Cerebral Cortex April 2006;16:537--540

doi:10.1093/cercor/bhj001
Advance Access publication July 20, 2005

Pain Suppresses Spontaneous Markus Ploner, Joachim Gross, Lars Timmermann,

Bettina Pollok and Alfons Schnitzler
Brain Rhythms
Department of Neurology, Heinrich-Heine-University,
40225 Düsseldorf, Germany

The neuronal activity of the resting human brain is dominated by (Mouraux et al., 2003; Ohara et al., 2004; Raij et al., 2004)
spontaneous oscillatory activity of primary visual, somatosensory which, in principle, corresponds to the effect of tactile stimuli.
and motor areas. These spontaneous brain rhythms are related to However, pain is a unique experience which disrupts ongoing
the functional state of a system. A higher amplitude of oscillatory behavior, demands attention and urges the individual to react

Downloaded from http://cercor.oxfordjournals.org/ at University of Aberdeen on November 16, 2014

activity is thought to reflect an idling state, whereas a lower (Melzack and Casey, 1968; Eccleston and Crombez, 1999). Thus,
amplitude is associated with activation and higher excitability of pain broadly interferes with sensory, motor and cognitive
the specific system. Here, we used magnetoencephalography to processes. Correspondingly, pain may not only selectively
investigate the effects of pain on spontaneous brain rhythms. Our modulate the function of the sensorimotor system but of
results show that a focally applied brief painful stimulus globally cortical systems in general. Therefore, we used the high
suppresses spontaneous oscillations in somatosensory, motor and spatial and temporal resolution of magnetoencephalography
visual areas. This global suppression contrasts with the regionally to investigate the global effects of pain on spontaneous
specific suppressions of other modalities and shows that pain oscillatory activity.
induces a widespread change in cortical function and excitability.
This global change in excitability may reflect the alerting function of Materials and Methods
pain which opens the gates for processing of and reacting to stimuli Twelve healthy right-handed male subjects with a mean age of 33 years
of existential relevance. (range 22--41 years) participated in the study. Informed consent was
obtained from all subjects before participation. The study was approved
Keywords: cutaneous laser stimulation, human, magnetoencephalography, by the local ethics committee and conducted in conformity with the
nociception, oscillations, pain, somatosensory Declaration of Helsinki.

Stimulation
Introduction Forty painful cutaneous laser stimuli, which evoke a highly synchro-
nized selective activation of nociceptive afferents without concomitant
From the earliest recordings of the human electroencephalo- activation of tactile afferents (Bromm and Treede, 1984) were delivered
gram, spontaneous oscillatory activity at frequencies around to the dorsum of the right hand. The laser device was a Tm:YAG-laser
10 Hz (alpha-band) and 20 Hz (beta-band) has been consis- (Carl Baasel Lasertechnik, Starnberg, Germany) with a wavelength of
tently observed over primary visual, somatosensory and motor 2000 nm, a pulse duration of 1 ms and a spot diameter of 6 mm. The laser
areas (Berger, 1929; Gastaut, 1952; Hari and Salmelin, 1997; beam was led through an optical fiber from outside into the recording
room. Stimulation site was slightly changed after each stimulus.
Niedermeyer, 1999). In each of these systems oscillations Interstimulus intervals were randomly varied between 10 and 14 s.
show a modality-specific reactivity. The occipital alpha-ryhthm Applied stimulus intensity was 600 mJ, which evoked moderately
is dampened by visual stimuli, whereas alpha- and beta- painful sensations. The subjects passively perceived the stimuli with
oscillations over the sensorimotor cortices — termed mu- closed eyes. In four of the subjects, in an additional recording, the left
rhythm — are attenuated by touch and limb movements (Hari hand was stimulated using the same parameters as in the right-hand
and Salmelin, 1997; Pfurtscheller, 1999). This modality speci- stimulation condition.
ficity is complemented by a spatial specificity with stimulus-
Recordings and Analysis
induced modulations of oscillations occurring predominantly During the recordings the subjects were comfortably seated with closed
in the contralateral hemisphere (Hari and Salmelin, 1997; eyes in a magnetically shielded room. Cortical activity was continuously
Pfurtscheller, 1999). Spatial distribution and reactivity suggest recorded with a Neuromag-122 whole-head neuromagnetometer.
that oscillatory activity is related to the functional state of Signals were digitized at 483 Hz.
a system. A higher amplitude of oscillatory activity is thought to As a first step, time windows and frequency bands of pain-induced
reflect an idling state of a system, whereas a lower amplitude is changes of cortical activity were identified. To this end, time frequency
representations (TFR) were calculated using a Fourier transform
associated with activation of a system (Hari and Salmelin, 1997; approach (Delorme and Makeig, 2004). For each trial the TFR comprised
Niedermeyer, 1999; Pfurtscheller, 1999). In addition, suppres- an epoch from 1500 ms before to 3000 ms after stimulus application. A
sion of oscillatory activity has been related to a higher degree global grand average TFR was obtained by averaging TFRs across trials,
of excitability in the sense of a thalamocortical gate which can sensors and subjects. This global grand average TFR showed prominent
be opened by endogenous or exogenous events (Steriade and pain-induced suppressions of cortical activity in the alpha- (7--15 Hz)
Llinas, 1988). and beta- (15--25 Hz) band in a time window between 500 and 1500 ms
after stimulus application. Thus, further analysis focused on these
Recently, the effects of pain on spontaneous oscillations have
frequency bands and on this time window.
been investigated. Neurophysiological studies revealed that In the next step, locations of pain-induced suppressions of cortical
phasic painful stimuli suppress oscillations over the sensorimo- activity were calculated. To this end cross-spectral density matrices
tor cortex predominantly of the contralateral hemisphere of power changes in the time window between 500 and 1500 ms as

Ó The Author 2005. Published by Oxford University Press. All rights reserved.
For permissions, please e-mail: journals.permissions@oupjournals.org
compared to a baseline period from –1000 to –10 ms were calculated.
From these matrices pain-evoked activity was localized using a spatial
filtering algorithm (Van Veen et al., 1997; Gross et al., 2001). The spatial
filter was employed with a realistic head model to estimate power in the
whole brain, and resulted in individual tomographic power maps with
voxel sizes of 6 3 6 3 6 mm. Further processing of tomographic power
maps was carried out using SPM99 (Wellcome Department of Cognitive
Neurology, Institute of Neurology, London, UK: http://www.fil.ion.
ucl.ac.uk/spm). Individual maps were spatially normalized to Talairach
space using parameters derived from normalization of individual
T1-weighted magnetic resonance images (Friston et al., 1995). Among
the five strongest local power maxima individual power maps consis-
tently showed maxima located in the bilateral central region and in the
occipital cortex. Mean group normalized power maps were calculated
for each of the three regions.
In a third step time courses of pain-induced power changes in the
bilateral central region and in the occipital cortex were determined.
Using the temporal spectral evolution (TSE) method (Salmelin and Hari, Figure 1. Time frequency representation (TFR) of pain-induced modulations of
spontaneous neuronal activity averaged across sensors, trials and subjects. Power

Downloaded from http://cercor.oxfordjournals.org/ at University of Aberdeen on November 16, 2014

1994), signals were band-pass filtered from 7 to 15 and from 15 to 25 Hz
respectively. Filtered signals were rectified, averaged across trials and increases and decreases (DP) from baseline are coded in red and blue, respectively.
across 10 sensors over the bilateral central region and 12 sensors over
the occipital cortex. The signals recorded from these sensors showed
clear modulation of oscillatory activity. Results did not depend on the were localized in the previously identified time window
number of sensors. From the individual time courses group mean time (500--1500 ms) and frequency bands (alpha, beta) relative to a
courses of pain-induced power changes were calculated. For each area 1000 ms prestimulus baseline. Figure 2 shows the group mean
and frequency band 95% confidence intervals of power changes were
calculated as twice the standard deviation of the 1000 ms prestimulus
locations of pain-induced power changes. Foci of suppression of
baseline. spontaneous oscillatory activity were located in the bilateral
For statistical comparison mean amplitudes of pain-induced power sensorimotor cortices and in the occipital cortex. Within the
changes during the time window between 500 and 1500 ms were bilateral sensorimotor cortices, suppressions in the alpha-band
determined for both frequency bands. The lateralization of pain-induced were located slightly more posterior than suppressions in the
modulations was analyzed by comparing mean amplitudes of right- and beta-band corresponding to location in primary somatosensory
left-hemispheric modulations using sequentially Bonferroni-corrected
and motor cortices respectively. Thus, pain suppresses the
two-tailed Wilcoxon signed-rank tests. Lateralization was visualized by
calculating a lateralization ratio (left hemispheric/right hemispheric) of sensorimotor mu-rhythm bilaterally as well as the occipital
pain-induced modulations to right- and left-sided stimulation. alpha-rhythm.
Third, time courses of pain-induced modulations were
Control Experiment calculated for each region and frequency band (Fig. 2). Time
In order to compare the effects of pain and touch on cortical activity courses show that the significant pain-induced suppression of
electrical stimulation of tactile afferents was carried out in 12 healthy about 2000 ms duration applies to the 10 Hz ‘sensory’ and 20 Hz
right-handed subjects (4 female, 8 male, mean age 32 years, range 24--44 ‘motor’ components of the mu-rhythm bilaterally and to the
years). Electrical stimuli were applied by using ring electrodes attached occipital alpha-rhythm. The suppression of the mu-rhythm is
to the middle and end phalanx of the index finger of the right hand.
stronger in the right, ipsilateral hemisphere than in the left,
Stimuli were rectangular constant voltage pulses of 0.3 ms duration with
an interstimulus interval of 3 s. Stimulus intensity was adjusted to 2- to contralateral hemisphere. This contrasts with the effect of
3-fold detection threshold intensity evoking clear and non-painful tactile stimuli applied to the right hand. Tactile stimuli induce
sensations. Time courses of tactile-induced power changes in the a short-lasting suppression of the mu-ryhthm mainly in the left,
bilateral central region and in the occipital cortex were determined contralateral hemisphere and no comparable suppression of the
using the same procedure as for the pain-induced effects. Mean occipital alpha-rhythm. Figure 3 illustrates the lateralization of
amplitudes of tactile-induced power changes were calculated during suppressions of the mu-rhythm to painful and tactile stimulation
a time window between 0 and 1000 ms for both frequency bands.
Statistical analysis and visualization was the same as for the painful
by showing a lateralization ratio (left hemispheric/right hemi-
stimulation condition. spheric) of suppressions. The figure illustrates the right hemi-
spheric lateralization of suppressions to right-sided painful
stimuli and the left hemispheric lateralization of suppressions
Results to right-sided tactile stimulation. To further clarify the lateral-
First, time windows and frequency bands of pain-induced ization of the pain-induced modulations we applied painful
modulations of oscillatory activity were determined. Thus, stimuli to the left hand in four of the subjects. The results show
global grand average time frequency representations (TFR) that left-sided painful stimuli also yield a right-lateralized
were calculated. Figure 1 shows that the brief painful stimuli suppression of the mu-rhythm. Thus, these findings show that
suppress cortical oscillatory activity between 500 and 1500 ms pain-induced modulations of the mu-rhythm are generally
after stimulus application. This suppression occurs in the alpha- lateralized to the right hemisphere and do not reflect an
band (7--15 Hz) and in the beta-band (15--25 Hz). [Note that the ipsilateral dominance.
early power increase below the alpha-band reflects evoked
responses which have been analyzed previously (Ploner et al.,
1999, 2000, 2002; Timmermann et al., 2001).] Discussion
Second, we determined locations of pain-induced suppres- The present findings reveal that brief painful stimuli yield
sions of cortical oscillations. Using a time-domain variant of the a global right-lateralized suppression of spontaneous oscilla-
DICS method (Gross et al., 2001) pain-induced power changes tions in sensory and motor systems.

538 Pain Suppresses Spontaneous Brain Rhythms d

Ploner et al.
 Downloaded from http://cercor.oxfordjournals.org/ at University of Aberdeen on November 16, 2014
Figure 2. Group mean locations and time courses of pain-induced modulations. Locations of 20 Hz and 10 Hz suppressions are coded in green and red respectively. Time courses of
pain-induced modulations (black lines) are compared to tactile-induced modulations (grey lines). The dotted lines show 95% confidence intervals of modulation for each modality,
area and frequency band.

of cortical areas (Pfurtscheller, 1999) our results demon-

strate that pain induces a widespread change in cortical
function and excitability. This global pain-induced change in
cortical function and excitability may be related to the unique
biological significance of pain which disrupts ongoing behav-
iour, demands attention and urges the individual to react
(Melzack and Casey, 1968; Eccleston and Crombez, 1999).
More specifically, our finding of a global change in excitability
may reflect the alerting function of pain, which may be
mediated by a right-lateralized cortico-subcortical network
dedicated to the detection of salient events (Downar et al.,
2000; Corbetta and Shulman, 2002). The right-sided lateraliza-
tion of this network together with a preponderance of the right
hemisphere in the processing of pain (Hari et al., 1997; Coghill
Figure 3. Lateralization ratio (left hemispheric/right hemispheric) of suppressions to
painful and tactile stimuli. Note that left- and right-lateralized suppressions of the mu- et al., 2001) and negative affect (Davidson, 1995) could well
rhythm correspond to bars to the left and the right side, respectively. Error bars account for the right-hemispheric lateralization of the observed
represent SEM; *P # 0.05, **P # 0.01, sequentially Bonferroni-corrected, two-tailed effects. The alerting function of pain along with a global
Wilcoxon signed rank tests. suppression of spontaneous brain rhythms may ‘open the gates’
of sensory and motor systems and prepare the individual for
processing of and reacting to stimuli of existential relevance.
Our results correspond with recent neurophysiological This pain-induced gating of sensory and motor information may
studies which showed a pain-induced suppression of the mu- be related to the recently described phenomenon of pain-
rhythm (Mouraux et al., 2003; Ohara et al., 2004) lateralized to induced facilitation of sensory (Ploner et al., 2004) and motor
the right, contralateral hemisphere (Raij et al., 2004). However, processing (Raij et al., 2004).
these studies focused on pain-induced effects on the mu-
rhythm and did not investigate global effects of pain on
spontaneous brain rhythms. Other studies investigating the Notes
effects of tonic pain on spontaneous oscillatory activity also This work was supported by grants from the Research Committee
revealed pain-induced decreases in alpha-power and mostly an of the Medical Faculty of the Heinrich-Heine-University and the
VolkswagenStiftung.
increase in beta-power (Backonja et al., 1991; Veerasarn and
Address correspondence to Markus Ploner or Alfons Schnitzler,
Stohler, 1992; Chen and Rappelsberger, 1994; Ferracuti et al., Department of Neurology, Heinrich-Heine-University, Moorenstrasse
1994; Chang et al., 2002). However, the effects of tonic pain 5, 40225 Düsseldorf, Germany. Email: ploner@neurologie.uni-
most probably comprise complex pain-coping strategies and, duesseldorf.de, schnitza@uni-duesseldorf.de.
thus, reflect neural mechanisms distinct from the modulations
induced by the brief painful stimuli of the present study.
References
Further, our results reveal for the first time that the effects of
Backonja M, Howland EW, Wang J, Smith J, Salinsky M, Cleeland CS
pain outreach the modality and topographically specific effects
(1991) Tonic changes in alpha power during immersion of the hand
exerted by other sensory and motor events (Hari and Salmelin, in cold water. Electroencephalogr Clin Neurophysiol 79:192--203.
1997; Pfurtscheller, 1999). Considering that spontaneous oscil- Berger H (1929) Über das Elektroenkephalogramm des Menschen. Arch
lations are related to the functional state and the excitability Psychiat Nervenkr 87:527--570.

Cerebral Cortex April 2006, V 16 N 4 539

Bromm B, Treede RD (1984) Nerve fibre discharges, cerebral potentials Mouraux A, Guerit JM, Plaghki L (2003) Non-phase locked electroen-
and sensations induced by CO2 laser stimulation. Hum Neurobiol cephalogram (EEG) responses to CO2 laser skin stimulations may
3:33--40. reflect central interactions between A partial differential- and C-fibre
Chang PF, Arendt-Nielsen L, Chen AC (2002) Dynamic changes and afferent volleys. Clin Neurophysiol 114:710--722.
spatial correlation of EEG activities during cold pressor test in man. Niedermeyer E (1999) The normal EEG of the waking adult. In:
Brain Res Bull 57: 667--675. Electroencephalography: basic principles, clinical applications, and
Chen AC, Rappelsberger P (1994) Brain and human pain: topographic related fields (Niedermeyer E, Lopes da Silva F, eds), pp. 149--173.
EEG amplitude and coherence mapping. Brain Topogr 7:129--140. Baltimore, MD: Williams & Wilkins.
Coghill RC, Gilron I, Iadarola MJ (2001) Hemispheric lateralization of Ohara S, Crone NE, Weiss N, Lenz FA (2004) Attention to a painful
somatosensory processing. J Neurophysiol 85:2602--2612. cutaneous laser stimulus modulates electrocorticographic event-
Corbetta M, Shulman GL (2002) Control of goal-directed and stimulus- related desynchronization in humans. Clin Neurophysiol
driven attention in the brain. Nat Rev Neurosci 3:201--215. 115:1641--1652.
Davidson RJ (1995) Cerebral asymmetry, emotion, and affective style. In: Pfurtscheller G (1999) EEG Event-related desynchronization (ERD) and
Brain asymmetry (Davidson RJ, Hugdahl K, eds), pp. 361--387. event-related synchronization (ERS). In: Electroencephalography:
Cambridge, MA: MIT Press. basic principles, clinical applications, and related fields
Delorme A, Makeig S (2004) EEGLAB: an open source toolbox for (Niedermeyer E, Lopes da Silva F, eds), pp. 958--967. Baltimore,
analysis of single-trial EEG dynamics including independent compo- MD: Williams & Wilkins.
nent analysis. J Neurosci Methods 134:9--21. Ploner M, Schmitz F, Freund HJ, Schnitzler A (1999) Parallel activation of

Downloaded from http://cercor.oxfordjournals.org/ at University of Aberdeen on November 16, 2014

Downar J, Crawley AP, Mikulis DJ, Davis KD (2000) A multimodal primary and secondary somatosensory cortices in human pain
cortical network for the detection of changes in the sensory processing. J Neurophysiol 81: 3100--3104.
environment. Nat Neurosci 3:277--283. Ploner M, Schmitz F, Freund HJ, Schnitzler A (2000) Differential
Eccleston C, Crombez G (1999) Pain demands attention: a cognitive- organization of touch and pain in human primary somatosensory
affective model of the interruptive function of pain. Psychol Bull cortex. J Neurophysiol 83: 1770--1776.
125:356--366. Ploner M, Gross J, Timmermann L, Schnitzler A (2002) Cortical
Ferracuti S, Seri S, Mattia D, Cruccu G (1994) Quantitative EEG representation of first and second pain sensation in humans. Proc
modifications during the Cold Water Pressor Test: hemispheric Natl Acad Sci USA 99:12444--12448.
and hand differences. Int J Psychophysiol 17:261--268. Ploner M, Pollok B, Schnitzler A (2004) Pain facilitates tactile processing
Friston KJ, Ashburner J, Frith CD, Poline J-B, Heather JD, Frackowiak RSJ in human somatosensory cortices. J Neurophysiol 92:1825--1829.
(1995) Spatial registration and normalization of images. Hum Brain Raij TT, Forss N, Stancak A, Hari R (2004) Modulation of motor-cortex
Mapp 2:1--25. oscillatory activity by painful Adelta- and C-fiber stimuli. Neuroimage
Gastaut H (1952) Etude électrocorticographique de la réactivité des 23:569--573.
rhythmes rolandiques. Rev Neurol 87:176--182. Salmelin R, Hari R (1994) Spatiotemporal characteristics of sensorimo-
Gross J, Kujala J, Hamalainen M, Timmermann L, Schnitzler A, Salmelin R tor neuromagnetic rhythms related to thumb movement. Neurosci-
(2001) Dynamic imaging of coherent sources: Studying neural ence 60:537--550.
interactions in the human brain. Proc Natl Acad Sci USA Steriade M, Llinas RR (1988) The functional states of the thalamus and
98:694--699. the associated neuronal interplay. Physiol Rev 68:649--742.
Hari R, Salmelin R (1997) Human cortical oscillations: a neuromagnetic Timmermann L, Ploner M, Haucke K, Schmitz F, Baltissen R, Schnitzler A
view through the skull. Trends Neurosci 20:44--49. (2001) Differential coding of pain intensity in the human primary
Hari R, Portin K, Kettenmann B, Jousmaki V, Kobal G (1997) Right- and secondary somatosensory cortex. J Neurophysiol 86:1499--1503.
hemisphere preponderance of responses to painful CO2 stimulation Van Veen BD, van Drongelen W, Yuchtman M, Suzuki A (1997) Localization
of the human nasal mucosa. Pain 72:145--151. of brain electrical activity via linearly constrained minimum variance
Melzack R, Casey KL (1968) Sensory, motivational, and central control spatial filtering. IEEE Trans Biomed Eng 44:867--880.
determinants of pain: a new conceptual model in pain. In: The skin Veerasarn P, Stohler CS (1992) The effect of experimental muscle pain
senses (Kenshalo DRJ, ed.). Springfield, IL: Charles C. Thomas. on the background electrical brain activity. Pain 49:349--360.

540 Pain Suppresses Spontaneous Brain Rhythms d

Ploner et al.