts K, Schlaug G, Jancke L, Stein H (1997) Motor cortex and hand
motor skills: structural compliance in the human brain. Hum Brain Mapp
5:206–215.
Bangert M, Schlaug G (2006) Specialization of the specialized in features of
external human brain morphology. Eur J Neurosci 24:1832–1834.
Bengtsson SL, Nagy Z, Skare S, Forsman L, Forssberg H, Ullen F (2005)
Extensive piano practicing has regionally specific effects on white matter
development. Nat Neurosci 8:1148 –1150.
Boroojerdi B, Ziemann U, Chen R, Butefisch CM, Cohen LG (2001) Mech-
anisms underlying human motor system plasticity. Muscle Nerve
24:602–613.
Butefisch CM, Boroojerdi B, Chen R, Battaglia F, Hallett M (2005) Task-
dependent intracortical inhibition is impaired in focal hand dystonia.
Mov Dis 20:545–551.
Classen J, Wolters A, Stefan K, Wycislo M, Sandbrink F, Schmidt A, Kunesch
E (2004) Paired associative stimulation. Suppl Clin Neurophysiol
57:563–569.
Elbert T, Pantev C, Wienbruch C, Rockstroh B, Taub E (1995) Increased
cortical representation of the fingers of the left hand in string players.
Science 270:305–307.
Ericsson KA, Krampe RT, Tesch-Ro¨mer C (1993) The role of deliberate
practice in the acquisition of expert performance. Psychol Rev
100:363–406.
Gaser C, Schlaug G (2003) Brain structures differ between musicians and
non-musicians. J Neurosci 23:9240 –9245.
Gilio F, Rizzo V, Siebner HR, Rothwell JC (2003) Effects on the right motor
hand-area excitability produced by low-frequency rTMS over human
contralateral homologous cortex. J Physiol (Lond) 551:563–573.
Hanajima R, Ugawa Y, Terao Y, Sakai K, Furubayashi T, Machii K, Kanazawa
I (1998) Paired-pulse magnetic stimulation of the human motor cortex:
differences among I waves. J Physiol (Lond) 509:607– 618.
Haslinger B, Erhard P, Altenmueller E, Hennenlotter A, Schwaiger M, Graefin
von Einsiedel H, Rummeny E, Conrad B, Ceballos-Baumann AO (2004)
Reduced recruitment of motor associtation areas during bimanual coor-
dination in concert pianists. Hum Brain Mapp 22:206–215.
Ilic TV, Meintzschel F, Cleff U, Ruge D, Kessler KR, Ziemann U (2002)
Short-interval paired-pulse inhibition and facilitation of human motor
cortex: the dimension of stimulus intensity. J Physiol (Lond)
545:153–167.
Jancke L, Shah NJ, Peters M (2000) Cortical activations in primary and
secondary motor areas for complex bimanual movements in professional
pianists. Brain Res Cogn Brain Res 10:177–183.
Figure 6. Effect of the baseline slope of the IOcurve on the MEP effects of PAS25. The dia-
gram shows hypothetical IOcurves of the two different groups of subjects (gray and black) that
differ in slope. In this example, PAS25 causes an equivalent 20% increase in the effectiveness of
a TMS stimulus. This leads to a 50% increase in MEP amplitude in the gray group but to a 100%
increase in the black group. Thus, if the PAS25 effect was only measured by its effect on MEPs
elicited at a single intensity of test stimulus (e.g., a preinterventional SI1mV), the result would
suggest that PAS25 is more powerful in the black group than in the gray group. However, the
difference may simply reflect a difference in initial IOcurves rather than a difference in synaptic
plasticity. A more complete demonstration of increased synaptic plasticity involves showing
that PAS increases the slope of the IOcurve (see text).
Rosenkranz et al. • Motorcortical Excitability and Plasticity in Musicians J. Neurosci., May 9, 2007 • 27(19):5200 –5206 • 5205
Jensen JL, Marstrand PCD, Nielsen JB (2005) Motor skill training and
strength training are associated with different plastic changes in the cen-
tral nervous system. J Appl Physiol 99:1558 –1568.
Kleim JA, Barbay S, Cooper NR, Hogg TM, Reidel CN, Remple MS, Nudo RJ
(2002) Motor learning-dependent synaptogenesis is localized to func-
tionally reorganized motor cortex. Neurobiol Learn Mem 77:63–77.
Kleim JA, Hogg TM, Vandenberg PM, Cooper NR, Bruneau R, Remple M
(2004) Cortical synaptogenesis and motor map reorganization occur
during late, but not early, phase of motor skill learning. J Neurosci
24:628– 633.
Krings T, Topper R, Foltys H, Erberich S, Sparing R, Willmes K, Thron A
(2000) Cortical activation patterns during complex motor tasks in piano
players and control subjects. A functional magnetic resonance imaging
study. Neurosci Lett 278:189 –193.
Kujirai T, Caramia MD, Rothwell JC, Day BL, Thompson PD, Ferbert A,
Wroe S, Asselman P, Marsden CD (1993) Corticocortical inhibition in
human motor cortex. J Physiol (Lond) 471:501–519.
Lotze M, Scheler G, Tan HR, Braun C, Birbaumer N (2003) The musician’s
brain: functional imaging of amateurs and professionals during perfor-
mance and imagery. NeuroImage 20:1817–1829.
Meister I, Krings T, Foltys H, Boroojerdi B, Muller M, Topper R, Thron A
(2005) Effects of long-term practice and task complexity in musicians
and nonmusicians performing simple and complex motor tasks: implica-
tions for cortical motor organization. Hum Brain Mapp 25:345–352.
Munte TF, Altenmuller E, Jancke L (2002) The musician’s brain as a model
of neuroplasticity. Nat Rev Neurosci 3:473– 478.
Orth M, Snijders AH, Rothwell JC (2003) The variability of intracortical
inhibition and facilitation. Clin Neurophysiol 114:2362–2369.
Quartarone A, Siebner HR, Rothwell JC (2006) Task-specific hand dysto-
nia: can too much plasticity be bad for you? Trends Neurosci 29:192–199.
Ragert P, Schmidt A, Altenmuller E, Dinse HR (2004) Superior tactile per-
formance and learning in professional pianists: evidence for meta-
plasticity in musicians. Eur J Neurosci 19:473– 478.
Ridding MC, Rothwell JC (1997) Stimulus/response curves as a method of
measuring motor cortical excitability in man. Electroencephalogr Clin
Neurophysiol 105:340 –344.
Ridding MC, Sheean G, Rothwell JC, Inzelberg R, Kujirai T (1995) Changes
in the balance between motor cortical excitation and inhibition in focal,
task specific dystonia. J Neurol Neurosurg Psychiatry 59:493–498.
Rosenkranz K, Williamon A, Butler K, Cordivari C, Andrew JL, Rothwell JC
(2005) Pathophysiological differences between musician’s dystonia and
writer’s cramp. Brain 128:918 –931.
Schlaug G, Jancke L, Huang Y, Staiger JF, Steinmetz H (1995) Increased
corpus callosum size in musicians. Neuropsychologia 33:1047–1055.
Stefan K, Kunesch E, Cohen LG, Benecke R, Classen J (2000) Induction of
plasticity in the human motor cortex by paired associative stimulation.
Brain 123:572–584.
Stefan K, Kunesch E, Benecke R, Cohen LG, Classen J (2002) Mechanisms of
enhancement of human motor cortex excitability induced by interven-
tional paired associative stimulation. J Physiol (Lond) 543:699 –708.
Stefan K, Wycislo M, Gentner R, Schramm A, Naumann M, Reiners K, Clas-
sen J (2006) Temporary occlusion of associative motor cortical plasticity
by prior dynamic motor training. Cereb Cortex 16:376 –385.
Stinear CM, Byblow WD (2004a) Impaired modulation of intracortical in-
hibition in focal hand dystonia. Cereb Cortex 14:555–561.
Stinear CM, Byblow WD (2004b) Elevated threshold for intracortical inhi-
bition in focal hand dystonia. Mov Disord 19:1312–1317.
Watanabe D, Savion-Lemieux T, Penhune VB (2007) The effect of early
musical training on adult motor performance: evidence for a sensitive
period in motor learning. Exp Brain Res 176:332–340.
Weise D, Schramm A, Stefan K, Wolters A, Reiners K, Naumann M, Classen
J (2006) The two sides of associative plasticity in writer’s cramp. Brain
129:2709–2721.
Wolters A, Sandbrink F, Schlottmann A, Kunesch E, Stefan K, Cohen LG,
Benecke R, Classen J (2003) A temporally asymmetric Hebbian rule
governing plasticity in the human motor cortex. J Neurophysiol
89:2339–2345.
Ziemann U, Rothwell JC, Ridding MC (1996) Interaction between intracor-
tical inhibition and facilitation in human motor cortex. J Physiol (Lond)
496:873–881.
Ziemann U, Ilic TV, Pauli C, Meintzschel F, Ruge D (2004) Learning mod-
ifies subsequent induction of long-term potentiation-like and long-term
depression-like plasticity in human motor cortex. J Neurosci 24:1666 –
1672.
5206 • J. Neurosci., May 9, 2007 • 27(19):5200 –5206 Rosenkranz et al. • Motorcortical Excitability and Plasticity in Musicians
