Neurophysiology of the Cortical Pain Network: Revisiting the Role of S1 in Subjective Pain Perception Via Standardized Low-Resolution Brain Electromagnetic Tomography (sLORETA)

The Journal of Pain, Vol 9, No 11 (November), 2008: pp 1058-1069 Available online at www.sciencedirect.com

Neurophysiology of the Cortical Pain Network: Revisiting the Role of S1 in Subjective Pain Perception Via Standardized Low-Resolution Brain Electromagnetic Tomography (sLORETA) Rony-Reuven Nir, Rina Lev, Ruth Moont, Yelena Granovsky, Elliot Sprecher, and David Yarnitsky Laboratory of Clinical Neurophysiology, Department of Neurology, Rambam Health Care Campus, and Faculty of Medicine, Technion–Israel Institute of Technology, Haifa, Israel.

Abstract: Multiple studies have supported the usefulness of standardized low-resolution brain electromagnetic tomography (sLORETA) in localizing generators of scalp-recorded potentials. The current study implemented sLORETA on pain event-related potentials, primarily aiming at validating this technique for pain research by identifying well-known pain-related regions. Subsequently, we pointed at investigating the still-debated and ambiguous topic of pain intensity coding at these regions, focusing on their relative impact on subjective pain perception. sLORETA revealed significant activations of the bilateral primary somatosensory (SI) and anterior cingulate cortices and of the contralateral operculoinsular and dorsolateral prefrontal (DLPFC) cortices (P < .05 for each). Activity of these regions, excluding DLPFC, correlated with subjective numerical pain scores (P < .05 for each). However, a multivariate regression analysis (R ⴝ .80; P ⴝ .024) distinguished the contralateral SI as the only region whose activation magnitude significantly predicted the subjective perception of noxious stimuli (P ⴝ .020), further substantiated by a reduced regression model (R ⴝ .75, P ⴝ .008). Based on (1) correspondence of the pain-activated regions identified by sLORETA with the acknowledged imaging-based pain-network and (2) the contralateral SI proving to be the most contributing region in pain intensity coding, we found sLORETA to be an appropriate tool for relevant pain research and further substantiated the role of SI in pain perception. Perspective: Because the literature of pain intensity coding offers inconsistent findings, the current article used a novel tool for revisiting this controversial issue. Results suggest that it is the activation magnitude of SI, which solely establishes the significant correlation with subjective pain ratings, in accordance with the classical clinical thinking, relating SI lesions to diminished perception of pain. Although this study cannot support a causal relation between SI activation magnitude and pain perception, such relation might be insinuated. © 2008 by the American Pain Society Key words: Pain network, somatosensory cortex, intensity coding, event-related potentials, source localization, sLORETA, contact-heat stimulation.

N

europhysiological investigations of pain processing have been conducted using experimental stimuli, evoking synchronized pain-induced responses detected as event-related potentials (ERPs). The most investigated pain-induced extracranial measure is Received April 14, 2008; Revised June 10, 2008; Accepted June 17, 2008. Address reprint requests to Dr. David Yarnitsky, Professor and Head, Department of Neurology, Rambam Health Care Campus, Haifa, POB. 9602 31096, Israel. E-mail: [email protected] 1526-5900/$34.00 © 2008 by the American Pain Society doi:10.1016/j.jpain.2008.06.008

1058

the N2-P2 component recorded at the vertex,13,44 which was found to correlate both with stimulus intensity and subjective pain reports.28,29,37 In the search for the brain generators involved in pain perception, source localization techniques implemented on electroencephalography (EEG) recordings have been used in past years. These techniques aim at solving the inverse problem, that is, the computation of images of electrical neural activity derived from extracranial measurements. Various localization approaches are mostly categorized into either distributed or dipole modeling. The former enables the detection of multiple brain gen-

Nir et al erators under specific conditions, whereas the latter uses the limited number of sources as a parameter rather than an outcome of the method.81 Due to the acknowledged involvement of multiple cortical regions in processing of noxious stimuli, distributed source modeling should appear preferable to dipole modeling in pain research. In an investigation of well-known minimum norm and other linear inverse solutions of distributed modeling used in source localization studies,66 the superiority of standardized low-resolution brain electromagnetic tomography (sLORETA)53,54 over the other methods was primarily attributed to its minimal localization error in the presence of a gaussian noise, and significant spatial activations underlying cortical functions were reported in studies using as many as 32 electrodes.25,61 The current study used sLORETA based on pain-ERPs predominantly to examine the suitability of sLORETA to identify the major regions of the pain matrix and thus validate its use in this field. For this purpose, we searched for sLORETA-based significant regions of activation in response to experimental application of noxious stimuli and their concordance with the previously described imaging-based pain-network. The latter includes the anterior cingulate cortex (ACC), primary (SI) and secondary (SII) somatosensory cortices, insular cortex (IC), and the dorsolateral prefrontal cortex (DLPFC).4,11,12,14,45,56,68 Consequent to this undertaking, we aimed at incorporating the activity in these pain regions in a regression analysis to delineate their relative impact on pain intensity coding, in an attempt to elucidate this still equivocal topic. The chief pain-related regions have been previously investigated for properties of intensity coding.8,15,16,18,20,31,48,55,60,71,72 Several of these studies concentrated on specific cortical regions, such as the somatosensory cortices15,16,71 and the operculoinsular cortex (OIC).31 Other studies characterized activity in several regions but did not perform regression analyses to evaluate their relative contribution to subjective pain perception.8,20,48,58,60 Conversely, regression analyses were performed in 2 positron emission tomography (PET) studies18,72, but conflicting conclusions were reached. The former has suggested multiple regions to correlate with subjective pain perception, including SI, SII, ACC, and insula, whereas the latter has reported that pain intensity coding was attributed to the periventricular gray and the posterior cingulate cortex. These inconsistent results motivated us to use pain-ERPs combined with source localization analysis to substantiate the role of the main pain-related regions in pain intensity coding.

Materials and Methods Subjects Seventeen healthy, right-handed volunteers (10 men, 7 women), 18 to 32 years of age (mean ⫾ SD: 25 ⫾ 3.9) participated in the study. The institutional review board of Rambam Health Care Campus approved the study protocol according to Helsinki Declaration, and written consent was obtained from each subject.

1059

Stimulator Contact heat-evoked potentials (CHEPs) were recorded in response to brief heat stimuli, which were defined as suitable for inducing a well-synchronized cortical response.27–29,32 A contact heat-evoked potential stimulator with a round thermode was used (PATHWAY sensory evaluation system; Medoc Ltd., Ramat-Yishai, Israel), contacting a cutaneous area of 572.5 mm2 (27 mm in diameter). The thermode is comprised of a heating thermofoil (Minco Products, Inc., Minneapolis, MN), which is covered with a 25-␮m layer of thermo-conductive plastic (Kapton, thermal conductivity at 23°C of 0.1– 0.35 W/mK). Two thermocouples are embedded 10 ␮m within this conductive coating, which contacts the skin directly, thus providing an estimate of the skin temperature at the thermode surface. The thermofoil permits a heating rate of up to 70°C/s, and the Peltier device incorporated in the thermode allows a cooling rate of 40°C/s. These maximal values were used in our study. Mean time from onset to target-peak-temperature of 52°C was 295 ⫾ 0.01 ms, and mean stimulus duration was 647 ⫾ 0.001 ms.

Recording of Contact-Heat ERPs CHEPs were recorded from 32 Ag/AgCl electrodes using an electrode cap (Easy Cap Q40; FMS Falk Minow Services, Herrsching, Germany) according to the 10%system (an extended montage of the standard 10%–20% system). All electrodes were referenced to the nose-tip. Extracranial activity was continuously recorded using a Quick Amp EEG system (Brain Products GmbH, Munich, Germany) within a 0.15-Hz and 100 Hz-bandpass and digitized at a sampling rate of 500 Hz. Throughout the experiments, ongoing real-time recordings were monitored, and the impedance at each electrode position was kept below 5 k⍀. A notch filter was set to 50 Hz to reduce electrical interference. An artifact rejection program controlled visual artifacts, with the maximal allowed voltage-step set to 50 ␮V. EEG data were stored on disk and analyzed offline (Recorder and Analyzer programs; Brain Products GmbH). Each recording epoch of 2400 ms included a baseline period of 400 ms before stimulus onset, which was subsequently used for baseline correction of single trials. Stimulus onset was marked by an electrical transistor-transistor logic (TTL) pulse, delivered from the stimulator to the Quick Amp (Brain Products GmbH) at the onset of each stimulus.

Procedure Subjects sat comfortably in a quiet room with an ambient temperature of ⬃23°C. A total of 5 contact-heat stimulation blocks were applied, each block comprised of 60 stimuli of the same intensity (Fig 1). To refrain from inducing thermal sensitization in the stimulated area and to reduce habituation of primary nociceptors, the 60 stimuli comprising each block were divided into two sets of 30 trials, which were separated by a 15-minute break. To further reduce habituation, the thermode was slightly moved to an adjacent area after each stimulus,29 and a 20-minute break was kept between the blocks.

1060

Neurophysiology of the Cortical Pain Network peak (PTP) N2-P2 amplitude was calculated thereafter. The latencies were measured at the peak of response.

sLORETA Source Localization Analysis

Figure 1. Experimental paradigm. Contact-heat stimuli were administered to the left, nondominant volar forearm. Each block of 60 stimuli of the same intensity was divided into 2 sets of 30 stimuli, which were separated by a 15-minute break. Between blocks, a 20-minute break was kept. Within sets, stimuli were delivered at randomized interstimulus intervals (ISIs) of 8 to 13 seconds. Baseline temperature was 35oC.

From a baseline temperature of 35°C, the intensities of the applied 5 blocks of stimuli increased to 35 (control condition), 41, 46, 49, and 52°C. Subjects were unaware of the applied intensities, which were given in random order across subjects. In all subjects, blocks were applied to the left, nondominant proximal volar forearm. Because the thermode was attached manually, attention was paid to maintaining relatively constant pressure on the thermode. To avoid expectancy effects, stimuli were delivered at randomized interstimulus intervals (ISIs) of 8 to 13 seconds. To reduce the initial novelty effect on resulting heat-ERPs, all subjects were given training stimuli at the beginning of the experiment. Since the aim of the current study was to investigate activation magnitudes of pain-related regions in response to noxious stimuli, here we will only report source localization analyses based on comparisons between cortical activations elicited by noxious stimuli of 52°C, which significantly evoked the highest subjective numerical pain scores (NPSs), and innocuous baseline stimuli of 35°C.

Subjective Numerical Pain Scores Subjects were asked to verbally rate subjective perception of each applied stimulus on a 0 to 10 numerical pain scale, in which 0 was defined as “no sensation,” 4 as “threshold of pain sensation,” and 10 as “worst imaginable pain.”20 NPSs were obtained 2 seconds after each stimulus, portraying individual nearest to real-time scorings of pain without generating speech-induced artifacts in the recorded ERPs.

ERP Analysis Recorded EEG data from the Cz position (vertex response; N2-P2 component) were analyzed. Amplitudes of the negative (N2) and positive (P2) vertex potentials were measured from baseline to peak, and the peak-to-

The LORETA-KEY software package (R. D. Pascual-Marqui, The KEY Institute for Brain-Mind Research, Zurich, Switzerland) was used for all source localization analyses. The sLORETA is a method for estimating the localization of cortical generators at specific time windows by providing a solution to the inverse problem, namely the computation of images of electrical neural activity derived from extracranial measurements. sLORETA is used to compute statistical maps from EEG or magnetoencephalography (MEG) data that indicate the locations of the putative underlying source generators.54,65 These maps are derived by performing a location-wise inverse weighting of the results of a minimum norm least squares (MNLS) analysis with their estimated variances. As such, it does not require assumptions about the number, localization, configuration, or extent of neuronal sources. sLORETA performs source localization in 6239 cortical gray matter voxels sized 5 mm3 rather than 7 mm3 offered by the previous LORETA version, and localization inference is based on standardized values of the current density estimates.81 The solution space of sLORETA is restricted to cortical and some hippocampal and amygdala gray matter defined via a reference brain from the Brain Imaging Center at the Montreal Neurological Institute.19,46 The sLORETA implementation incorporates a 3-shell spherical head model registered to a recognized anatomical brain atlas.67 Individual 3-D electrodes are positioned by the Talairach coordinate system according to the spatial association between anatomical brain landmarks and scalp positions.73

Statistical Analysis Based on log-transformed power of the estimated electric current density (ampere per meter squared, A/m2), differences in activity between conditions of 35°C and 52°C were statistically assessed through a nonparametric permutation test with 5000 randomizations accounting for multiple voxel-by-voxel comparisons.51 sLORETA maps were statistically analyzed using the paired Student t test, and the threshold was set to P ⬍ .05 by the aforementioned nonparametric permutation test. Resulting activation clusters were large, owing to the fact that the current densities represent blurred estimates of the cortical sources. At the cluster level, all voxels showing a statistically significant amplitude increase in response to noxious stimuli of 52°C compared with 35°C were detected. Brain regions were defined as active when activation values in response to 52°C compared with 35°C stimuli exceeded a t-threshold corresponding to P ⬍ .05. Clusters of significant activation were defined as regions of interest (ROIs) and included the SI, ACC, OIC, and DLPFC. For further activation magnitude analysis, each ROI was characterized by a single-voxel local spatial maximum. To accept a spatial local maximum, the thresholds

Nir et al

1061

Separate Correlations Between Average NPSs and AUCs of Cortical Pain-ROIs Across All Subjects

Table 1.

LOCATION REGION

SIDE

BA

X

Y

Z

r

c i bi c c

3 3 24 13 8

40 ⫺40 ⫺5 and 5 35 49

⫺21 ⫺21 ⫺3 ⫺18 28

61 61 32 18 37

0.73 0.57 0.56 0.53 0.26

SI ACC* OIC DLPFC

P

VALUE

.002 .027 .032 .043 .350

Abbreviations: AUCs, areas under the curve; NPSs, numerical pain scores; ROIs, regions of interest; SI, primary somatosensory cortex; ACC, anterior cingulate cortex; OIC, operculoinsular cortex; DLPFC, dorsolateral prefrontal cortex; c, contralateral; i, ipsilateral; bi, bilateral; BA, Brodmann’s Area; r, correlation coefficient. *The sum of ipsilateral and contralateral activation of ACC was used. Locations are according to the Talairach coordinate system (x, mediolateral; y, rostrocaudal; z, dorsal-ventral).

of the t values in the particular voxel had to be greater compared with all surrounding voxels (see coordinates of spatial maximum of each ROI in Table 1). All voxel locations are given according to the Talairach coordinate system.67 Current density values of a spatial local maximum of each ROI were measured at 500 Hz in a time window of 250 to 750 ms after stimulus onset, yielding high-resolution temporal curves. Subsequently, areas under these curves (AUCs) were calculated and then correlated with subjective pain ratings. Due to the midline position of the ACC, the sum of AUCs of ipsilateral and contralateral sides was used. Relationships between continuous variables (ERP amplitude and latency, subjective NPS, AUC) were assessed using Pearson correlations and linear regression. Comparisons between discrete test conditions (ie, subjective NPS, temperature level) were performed using multiway repeated measures designs, using multifactor mixed model ANOVA, and subsequent Tukey-Kramer honestly significant difference post-hoc tests. Linear relationships between multiple continuous variables (subjective NPS, temperature value, ERP amplitude) were assessed with multivariate linear regression models. The present study did not aim at addressing differences in the examined parameters as a function of gender. Results are reported as mean ⫾ SD. P ⬍ .05 was regarded as significant. JMP software (SAS Institute, Cary, NC) was used for the statistical analyses.

Results Contact-Heat ERPs and Subjective Numerical Pain Scores In all subjects, application of thermal stimuli to the volar forearm induced clear and reproducible ERPs timelocked to contact-heat stimuli (Fig 2), and well-perceived heat sensations. Mean amplitudes of the negative (N2),

positive (P2), and PTP (N2-P2) components were 6.98 ⫾ 6.35, 17.27 ⫾ 7.46, and 24.25 ⫾ 11.08 ␮V, respectively. Latencies of N2 and P2 peaks were 363.54 ⫾ 60.7 and 514.62 ⫾ 67.78 ms, respectively. Average subjective NPSs ranged from 1.0 to 9.13 on a 10-point scale, with mean value of 5.37 ⫾ 2.26. Average subjective NPSs corresponding to the noxious stimulus intensity of 52°C were investigated for distribution normality using the Shapiro-Wilk Test, which revealed that no statistical difference was found when compared with a normal distribution (P ⫽ .669). Significant correlation was observed between average PTP amplitudes of N2-P2 components and subjective NPSs in response to repetitive contactheat stimuli of 52°C (r ⫽ .74; P ⬍ .001), depicted by the function NPS ⫽ 0.58 ⫹ 0.24PTP Amplitude.

sLORETA Source Localization Analysis Source estimation analysis revealed significant activation of SI, ACC, OIC, and DLPFC (coordinates are shown in Table 1) in response to noxious contact-heat stimulus intensity of 52°C compared with the baseline innocuous stimulation of 35°C (P ⬍ .05 for each; Fig 3). Subsequently, we correlated AUCs of current densities of the mapped areas with the psychophysical measures, to investigate areas having significant correlation with the subjective perception of noxious stimuli. Discrete examination of the relationship between the average subjective NPS and AUC of each cortical area confirmed this relationship was significant at the bilateral SI and ACC, and contralateral OIC, whereas insignificant at the DLPFC (P ⬍ .05 for each; Table 1). Yet, when these regions were considered simultaneously in a multiple regression analysis (regression parameters: R ⫽ .80; P ⫽ .024), only the contralateral SI proved to have had a significant contribution to subjective pain perception (P ⫽ .020; Table 2). A reduced regression model using the contralateral SI alone as a predictor of the resulting subjective pain perception (R ⫽ .75; P ⫽ .008) yielded the linear function NPS(52) ⫽ 2.933⫹0.0057AUC(SIC) (P ⫽ .005; Fig 4). To further validate our results, we aimed at showing that the cortical SI pain response specifically resulted from the stimulation applied to the volar forearm by analyzing a segment of Brodmann’s Area 3 representing the lower limb in SI. As expected, it was found to demonstrate practically null activation. Furthermore, we constructed an additional multivariate regression model, which tested the relationship between pain perception and the AUCs of current densities of contralateral and ipsilateral SI segments, representing the upper limbs. This regression model (R ⫽ .73; P ⫽ .003) confirmed a significant correlation (P ⫽ .003) between subjective NPS values for 52°C stimuli and the activation level of SI only contralateral to the stimulated forearm. The importance of sampling current density values at a high temporal resolution and deriving corresponding AUCs was also demonstrated through an analysis that incorporated only peak current density values of each ROI, and subsequently tested their correlations with pain perception. Separate correlation tests revealed a significant relationship only between average peak amplitudes

1062

Neurophysiology of the Cortical Pain Network

Subject 1

Subject 2 NPS: 4.56

Amplitude [µV]

-15

-10

-5

-5

0

0

5

5

10

10

15

15

20

20

25

25

30

30

35

0.0

0.2

0.4

0.6

0.8

1.0

Subject 3

0.2

0.4

0.6

0.8

1.0

NPS: 9.13

-10 -5

-5

Amplitude [µV]

0.0

-15

NPS: 5.45

-10

0

0

5

5

10

10

15

15

20

20

25

25

30

30

35

35 0.0

0.2

0.4

0.6

0.8

0.0

1.0

Subject 5

0.2

0.4

0.6

0.8

1.0

Subject 6 -15

NPS: 6.45

-15

Amplitude [µV]

35

Subject 4

-15

-10

-10

-5

-5

0

0

5

5

10

10

15

15

20

20

25

25

30

30

NPS: 6.40

35

35 0.0

0.2

0.4

0.6

0.8

0.0

1.0

Subject 7

0.2

0.4

0.6

0.8

1.0

Subject 8

-15

-15

NPS: 3.13

-10

NPS: 8.67

-10

-5

Amplitude [µV]

35°C 52°C

NPS: 5.37

-15

-10

-5

0

0

5

5

10

10

15

15

20

20

25

25

30

30

35

35

0.0

0.2

0.4

0.6

Time [s]

0.8

1.0

0.0

0.2

0.4

0.6

0.8

1.0

Time [s]

Figure 2. Average NPS values (0 –10 scale) to contact-heat stimuli of 52°C, and ERPs recorded at Cz in response to stimulus intensities of 52°C and 35°C (baseline) applied to the volar forearm in 8 randomly chosen subjects. In response to noxious stimuli of 52°C, N2-P2 amplitudes significantly correlated with subjective NPS values (r ⫽ .74; P ⬍ .001; n ⫽ 17). NPS, Numerical Pain Score.

Nir et al

1063

Figure 3. Temporal activation profiles (left) and anatomical modeling (right) of the spatial local maximum of each cortical pain-related region of interest, demonstrating significant activation after application of noxious stimuli of 52oC compared with innocuous baseline stimuli of 35oC applied to the left volar forearm (coordinates are shown in Table 1). These regions include the bilateral SI (A) and ACC (B), and contralateral OIC (C), and DLPFC (D). Left: for each region, average (n ⫽ 17) normalized current density values are depicted

1064

Neurophysiology of the Cortical Pain Network

Multivariate Regression Analysis Incorporating Average AUCs of Pain-ROIs and Subjective Pain Scores

Table 2.

PARAMETER Intercept SI ACC* OIC DLPFC

SIDE

ESTIMATE

SE

t RATIO

— c i bi c c

3.000 0.008 0.001 ⫺0.002 ⫺0.001 ⫺0.001

0.750 0.003 0.001 0.002 0.001 0.003

4.00 2.53 0.89 ⫺1.16 ⫺0.61 ⫺0.22

P

VALUE

.002 .020 .395 .270 .236 .831

Abbreviations: AUCs, areas under the curve; ROIs, regions of interest; SI, primary somatosensory cortex; ACC, anterior cingulate cortex; OIC, operculoinsular cortex; DLPFC, dorsolateral prefrontal cortex; c, contralateral; i, ipsilateral; bi, bilateral. NOTE. Only the contralateral SI significantly correlated with the subjective perception of noxious stimuli. *The sum of ipsilateral and contralateral activation of ACC was used.

of the contralateral SI and subjective NPSs (r ⫽ .60; P ⫽ .017), and a regression analysis that used all ROIs, namely SI, ACC, OIC, and DLPFC, was insignificant.

Discussion Basing sLORETA source localization analysis on pain ERPs, the present study has validated the suitability of sLORETA for pain research, since the detected pain-related regions equated the most commonly activated sites in imaging pain studies (Fig 3). Furthermore, we investigated the relative role of these regions in pain intensity coding, a topic still equivocal. Importantly, although average subjective pain ratings separately correlated with AUCs of current densities of the majority of the detected regions, a regression analysis found it was only the contralateral SI that manifested significant correlation with subjective pain perception (Fig 4). The novelty of the present study is exhibited through the aforementioned results, which are of importance in light of the documented inconsistency in the literature of SI activation. The following discussion elaborates upon this divergence, and also addresses our main findings in light of previous intensity coding investigations.

Validity of sLORETA in Pain Research Numerous studies have supported the usefulness of the earlier LORETA version in localizing generators of scalp-recorded potentials,1,2,53,70,80,82 including recent research on epilepsy,6 schizophrenia,30 and Alzheimer’s disease,5 in which spontaneous EEG was continuously recorded with no simultaneous stimulation. The localization accuracy of the earlier version has been demon-

strated in cross-modality comparisons including combined EEG/functional magnetic resonance imaging (fMRI) studies,49,79 showing that LORETA and fMRI localizations overlap within 14 to 16 mm. Furthermore, crossvalidation studies of epileptic foci localization43,47 have demonstrated strong correspondence with PET data,84 and studies using 32 electrodes have reached significant conclusions regarding cortical processing in schizophrenia25 and delirium states.61 The current ERP-based study targeted to investigate the suitability of sLORETA for pain research by applying series of brief noxious contact-heat stimuli of 52°C, and comparing corresponding cortical responses to those exerted by baseline stimuli of 35°C. The sensitivity of sLORETA was demonstrated by its capability to identify the most commonly activated cortical pain-related regions reported in imaging studies, namely the ACC, SI, OIC, and DLPFC.4,11,12,14,45,56,68 The specificity of sLORETA was stressed by revealing that the noxious stimuli applied to the volar forearm induced null activation in a segment of Brodmann’s Area (BA) 3 representing the lower leg in SI, while exerting significant activation in a segment of the same BA representing the contralateral forearm. Further establishing the specificity of sLORETA, a regression analysis testing the correlation between subjective pain ratings and AUCs of current densities of ipsilateral and contralateral SI revealed that the SI activation contralateral to the stimulated side prevailed. Thus, the validity of sLORETA in research of pain processing was demonstrated.

Investigation of Pain Intensity Coding Using sLORETA Once verifying the sensitivity and specificity of sLORETA in identifying pain-related regions, we used sLORETA-based analysis to investigate the still equivocal topic of pain intensity coding. The present study used a temporal resolution of milliseconds to form reliable high-resolution time curves of current densities at the maximally activated voxels, the areas under which were later correlated with pain perception. This enabled assessing the contribution of each of the activated cortical structures to pain intensity coding. Our results indicate that significant correlations between average subjective NPS values and the noxious stimuli-induced activation magnitude characterized the bilateral SI and ACC, and contralateral OIC. Yet, a regression analysis distinguished the contralateral SI as the sole region having a significant impact on the subjective perception of noxious stimuli. Our results are in agreement with primate-derived evidence for correlation of intensity and perception of

4™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™ during the time interval of 250 to 750 ms after stimulus onset. Areas under these curves (AUCs) were tested for correlation with subjective pain perception. Gray vertical bars indicate time-interval of 10 ms of peak activation magnitudes. Right: for each ROI, average (n ⫽ 17) statistical paired t test maps representing the level of significance of the difference in response to noxious (52oC) versus innocuous (35oC) stimuli were generated for the 10-ms time interval of peak activation magnitude. t values from a nonparametric testing are color coded such that yellow voxels denote P ⬍ .05, and red voxels denote P ⬍ .1. SI, primary somatosensory cortex; OIC, operculoinsular cortex; ACC, anterior cingulate cortex; DLPFC, dorsolateral prefrontal cortex; c, contralateral; i, ipsilateral.

Numerical Pain Scale

Nir et al

1065 10,21,22,23,62

10 9 8 7 6 5 4 3 2 1 0 0

200

400

600

800

AUC of current density in contralateral SI [A/m2] Figure 4. Average activation of contralateral SI and subjective NPSs in response to noxious stimuli of 52oC. A reduced regression model, which used the AUC of the contralateral SI alone as a predictor of the subjective perception of noxious stimuli, yielded a linear relationship (R ⫽ .75; P ⫽ .008).

painful heat with the firing rate of SI wide-dynamicrange (WDR) neurons.17,39,41 Additionally, SI ablation in primates leads to a severe deficit in the detection and discrimination of noxious stimuli.40 A case report on a patient with ischemic lesion including SI and SII57 supports these observations by the patient’s inability to describe neither the intensity nor location of the applied stimulus, attributed to the SI lesion. As the aforementioned reports, our results correspond with the concept that the SI is an instrumental region in the sensory-discriminative aspect of pain processing, which encodes sensory properties of noxious stimuli, such as intensity18,60 and bodily location3,69 Yet, the present study did not aim at investigating properties of SI activation in response to innocuous thermal stimuli, as well as stimuli of other modalities.

Detection of SI Activation The detection yield of SI activation and its role in pain processing is presently under debate.16 In their review, Peyron et al56 reported that pain-induced SI activity was observed in merely half of PET and fMRI studies, and likewise, electrophysiological laser pain studies have also encountered difficulties in demonstrating SI activation26 using scalp-derived EEG7,9,76,77 or MEG.24,34,50,83 Conversely, reports using similar recording and analysis techniques,35,36,58,59,71,74 or differing by either using direct recording of subdural/cortical electrodes,52,78 or dipolar modeling techniques,63 did demonstrate SI activation in response to a laser-evoked afferent barrage of A-␦ as well as of C-fiber nociceptors. Peyron et al56 suggested several causes for these inconsistent SI activations, for example, the stimulated skin area and body site. This is in accordance with Treede’s notion on requirement of substantial temporal and spatial summation for SI detection.75 In this regard, the size of receptive fields (RFs) of nociceptive neurons in painrelated regions can also account for the reported divergent results; most SI nociceptive neurons have small RFs,17,38,39,42 whereas SII, ACC, and insular nociceptive

neurons have larger RFs. It is likely, therefore, that a stimulation of a given skin area would induce a greater activation of SII, ACC, and insular neurons compared with SI, thereby generating a less easily detectable activity in the latter. This argument can account for the robust SI activation in the present study, since our contact skin area was larger compared with laser studies. Likewise, the stimulus duration in our study can account for the more robust activation of SI as compared with laser-induced electrophysiological recordings, as the latter stimulus duration is substantially shorter. In addition, assuming that a portion of SI nociceptive neurons can be suppressed by continuous prolonged noxious stimulation, demonstrating inhibitory effects within SI,17,39,42 series of relatively short stimuli (of less than 1 second) as used in this study could exert more excitatory activity, optimizing the use of SI in pain perception.64

Pain Intensity Coding: A Still-Debated Topic Although serious progress has been made in recent years in investigation of pain intensity coding, considerable discrepancies regarding cortical structures and their functionality in pain processing still exist. This reason and notable limitations of relevant studies motivated us to revisit this still debated topic by incorporating the activity of the identified pain regions in a regression analysis for delineating their relative contribution to subjective pain perception. One limitation shared by several intensity coding studies is their concentration on specific pain-related regions; using MEG and focused on SI and SII, Timmermann et al71 suggested SI represented the perceived stimulus intensity, while the activation pattern of SII pointed against a significant contribution to the sensory-discriminative aspect of pain perception. Chen et al16 characterized the temporal activation of the same regions using fMRI, and concluded both SI and SII encoded a temporal signature specific to the perceptual characteristics of both noxious and innocuous stimuli. However, neither study tested direct correlations between activation magnitudes and pain ratings. In an MEG study, Chen et al17 explored the somatosensory cortices, but was unable to evaluate the role of SI in pain intensity coding since SI activation was detectable in only half of the subjects. Iannetti et al.31, using EEG and concentrating on OIC and SI, have concluded it was the OIC which predominantly encoded pain intensity. Several studies used imaging techniques to investigate intensity coding in multiple regions, though their main aim was other than to assess the functionality of these regions in terms of their relative impact on pain perception; Moulton et al48 concluded that nociceptive responses in SI discriminate between painful temperatures at a level unmatched in other cortical areas, though a correlation was not assessed between its activation and subjective pain ratings. Porro et al60 reported several regions, including SI and ACC but not SII, demonstrated signal intensity changes that significantly correlated

1066

Neurophysiology of the Cortical Pain Network

with individual psychophysical pain intensity curves. In contrast, other studies were not as definite as to the instrumental role of SI in intensity coding. Derbyshire et al20 found a wide range of regions that linearly related to perceived stimulus intensity, though admittedly inconclusive regarding SI. Bornhövd et al8 used correlation tests to demonstrate that SI responses did not fully correspond to perceived pain, showing decreased activation in the highest stimulus intensity. Peyron et al55 did not include the SI in the intensity coding matrix, but rather the bilateral insula/SII cortices, though not testing correlations between pain ratings and activation levels at those regions. Regression analyses between pain-related regions were incorporated in PET studies by Tölle,72 Coghill,18 and their colleagues. The former investigation concluded that coding of pain intensity is attributed to the periventricular gray and the posterior cingulate cortex, whereas the latter research attributed this function to multiple regions, including SI, SII, ACC, and insula. In the current study, it is noted that activity in the DLPFC did not correlate with perceived pain magnitude. This is probably due to the fact that this region does not receive direct data from the ascending pathways but rather serves in modulation of pain perception.12,18 There are conflicting reports on the timing of activation of the various components of the central pain-network regions.26 Although this study did not aim at exploring the temporal domain of the pain-network activation, our data is derived form a high temporal resolution EEG-based sLORETA analysis, through which we rigorously constructed AUCs at a resolution of 500 Hz,

reflecting temporal activation profiles of the predominant pain-related regions. Naturally, this methodology sheds light on existing imaging-based temporal characterization of pain intensity coding,8,15,48 because their measurement of haemodynamic brain responses limits them to a minimal temporal resolution of a few seconds. Our finding show similar temporal activation profiles for the major pain-network components, including SI, OIC, and ACC. As to the correspondence between temporal activities of SI and the OIC, this is in line with several MEG-based reports showing similarity between time courses of activity in SI and opercular sources.26,36,57 This concomitance, which was reported in responses to intracutaneous A-␦ stimulation,33 had already been observed in the early EEG study of Tarkka and Treede69 that suggested SI and SII activities to thermal stimuli were organized in parallel, rather than as sequential responses. As to the ACC, latency times of its activation, although slightly variable across studies, lie within a time window included in the temporal response profiles attributed to SI and opercular sources.26 In conclusion, the current ERP study demonstrates the feasibility of sLORETA in analysis of the pain-network activity in the brain, in response to contact-heat stimulation. Our results show the different stimulus-response relationship of the contralateral SI compared with those of ACC, OIC, and DLPFC, with the first clearly coding pain intensity, unlike the others. Activity patterns demonstrated by sLORETA are distinguishable from the classical fMRI, PET, and laser studies, calling for further exploration of data using this approach.

References

span in mild cognitive impairment and Alzheimer’s disease. Eur J Neurosci 25:3742-3757, 2007

1. Anderer P, Pascual-Marqui RD, Semlitsch HV, Saletu B: Electrical sources of P300 event-related brain potentials revealed by low resolution electromagnetic tomography. I: effects of normal aging. Neuropsychobiology 37:20-27, 1998

6. Bela C, Monika B, Marton T, Istvan K: Valproate selectively reduces EEG activity in anterior parts of the cortex in patients with idiopathic generalized epilepsy: A low resolution electromagnetic tomography (LORETA) study. Epilepsy Res 75:186-191, 2007

2. Anderer P, Pascual-Marqui RD, Semlitsch HV, Saletu B: Differential effects of normal aging on sources of standard N1, target N1 and target P300 auditory event-related brain potentials revealed by low resolution electromagnetic tomography (LORETA). Electroencephalogr Clin Neurophysiol 108:160-174, 1998 3. Andersson JL, Lilja A, Hartvig P, Langstrom B, Gordh T, Handwerker H, Torebjork E: Somatotopic organization along the central sulcus, for pain localization in humans, as revealed by positron emission tomography. Exp Brain Res 117:192-199, 1997 4. Apkarian AV, Darbar A, Krauss BR, Gelnar PA, Szeverenyi NM: Differentiating cortical areas related to pain perception from stimulus identification: Temporal analysis of fMRI activity. J Neurophysiol 81:2956-2963, 1999 5. Babiloni C, Cassetta E, Binetti G, Tombini M, Del Percio C, Ferreri F, Ferri R, Frisoni G, Lanuzza B, Nobili F, Parisi L, Rodriguez G, Frigerio L, Gurzi M, Prestia A, Vernieri F, Eusebi F, Rossini PM: Resting EEG sources correlate with attentional

7. Bentley DE, Youell PD, Crossman AR, Jones AK: Source localisation of 62-electrode human laser pain evoked potential data using a realistic head model. Int J Psychophysiol 41:187-193, 2001 8. Bornhövd K, Quante M, Glauche V, Bromm B, Weiller C, Buchel C: Painful stimuli evoke different stimulus-response functions in the amygdala, prefrontal, insula and somatosensory cortex: A single-trial fMRI study. Brain 125(Pt 6): 1326-1336, 2002 9. Bromm B, Chen AC: Brain electrical source analysis of laser evoked potentials in response to painful trigeminal nerve stimulation. Electroencephalogr Clin Neurophysiol 95: 14-26, 1995 10. Bushnell MC, Duncan GH: Sensory and affective aspects of pain perception: Is medial thalamus restricted to emotional issues? Exp Brain Res 78:415-418, 1989 11. Bushnell MC, Duncan GH, Hofbauer RK, Ha B, Chen JI, Carrier B: Pain perception: Is there a role for primary somatosensory cortex? Proc Natl Acad Sci U S A 96:7705-7709, 1999

Nir et al 12. Bushnell MC, Apkarian AV: Representation of pain in the brain, in McMahon SB, Koltzenburg M (eds): Wall and Melzack’s Textbook of Pain, 5th ed. Elsevier Churchill Livingstone, Philadelphia, PA, 2006, pp 107-124 13. Carmon A, Mor J, Goldberg J: Evoked cerebral responses to noxious thermal stimuli in humans. Exp Brain Res 25:103107, 1976 14. Casey KL, Minoshima S, Morrow TJ, Koeppe RA: Comparisons of human cerebral activation pattern during cutaneous warmth, heat pain, and deep cold pain. J Neurophysiol 76:571-581, 1996 15. Chen JI, Ha B, Bushnell MC, Pike B, Duncan GH: Differentiating noxious- and innocuous-related activation of human somatosensory cortices using temporal analysis of fMRI. J Neurophysiol 88:464-474, 2002 16. Chen WT, Yuan RY, Shih YH, Yeh TC, Hung DL, Wu ZA, Ho LT, Lin YY: Neuromagnetic SII responses do not fully reflect pain scale. Neuroimage 31:670-676, 2006 17. Chudler EH, Anton F, Dubner R, Kenshalo DR Jr: Responses of nociceptive SI neurons in monkeys and pain sensation in humans elicited by noxious thermal stimulation: Effect of interstimulus interval. J Neurophysiol 63:559-569, 1990 18. Coghill RC, Sang CN, Maisog JM, Iadarola MJ: Pain intensity processing within the human brain: A bilateral, distributed mechanism. J Neurophysiol 82:1934-1943, 1999 19. Collins DL, Neelin P, Peters TM, Evans AC: Automatic 3D intersubject registration of MR volumetric data in standardized Talairach space. J Comput Assist Tomogr 18:192-205, 1994 20. Derbyshire SW, Jones AK, Gyulai F, Clark S, Townsend D, Firestone LL: Pain processing during three levels of noxious stimulation produces differential patterns of central activity. Pain 73:431-45, 1997 21. Dong WK, Ryu H, Wagman IH: Nociceptive responses of neurons in medial thalamus and their relationship to spinothalamic pathways. J Neurophysiol 41:1592-1613, 1978 22. Dong WK, Salonen LD, Kawakami Y, Shiwaku T, Kaukoranta EM, Martin RF: Nociceptive responses of trigeminal neurons in SII-7b cortex of awake monkeys. Brain Res 484:314-324, 1989

1067 29. Greffrath W, Baumgartner U, Treede RD: Peripheral and central components of habituation of heat pain perception and evoked potentials in humans. Pain 132:301-311, 2007 30. Horacek J, Brunovsky M, Novak T, Skrdlantova L, Klirova M, Bubenikova-Valesova V, Krajca V, Tislerova B, Kopecek M, Spaniel F, Mohr P, Hoschl C: Effect of low-frequency rTMS on electromagnetic tomography (LORETA) and regional brain metabolism (PET) in schizophrenia patients with auditory hallucinations. Neuropsychobiology 55:132-142, 2007 31. Iannetti GD, Zambreanu L, Cruccu G, Tracey I: Operculoinsular cortex encodes pain intensity at the earliest stages of cortical processing as indicated by amplitude of laserevoked potentials in humans. Neuroscience 131:199-208, 2005 32. Iannetti GD, Zambreanu L, Tracey I: Similar nociceptive afferents mediate psychophysical and electrophysiological responses to heat stimulation of glabrous and hairy skin in humans. J Physiol 577(Pt 1):235-248, 2006 33. Inui K, Tran TD, Oiu Y, Wang X, Hoshiyama M, Kakigi R: A comparative magnetoencephalographic study of cortical activations evoked by noxious and innocuous somatosensory stimulations. Neuroscience 120:235-248, 2003 34. Kakigi R, Koyama S, Hoshiyama M, Kitamura Y, Shimojo M, Watanabe S: Pain-related magnetic fields following painful CO2 laser stimulation in man. Neurosci Lett 192:45-48, 1995 35. Kakigi R, Watanabe S, Yamasaki H, Maeda K: Pain-related brain activities: magnetoencephalographic studies. Electroencephalogr Clin Neurophysiol Suppl 49:245-249, 1999 36. Kanda M, Nagamine T, Ikeda A, Ohara S, Kunieda T, Fujiwara N, Yazawa S, Sawamoto N, Matsumoto R, Taki W, Shibasaki H: Primary somatosensory cortex is actively involved in pain processing in human. Brain Res 853:282-289, 2000 37. Kanda M, Matsuhashi M, Sawamoto N, Oga T, Mima T, Nagamine T, Shibasaki H: Cortical potentials related to assessment of pain intensity with visual analogue scale (VAS). Clin Neurophysiol 113:1013-1024, 2002 38. Kenshalo DR Jr, Isensee O: Responses of primate SI cortical neurons to noxious stimuli. J Neurophysiol 50:14791496, 1983

23. Dong WK, Chudler EH, Sugiyama K, Roberts VJ, Hayashi T: Somatosensory, multisensory, and task-related neurons in cortical area 7b (PF) of unanesthetized monkeys. J Neurophysiol 72:542-564, 1994

39. Kenshalo DR Jr, Chudler EH, Anton F, Dubner R: SI nociceptive neurons participate in the encoding process by which monkeys perceive the intensity of noxious thermal stimulation. Brain Res 454:378-382, 1988

24. Forss N, Raij TT, Seppa M, Hari R: Common cortical network for first and second pain. Neuroimage 24:132-142, 2005

40. Kenshalo DR, Willis WD: The role of the cerebral cortex in pain sensation. In Jones EG, Peters A (eds): Cerebral Cortex, Normal and Altered States of Function (Vol. 9). Plenum, New York, NY, 1991, pp 153-212

25. Gallinat J, Mulert C, Bajbouj M, Herrmann WM, Schunter J, Senkowski D, Moukhtieva R, Kronfeldt D, Winterer G: Frontal and temporal dysfunction of auditory processing in schizophrenia. Neuroimage 17:110-127, 2002 26. Garcia-Larrea L, Frot M, Valeriani M: Brain generators of laser-evoked potentials: From dipoles to functional significance. Neurophysiol Clin 33:279-292, 2003 27. Granovsky Y, Matre D, Sokolik A, Lorenz J, Casey KL: Thermoreceptive innervation of human glabrous and hairy skin: a contact heat evoked potential analysis. Pain 115:238247, 2005 28. Granovsky Y, Granot M, Nir RR, Yarnitsky D: Objective correlate of subjective pain perception by contact heatevoked potentials. J Pain 9:53-63, 2008

41. Kenshalo DR, Iwata K, Sholas M, Thomas DA: Response properties and organization of nociceptive neurons in area 1 of monkey primary somatosensory cortex. J Neurophysiol 84:719-729, 2000 42. Lamour Y, Willer JC, Guilbaud G: Rat somatosensory (SmI) cortex, I: characteristics of neural responses to noxious stimulation and comparison with responses to non-noxious stimulation. Exp Brain Res 49:35-45, 1983 43. Lantz G, Grave de Peralta R, Spinelli L, Seeck M, Michel CM: Epileptic source localization with high density EEG: how many electrodes are needed? Clin Neurophysiol 114:63-69, 2003 44. Lorenz J, Garcia-Larrea L: Contribution of attentional

1068

Neurophysiology of the Cortical Pain Network

and cognitive factors to laser evoked brain potentials. Neurophysiol Clin 33:293-301, 2003

ral and intensity coding of pain in human cortex. J Neurophysiol 80:3312-3320, 1998

45. May A, Kaube H, Buchel C, Eichten C, Rijntjes M, Juptner M, Weiller C, Diener HC: Experimental cranial pain elicited by capsaicin: A PET study. Pain 74:61-66, 1998

61. Reischies FM, Neuhaus AH, Hansen ML, Mientus S, Mulert C, Gallinat J: Electrophysiological and neuropsychological analysis of a delirious state: The role of the anterior cingulated gyrus. Psychiatry Res 138:171-181, 2005

46. Mazziotta J, Toga A, Evans A, Fox P, Lancaster J, Zilles K, Woods R, Paus T, Simpson G, Pike B, Holmes C, Collins L, Thompson P, MacDonald D, Iacoboni M, Schormann T, Amunts K, Palomero-Gallagher N, Geyer S, Parsons L, Narr K, Kabani N, Le Goualher G, Boomsma D, Cannon T, Kawashima R, Mazoyer B: A probabilistic atlas and reference system for the human brain: International Consortium for Brain Mapping (ICBM). Philos Trans R Soc Lond B Biol Sci 356:1293-1322, 2001 47. Michel CM, Lantz G, Spinelli L, De Peralta RG, Landis T, Seeck M: 128-channel EEG source imaging in epilepsy: Clinical yield and localization precision. J Clin Neurophysiol 21: 71-83, 2004 48. Moulton EA, Keaser ML, Gullapalli RP, Greenspan JD: Regional intensive and temporal patterns of functional MRI activation distinguishing noxious and innocuous contact heat. J Neurophysiol 93:2183-2193, 2005

62. Robinson CJ, Burton H: Somatic submodality distribution within the second somatosensory (SII), 7b, retroinsular, postauditory, and granular insular cortical areas of M. fascicularis. J Comp Neurol 192:93-108, 1980 63. Schlereth T, Baumgartner U, Magerl W, Stoeter P, Treede RD: Left-hemisphere dominance in early nociceptive processing in the human parasylvian cortex. Neuroimage 20:441-454, 2003 64. Schnitzler A, Ploner M: Neurophysiology and functional neuroanatomy of pain perception. J Clin Neurophysiol 17: 592-603, 2000 65. Sekihara K, Sahani M, Nagarajan SS: A simple nonparametric statistical thresholding for MEG spatial-filter source reconstruction images. Neuroimage 27:368-376, 2005

49. Mulert C, Jager L, Schmitt R, Bussfeld P, Pogarell O, Moller HJ, Juckel G, Hegerl U: Integration of fMRI and simultaneous EEG: Towards a comprehensive understanding of localization and time-course of brain activity in target detection. Neuroimage 22:83-94, 2004

66. Soufflet L, Boeijinga PH: Linear inverse solutions: Simulations from a realistic head model in MEG. Brain Topogr 18:87-99, 2005

50. Nakamura Y, Paur R, Zimmermann R, Bromm B: Attentional modulation of human pain processing in the secondary somatosensory cortex: A magnetoencephalographic study. Neurosci Lett 328:29-32, 2002

68. Talbot JD, Marrett S, Evans AC, Meyer E, Bushnell MC, Duncan GH: Multiple representations of pain in human cerebral cortex. Science 251:1355-1358, 1991

51. Nichols TE, Holmes AP: Nonparametric permutation tests for functional neuroimaging: A primer with examples. Hum Brain Mapp 15:1-25, 2002 52. Ohara S, Crone NE, Weiss N, Treede RD, Lenz FA: Cutaneous painful laser stimuli evoke responses recorded directly from primary somatosensory cortex in awake humans. J Neurophysiol 91:2734-2746, 2004 53. Pascual-Marqui RD, Michel CM, Lehmann D: Low resolution electromagnetic tomography: A new method for localizing electrical activity in the brain. Int J Psychophysiol 18: 49-65, 1994 54. Pascual-Marqui RD: Standardized low-resolution brain electromagnetic tomography (sLORETA): Technical details. Methods Find Exp Clin Pharmacol 24(Suppl D):5-12, 2002 55. Peyron R, Garcia-Larrea L, Gregoire MC, Costes N, Convers P, Lavenne F, Mauguiere F, Michel D, Laurent B: Haemodynamic brain responses to acute pain in humans: Sensory and attentional networks. Brain 122(Pt 9):1765-1780, 1999 56. Peyron R, Laurent B, Garcia-Larrea L: Functional imaging of brain responses to pain: A review and meta-analysis (2000). Neurophysiol Clin 30:263-288, 2000 57. Ploner M, Freund HJ, Schnitzler A: Pain affect without pain sensation in a patient with a postcentral lesion. Pain 81:211-214, 1999 58. Ploner M, Schmitz F, Freund HJ, Schnitzler A: Parallel activation of primary and secondary somatosensory cortices in human pain processing. J Neurophysiol 81(6):3100-3104,1999 59. Ploner M, Schmitz F, Freund HJ, Schnitzler A: Differential organization of touch and pain in human primary somatosensory cortex. J Neurophysiol 83:1770-1776, 2000 60. Porro CA, Cettolo V, Francescato MP, Baraldi P: Tempo-

67. Talairach J, Tournoux P: Co-Planar Stereotaxic Atlas of the Human Brain. Stuttgart, Germany, Thieme, 1988

69. Tarkka IM, Treede RD: Equivalent electrical source analysis of pain-related somatosensory evoked potentials elicited by a CO2 laser. J Clin Neurophysiol 10:513-519, 1993 70. Thatcher RW, North D, Biver C: Evaluation and validity of a LORETA normative EEG database. Clin EEG Neurosci 36:116-122, 2005 71. Timmermann L, Ploner M, Haucke K, Schmitz F, Baltissen R, Schnitzler A: Differential coding of pain intensity in the human primary and secondary somatosensory cortex. J Neurophysiol 86:1499-1503, 2001 72. Tölle TR, Kaufmann T, Siessmeier T, Lautenbacher S, Berthele A, Munz F, Zieglgansberger W, Willoch F, Schwaiger M, Conrad B, Bartenstein P: Region-specific encoding of sensory and affective components of pain in the human brain: A positron emission tomography correlation analysis. Ann Neurol 45:40-47, 1999 73. Towle VL, Bolanos J, Suarez D, Tan K, Grzeszczuk R, Levin DN, Cakmur R, Frank SA, Spire JP: The spatial location of EEG electrodes: Locating the best-fitting sphere relative to cortical anatomy. Electroencephalogr Clin Neurophysiol 86:1-6, 1993 74. Tran TD, Inui K, Hoshiyama M, Lam K, Qiu Y, Kakigi R: Cerebral activation by the signals ascending through unmyelinated C-fibers in humans: A magnetoencephalographic study. Neuroscience 113:375-386, 2002 75. Treede RD, Vogel H, Rios M, Krauss G, Lesser RP, Lenz FA: Pain-related evoked potentials from parasylvian cortex in humans. Electroencephalogr Clin Neurophysiol Suppl 49: 250-254, 1999 76. Valeriani M, Rambaud L, Mauguiere F: Scalp topography and dipolar source modelling of potentials evoked by CO2 laser stimulation of the hand. Electroencephalogr Clin Neurophysiol 100:343-353, 1996

Nir et al 77. Valeriani M, Le Pera D, Niddam D, Arendt-Nielsen L, Chen AC: Dipolar source modeling of somatosensory evoked potentials to painful and nonpainful median nerve stimulation. Muscle Nerve 23:1194-1203, 2000 78. Valeriani M, Barba C, Le Pera D, Restuccia D, Colicchio G, Tonali P, Gagliardo O, Treede RD: Different neuronal contribution to N20 somatosensory evoked potential and to CO2 laser evoked potentials: An intracerebral recording study. Clin Neurophysiol 115:211-216, 2004 79. Vitacco D, Brandeis D, Pascual-Marqui R, Martin E: Correspondence of event-related potential tomography and functional magnetic resonance imaging during language processing. Hum Brain Mapp 17:4-12, 2002 80. Volpe U, Mucci A, Bucci P, Merlotti E, Galderisi S, Maj M: The cortical generators of P3a and P3b: A LORETA study. Brain Res Bull 73:220-230, 2007

1069 81. Wagner M, Fuchs M, Kastner J: Evaluation of sLORETA in the presence of noise and multiple sources. Brain Topogr 16:277-280, 2004 82. Weber DL, Clark CR, McFarlane AC, Moores KA, Morris P, Egan GF: Abnormal frontal and parietal activity during working memory updating in post-traumatic stress disorder. Psychiatry Res 140:27-44, 2005 83. Yamasaki H, Kakigi R, Watanabe S, Naka D: Effects of distraction on pain perception: Magneto- and electro-encephalographic studies. Brain Res Cogn Brain Res 8:73-76, 1999 84. Zumsteg D, Friedman A, Wennberg RA, Wieser HG: Source localization of mesial temporal interictal epileptiform discharges: Correlation with intracranial foramen ovale electrode recordings. Clin Neurophysiol 1:2810-2818, 2005