- Email: [email protected]
Empathy for pain in fibromyalgia patients: An EEG study
International Journal of Psychophysiology 146 (2019) 43–53
Contents lists available at ScienceDirect
International Journal of Psychophysiology journal homepage: www.elsevier.com/locate/ijpsycho
Empathy for pain in fibromyalgia patients: An EEG study a,⁎
a
a
a
a
Marina de Tommaso , Katia Ricci , Giuliana Conca , Eleonora Vecchio , Marianna Delussi , Sara Invittob a b
T
Applied Neurophysiology and Pain Unit, TIRES Center, Bari Aldo Moro University, Italy Human Anatomy and Neuroscience Lab, DiSTeBA, University of Salento, Lecce, Italy
ARTICLE INFO
ABSTRACT
Keywords: Empathy Pain Fibromyalgia Laser evoked potentials Event related spectral perturbation
Background: Recent research in cognitive neurosciences highlights how the neural circuitries are activated during pain responses in empathic context. Aims: The present study was designed to test if healthy subjects and Fibromyalgia (FM) patients, both evaluated by Laser Evoked Potentials (LEPs) and Event-Related Spectral Perturbation (ERSP), might reveal the empathic response to the partner's nociceptive stimulation. Methods: The emphatic nociceptive paradigm was recorded through 64 channels EEG and laser stimulation of the right hand in a shared visual open setting (Open Condition) or in a blind setting (Blind condition) where the subjects didn't receive visual information about partner nociceptive condition. Twenty one healthy subjects and 19 FM patients were evaluated in pairs. All subjects were tested by the Empathy for Pain Scale (EPS). Results: The averaged LEPs were similar between patients and controls in the different conditions. In attendance of the partner's stimulation, FM patients desynchronized the same fronto-central regions as before own stimulation, while healthy subjects shared the other's pain by activating scalp areas compatible with visual attention. These EEG features were more represented in subjects with higher EPS scores. Conclusions: While empathic features of healthy subjects seemed influenced by the specific visual attentional task, patients expressed an EEG pattern compatible with somatosensory circuits activation in the expectation of own and other's pain. The visual empathic involvement in other's noxious stimulation could evoke a different EEG response depending upon the experience of chronic pain.
1. Introduction Empathy is the ability to consciously share the affective state of another individual, who is recognized as the source of that state, and generates similar perception in a ‘mirroring’ condition (de Vignemont and Singer, 2006). Recent research in the field of cognitive neuroscience has shed light on the neural circuits that are activated during responses to pain in an empathic context. Existing neuroscientific evidence suggests that empathizing with the pain of another person implies a topographic activity in regions similar to those associated with the direct experience of pain (Decety, 2009; Lamm et al., 2011). Neuroimaging studies on empathy of pain have shown that the cortical areas involved in the processing of one's own pain can also be activated during the empathic perception of another person's pain experience (Decety, 2009; Lamm et al., 2011). The empathic response is based on the same somatosensory circuits that underlie pain processing and on cortical areas involved in affective and emotional responses. It is
influenced by cortical areas included in the somatosensory system that regulates the pain elaboration, as well as by the emotional reaction. The anticipation of pain in others also evokes pain-related cortical responses, which seem to correspond with the severity of the pain expected (Jackson et al., 2006). Despite the existence of a large amount of research on empathy for pain, how the modality to share the others' pain could reflect the one's own pain elaboration is still an unresolved issue. In fact, most neuroimaging studies are based on the empathic reaction to another person's suffering, without previous personal experiences. The most commonly used paradigms are based on viewing pictures of people with limbs in painful situations; another type of paradigm is based on an interaction with a person subjected to a painful stimulation that was preceded by a visual or acoustic cue (Lamm et al., 2011). In the latter paradigm, sensory-motor processes, including mimicry, might strongly contribute to neural responses. The intrinsic modality of pain processing and affective reactions may influence the empathic correspondence with another person's suffering, with an
⁎ Corresponding author at: Applied Neurophysiology and Pain Unit, TIRES Center, Bari Aldo Moro University, Policlinico General Hospital, Giovanni XXIII Building, Via Amendola 207 A, 70124 Bari, Italy. E-mail address: [email protected] (M. de Tommaso).
https://doi.org/10.1016/j.ijpsycho.2019.09.007 Received 23 January 2019; Received in revised form 16 August 2019; Accepted 24 September 2019 Available online 21 October 2019 0167-8760/ © 2019 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
activation of the same nociceptive cortical circuits underlying personal experience. Thus, a vision of the other person's pain may evoke an empathic response, which can be reinforced by self-personal pain experience. Within the interaction paradigm (Bernhardt and Singer, 2012; Lang et al., 2011; Perry et al., 2010), we can use the concept of empathy, no longer understood as ‘looking at the other’ (Balconi and Vanutelli, 2017), but as a social system that can be generated from a context (Fitzgibbon et al., 2010). In this ‘new’ paradigm we hypothesized to construct a model of shared nociception, where the subjects became dyads (Balconi et al., 2017) and the setting allows a shared nociception. The empathic response seemed to be altered in patients with diseases that involved the nociceptive system and in disorders of affective and social relationships (Betti and Aglioti, 2016). Fibromyalgia is a chronic disabling disorder, characterized by wide spread pain, sleep disturbances, fatigue and other symptoms, which comprise the complex diagnosis (Wolfe et al., 2010). Vandenbroucke explored the spatial discrimination of tactile stimuli in FM patients and controls (Vandenbroucke et al., 2014), during the vision of videos depicting pain- and non-pain-related scene. Authors reported that the vision of others' pain facilitated spatial somatosensory detection in patients and controls. In an EEG study, FM patients showed increased activations associated with pain and non-pain pictures, suggesting that even innocuous stimuli with somatic connotations may involve the emotional state (Fallon et al., 2015). Üçeyler et al. found higher scores for personal distress and empathic concern in FM patients compared to controls, but they did not assess how these features could have impact on central pain elaboration (Üçeyler et al., 2015). The empathic response could thus make light on some aspects of pain processing in this complex disease. Laser-evoked potentials (LEPs) are a useful tool to explore the cortical activation under A-delta fiber input (Treede et al., 2003), including the psychophysiological changes due to empathic processes (Martini et al., 2013; Valeriani et al., 2008). The present study was designed to test 1) if in healthy subjects and FM patients the LEPs and the EEG rhythms might change under the vision of own or other's stimulated hand, during EEG recording session, adapted for a dyad 2) if these changes could be different between patients and controls, 3) if neurophysiological differences could depend upon the individual empathic behavior, expressed by the Empathy for Pain scale (Giummarra et al., 2015).
experimental tasks were performed in the morning. The subject's pairs were chosen in a pseudo-random way between participants who were not friends and do not know other's medical history. The study was approved by the Ethical Committee of Policlinico General Hospital of Bari (Italy). All subjects signed an informed consent. The study was conducted in accordance with the World Medical Association Declaration of Helsinki, which was published on the website of the Journal of American Medical Association. 2.2. Procedure While each subject was sitting in a comfortable chair, recordings were made using 31 scalp electrodes, (FP1,FP2,F3,Fz,F4,T3,T4,T5, T6,Cz,C3,C4,P3,P4,Pz,O1,Oz,O2,CP2,CP1,AF3,CPZ,AF4,CP6,CP5,FC5, FC1,FC2,FC6,TP6,TP7, according to the 10–20 International system) connected to a MICROMED amplifier. Four additional electrodes were positioned above and below both eyes to record the electrooculogram. The reference electrode was positioned at the nasion, and the ground electrode was positioned in the middle between the nasion and the FPZ. The EEG montage was prepared in the same room while subjects were separated by a panel screen, which avoided one person from being able to see the other person (Fig. 1). All the laser stimuli were thus triggered on the EEGs, and the tracks from both subjects were simultaneously visualized. Cutaneous heat stimuli were delivered with a CO2 laser (wavelength = 10.6 lm, 2-mm beam diameter; ELEN, Florence, Italy) on the dorsum of the right hand. The duration of stimulation was 30 ms. The laser beam light was directed to the hand during the entire stimulation session, independent of the stimuli delivering. For each subject of the dyad, the Pain Threshold (Pth) was established by increasing and decreasing the laser intensity in 0.5-watt steps. For the Pth, we considered the lowest intensity that enabled the subjects to perceive at least 50% of the stimuli as a painful pinprick. The laser intensity was two steps (1 step 1.5 W) over the Pain Threshold, in accordance with the procedure suggested by Treede (Treede, 2003; Treede et al., 2003) (Table 2). After the stimulation of the hand, we asked all subjects to rate the laser pain on a 0–100 visual analogue scale (VAS), where ‘0’ means feeling no pain and ‘100’ means the worst imaginable suffering. We stimulated one subject at a time, recording the EEG simultaneously from both participants. In the “Blind” condition, participants were separated by a panel and were invited to look at the dorsum of their right-hand during laser stimulation, while the partner was invited to look at her right non-stimulated hand. In the “Open” situation, the panel was removed, and each subject was invited to look at the other's stimulated hand, while being not stimulated, and to look at her own hand, while being stimulated (Fig. 1) We choose to limit the experiment to 30 stimuli in each stimulated situation (60 stimuli in total per subject) to avoid a time consuming task and damage to the skin. Also, to avoid fatigue or sensitization of nociceptors, we shifted the irradiated spot after each stimulus. The technician alerted both subjects just before 1 s of the stimulus delivering. The blind session preceded the open one. The dorsum of the hands was marked with a red sign to facilitate eyes fixation. Subjects were instructed to minimize eye movement and relax jaw muscles.
2. Methods 2.1. Subjects Twenty-one right-handed healthy women aged 25–40 years (32.3 ± 8.2 years) and 21 FM women, aged 26–40 years (31.8 ± 7.9 years) were evaluated during EEG recordings in pairs setting (Fig. 1). Healthy subjects were selected from the hospital and university staff. The FM patients were diagnosed at the Applied Neurophysiology and Pain Unit of Bari University, Bari Policlinico General Hospital, in accord with the ACR criteria (2010). None of the subjects reported neurological, psychiatric, or general medical diseases, except for FM diagnosis in the patients' group. All patients completed the Anxiety and Depression Zung Scales, and only patients with anxiety scores below 45 and depression scores below 50 were admitted to the study (Milic et al., 2019; Zung, 1965; Zung, 1971). No subject reported currently taking any drugs that act on the central nervous system in the last 3 months or any analgesic treatment in the last 48 h. The FM patients were admitted to the study after their first visit at the Applied Neurophysiology and Pain Unit of Bari University, and before taking the suggested treatment. The FM patients were examined before treatment prescription, and were also invited not to take analgesic drugs in the 48 h preceding the experiment, or to warn physicians about this possible cause of exclusion before the experiment started. The
2.3. Psychological tests Before the recording sessions, a psychologist administered the Empathy for Pain Scale (EPS) (Giummarra et al., 2015) to all subjects. In this scale, which was recently validated in healthy subjects, empathy for pain was measured on a 5-point scale (1 = strongly disagree; 5 = strongly agree) across 4 scenarios using 12 identical rated items. The scenarios were: (1) a person undergoing a surgical procedure (for example, as part of a television hospital drama); (2) a person who has recently had a surgical procedure (for example, they have stitches or bandaged amputation stump); (3) a person being accidentally injured 44
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 1. Study design. The technician alerted the subjects just before the delivering of laser stimuli.
(0.8 ± 0.5 in FM and 0.7 + 0.5 in control group). Laser evoked potentials (LEPs) obtained in the stimulated conditions were precomputed in the time interval 0–800 ms, using 70 Hz as low band pass filter and a baseline removing, considering the 100 ms preceding the laser stimulus. We precomputed Event Related Responses amplitude in the total considered time interval, and Statistical Probability Maps (SPMs) were obtained in the 800 ms (480 time frames at 256 Hz) following the laser stimuli. In particular, we checked the time intervals 160–190 for the N1, 210–230 for the N2 and 310–330 ms for the P2 component, taking into consideration previous normative data (de Tommaso et al., 2017a, 2017b), confirmed by visual analysis of single tracks. We also measured latencies of main components, using single tracks of Cz channel, referred to the nasion, for the N2 and P2 components, and the T3 channel, referred to Fz, for the N1 component, according to Treede et al. (2003). The EEGLAB ERSPs (Event Related Spectral Power perturbations) function was used to calculate the power spectrum changes related to laser stimulus delivering (Chien et al., 2014). In detail, Event-Related Spectral Perturbation (ERSP) and Inter-Trial Phase Coherence (ITC) images were based on 512 frames sampled at 256 Hz. Each trial contained samples from −1000 ms before to 996 ms after the time locking event, at 3 cycles at lowest frequency to 25.6 at highest. After a preliminary analysis of the relevant phenomena occurring in the total time window, two hundred time points (−443.4 to 439.5 ms) were generated and closest points for time variable were found. Time values for time/freq decomposition were not perfectly uniformly distributed, so the window size used was 285 samples (1113.28 ms) wide. The100 logspaced frequencies from 3.0 Hz to 128.0 Hz were computed. The mean baseline spectrum was calculated, and RMS was applied across channels (mask for at least 1 non-zeros values at each time/freq). The ERSP represented mean changes across the 31 scalp electrodes time locked to the laser stimulus. In the not stimulated condition, the possible brain response was obviously not perfectly time locked to the stimulus, though the analysis could be useful for the detection of possible EEG changes in approximate time intervals.
Table 1 Demographic and clinical features of Fibromyalgia (FM) patients. The Wide Pain Index here reported is included in the ACR criteria (Wolfe et al., 2010), and refers to the diffusion of pain in 19 sites of the body. N°
Age (years)
Age of illness (years)
Wide Pain Index
FM patients
19 F
Healthy subjects
21 M
M 31.5 SD 8.2 M 32.3 SD 8.2
M 5.5 SD 2.3
M 12.90 SD 4.77
(for example, in a car accident); and (4) a person being physically assaulted. The 12 response items were: distress, discomfort, disgust, fear, restlessness, sense of compassion, sense of what it feels like, a need to get help, a desire to look away, non-painful sensations, painful sensations, and visceral sensations (for example, nausea). We considered the total score evaluated across the four scenarios (Giummarra et al., 2015). Higher scores corresponded to higher levels of empathy. 2.4. Analysis of EEG data Preprocessing was performed in MATLAB, using the EEGLAB 14_1_1 tool. The data were first high-passed filtered at 1 Hz to remove slow drifts. Next, a notch filter at 50 Hz (L: 48, H: 52) and 100 Hz (L: 99, H: 101) was applied to remove power line noise artifacts. We applied the automatic channels and data segments. Artifacts components were then automatically removed, considering the components recorded on the EOG channels. Bad channels were identified by a semiautomatic method, based on visual detection and channel statistic. To precompute channels measures, spherical interpolation of missing channels and deletion of ICA artifact components pre-tagged in each dataset was performed. Channels presenting with distributions of potential values further away from a Gaussian distribution than other scalp channels, were also removed. Two cases in the FM group were not included in the analysis for an excess of bad channels. Two cases in the FM group and 1 case in the control group presented with 2 bad channels, and 2 cases in the FM and 3 in the control group with 1 bad channel. In both groups, we deleted on average 1 EEG segment for each recording session
2.5. Statistical analysis For the primary hypothesis, we conducted preliminary analysis in the single groups. In controls and patients groups, the laser evoked 45
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 2. Right: Grand average of Laser Evoked Potentials recorded in 21 healthy females in the conditions “Blind-Stimulated” and “Open-Stimulated”. Left: Event Related Spectral Perturbation (ERSP) computed on Cz channel. Spectral frequencies in the 3–110 Hz range are reported in Wavelet representation. Significant change between “Blind-Stimulated” and “Blind-Not Stimulated” conditions is present in the 200–400 ms time following the laser stimulus. In the lower part of the figure, the two dimensions maps represent the 9–11 Hz frequencies scalp distribution, and the results of parametric 2 ways ANOVA test, corrected for multiple comparisons.
Fig. 3. Right: Event Related Spectral Perturbation (ERSP) computed on O2 channel in 21 healthy controls. Spectral frequencies in the 3–110 Hz range are reported in Wavelet representation. An increase of beta activity is present in the“Open-Not Stimulated” situation, in the time close to the laser stimulus delivering to the partner. Left: the two dimensions maps represent the 15–20 Hz activity in the 0–50 ms time interval, and the results of parametric 2 ways ANOVA test, corrected for multiple comparisons. In the “Open-Not Stimulated” conditions, there was an increase of beta activity on occipital regions, during the partner's stimulation.
46
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 4. Right: Grand average of Laser Evoked Potentials recorded in 19 FM patients in the conditions “Blind-Stimulated” and “Open-Stimulated”. Left: Event Related Spectral Perturbation ERSP) computed on Cz channel. Spectral frequencies in the 3–110 Hz range are reported in Wavelet representation. Significant spectral changes are present during the 350–380 ms time following the laser stimulus, between the “Blind-Stimulated” and “Blind-Not Stimulated” conditions. The comparison of spectral components between the “Open-Stimulated” and “Open -Not Stimulated” conditions is not significant. In the lower part of the figure, the two dimensions maps represent the 2–5 Hz frequencies scalp distribution, and the results of parametric tests two ways ANOVA, corrected for multiple comparisons.
potentials and EEG spectral components were observed, and statistical analysis was performed applying the parametric statistic model of the EEGlab tool, corrected for Bonferroni multiple comparison, with a 2 × 2 ANOVA model, including the comparison of Blind-Open vs Stimulated-Not stimulated. For the comparison between groups, we applied the non-parametric statistic model with 2 × 2 ANOVA model (conditions Blind-Open vs FM patients-Controls) separately for the “Stimulated” and “Not Stimulated” sessions. Multiple comparisons bias was corrected by the Bonferroni method. Latencies of LEPs and the VAS values were compared by 2 × 2 ANOVA model with condition “Open-Stimulated” vs “BlindStimulated”- and groups -controls vs FM patients- as factors. The EPS scores were compared between groups, by the Student's ttest for unpaired data and correlated with relevant EEG and LEPs features by the linear regression analysis. For these analyses we used the SPSS software vers. 21.
with EPS below the median score, the remainder 11 in the subgroup with EPS above the median score. 3.1. Pain rating The pain threshold was similar between patients and controls. During the not-stimulated conditions, healthy subjects and FM patients did not report any sensation during the partner's laser stimulation, so these conditions were not considered in the analysis. Patients showed higher pain ratings, independent of the blind or open condition. The statistic comparison, showed an increase of pain sensation in FM patients, as an effect of group (Table 2). 3.2. Laser evoked potentials and Event Related Spectral Perturbation (ERSP) The tracks of 2 FM patients were discarded for excess of bad channels, so the final evaluation included 21 controls and 19 FM patients. In Table 1 demographic and clinical data are reported. Analysis in control group: Laser Evoked Potentials-The temporal N1 and the vertex N2P2 complex (Fig. 2 right) were similar in the condition Blind-Stimulated vs Open-Stimulated for amplitude and latency (see below). ERSP: the laser stimuli determined a power increase of the 3–12 Hz spectral components in the 200–400 post-stimulus interval. This activity was represented on the central regions. While in the Blind-stimulated situation the increase of alpha power in the 9–11 Hz range was
3. Results Empathy for Pain Scale EPS: patients seemed slightly more empathic than controls, according to the EPS scores, but the difference was not significant. (controls: 30.23 ± 9.92; FM patients: 35.48 ± 9). We subdivided the samples in subgroups, according to the median of EPS scores (30.87 in the total of cases), and we found that 14 controls were in the subgroup below the median score, 7 controls in the subgroup with EPS above the median score. Eight patients were in the subgroup 47
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 5. Two dimension maps of 19 FM patients, representing the beta activity in the frequency range 25–30 Hz, in the time interval preceding the laser stimulus. The results of parametric test two ways ANOVA, corrected for multiple comparisons, are also reported. Table 2 Laser evoked potentials amplitude and latencies, pain threshold and rating in patients (n°19) and controls (n°21). Mean and Standard Deviations (in brackets) and results of ANOVA analysis are reported.
Controls
Open Blind
FM
Open Blind
Anova
Group F P Condition F P Group × condition F P
N1 (ms)
N2 (ms)
P2 (ms)
N1 (uV)
N2P2 (uV)
Pain threshold (Watt)
VAS (0−100)
171.2 (23.4) 176.6 (18.2) 180.3 (10.1) 181.2 (12.2)
235.5 (12.2) 228.2 (13.3) 238.2 (15.5) 229.3 (13.3)
353.34 (23.4) 345.3 (30.7) 368.5 (34.3) 359.3 (29.9)
7.3 (2.5) 6.8 (3.9) 6.8 (3.3) 5.9 (2.2)
22.2 (8.99) 20.3 (11.32) 18.9 (12.2) 20.1 (10.5)
14.4 (2.32)
40.1 (21.1) 43.3 (20.2) 61.5 (29.9) 66.6 (23.3)
1.48 0.23
0.34 0.56
2.87 0.098
0.01 0.89
3.23 0.078
2.88 0.097
0.58 0.45
0.58 0.45
2.52 0.12
0.58 0.45
3.04 0.089
2.87 0.098
0.11 0.89.
0.27 0.76
0.34 0.71
0.89 0.4
2.23 0.12
2.33 0.11
relevant in respect to the Blind-not stimulated condition, the change between the Open-stimulated and Open-not stimulated condition was not significant. The interaction among conditions was not significant (Fig. 2 left). In the “Open-not stimulated” condition, there was an increase of the 3–7 Hz and 15–20 Hz spectral power in the 50 ms preceding and following the partner stimulation, compared to the “Blind-not stimulated” condition. The interaction among Stimulation vs Not Stimulation and Blind vs Open, was significant for the 15–20 Hz beta rhythm on occipital regions. (Fig. 3). Analysis in FM patients. Laser evoked potentials. In the patients group, LEP amplitude and latencies were similar between blind-stimulated and open-stimulated conditions (see below).
12.8 (4.35)
5.60 0.023
ERSP. The laser stimuli determined power increase in the 3–12 frequencies range, in the 250–380 ms time range following the stimulus. This activity had a prevalent representation over the vertex. The 3–12 Hz spectral power was significantly greater in the blind-stimulated condition, compared to the blind-not stimulated one, on central and frontal electrodes. In open condition, the significance between stimulated vs not stimulated was also present, but only on the frontal electrodes (Fig. 4). The beta band in the 25–30 Hz range decreased over the vertex in the Open-Not stimulated conditions, and also in the BlindStimulated situation, in the pre-stimulus interval. The interaction among the conditions, showed significant changes of beta power on the central regions in the 200–250 ms pre-stimulus time (Fig. 5).
48
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 6. Grand average of N2P2 vertex complex obtained in 21 controls and 19 FM patients in blind and open conditions.
paradigm was different from the majority or related studies, which were performed using functional magnetic resonance imaging (fMRI) while viewing pictures of limbs of persons in painful situations or during an interaction with a person subjected to a painful stimulation that was announced by a visual or acoustic warning (Lamm et al., 2011). Other studies have used the bio-electrical signals to describe the motor-system changes induced by empathic pain (Bucchioni et al., 2016). Few experiments have used LEPs (Valeriani et al., 2008) to show the empathic modulation induced by pictures of different pain situations. In this sense, our experiment reconstructed a “twin-type” model, which could facilitate mutual empathic participation during the presence or absence of painful stimulations. Despite the subjects received a warning of the laser stimulus, even in the “blind” situation, it seemed that the visual perception of the other person exerted an influence on the cortical pain processing, also while being not stimulated, which had different features in healthy controls and FM patients. In the following paragraphs, we will discuss in detail the results of LEPs and ERSP, and relevant general considerations. In patients and controls, the amplitude and latency of the LEPs did not change significantly due to the presence of the partner, compared to the condition of visual segregation. This means that the presence of the unstimulated partner, did not exert a relevant distractive effect from the own stimulated hand. Also, the subjective pain perception, was only slightly reduced in the open situation, confirming that both patients and controls, maintained attention toward painful stimuli. In fact the LEPs, and especially the vertex components, are reduced in amplitude in relation to the loss of relevance of the painful stimulus, a phenomenon not present in the “open” stimulation in patients and controls (Mouraux et al., 2011) The spectral compound of LEPs, mainly represented in the 2–11 Hz range, was obviously absent in not stimulated conditions (Xia et al., 2016). However, the statistical significance of the comparison between the stimulated and not stimulated situation, was absent in the “open” condition in the healthy group, and limited to the frontal electrodes in FM patients. Despite the interaction between factors stimulation vs not stimulation and open vs blind was not significant, something could occur in the spectral compound of EEG during the observation of the stimulated partner, which would be similar to the EEG changes related to the own hand stimulation. This phenomenon could be interesting, and in line with the assumption that pain empathy mechanisms generate in the same cortical areas devoted to individual pain processing
3.3. Comparison between patients and controls 3.3.1. LEPs The N1, N2 and P2 amplitudes and latencies were similar between groups in blind-stimulated and open-stimulated conditions. (Table 2; Fig. 6). 3.3.2. ERSP In the stimulated condition, the EEG power was similar between patients and controls in the 3–12 Hz frequency range. In the not stimulated condition, the occipital activity present in controls during the vision of the other's stimulated hand, was less represented in patients. The comparison between groups was significant in the time interval immediately preceding and following the partner's laser stimulation. (Fig. 7). 3.3.3. LEPs and ERSP in EPS subgroups The LEPs amplitude and latencies and the subjective pain rating were similar in EPS subgroups. In the Open-Not Stimulated condition, control subjects included in the group with higher empathy scores, displayed increased beta activity in the 28–30 Hz range over the parieto-occipital regions in the 50 ms. preceding and following the partner stimulation, as compared to other subjects. This pattern did not reach statistical significance (Fig. 8) However, the linear correlation between 28 and 30 Hz beta values averaged over the CP6 and O2 electrodes and the EPS score was significant (standardized beta coefficient: 0.48 t 2.36 p 0.021). In the Open-Not Stimulated condition, patients included in the EPS group with higher scores, displayed reduced beta rhythm on the central regions in the time preceding the other's stimulus delivering. (Fig. 9). The linear regression analysis between the average power of beta rhythm in the 25–30 Hz range computed over the fronto-central electrodes and EPS scores, was significant (standardized beta coefficient 0.72, t 3,59 p 0.0032). 4. Discussion The results of the present study seemed to indicate that seeing another person in a dyad-setting LEP recording could modulate the cortical response in an empathic way. In fact, the blind vs open condition seemed to determine some changes in EEG rhythms. Our experimental 49
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 7. Upper Event Related Spectral Perturbation (ERSP) computed on O2 channel in 21 control subjects and 19 FM patients in the Blind Not Stimulated and Open Not Stimulated conditions Spectral frequencies in the 3–110 Hz range are reported in Wavelet representation. The interaction between conditions “Blind Not Stimulated vs “Open Not Stimulated” and groups showed significant difference in beta rhythm range.
(Decety, 2009; Lamm et al., 2011). Unfortunately, it was not relevant, and probably variable in terms of individual response, time and frequency of appearance. The analysis of the EEG spectral perturbation in beta range, revealed significant changes in control subjects during the observation of the partner's hand during the stimulation. These changes, were not time locked to the stimulus delivering, as was expected, but occurred in temporal proximity of the partner's hand stimulation. In the blind not stimulated condition, control subjects exhibited a reduction of beta rhythm over the right occipito-parietal cortex, for a probable deactivation of visual attentional mechanism toward the own not stimulated hand. Numerous studies implicate a specific role of beta oscillations in pain processing (Hauck et al., 2015), and changes in beta rhythms synchronization were observed during the
observation of others' pain (Riečanský et al., 2015). However, these studies generally report that the beta rhythm desynchronization occurs over scalp derivations in proximity of somatosensory areas activation, while our control group had a relevant increase of beta power on the right parieto-occipital electrodes, as for an activation of focused visual attention. In addition, the most of researches demonstrated that the alpha–mu band is related to pain perception and plays a significant role in empathy for pain (Li et al., 2017; Perry et al., 2010), while in our experiment changes of alpha-mu power were present, but did not characterize in a substantial way the observation of the partner's pain stimulation. The task was based on visual focused attention on own or the partner's hand, in presence or in absence of the stimulation. The beta rhythm increases in relation to the expectancy of visual or auditory 50
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 8. Upper: event Related Spectral Perturbation (ERSP) computed on O2 channel in control subjects grouped for empathic score. Spectral frequencies in the 3–110 Hz range are reported in Wavelet representation. Lower: two dimension maps reporting the topographic distribution of beta rhythm spectral changes in the 50 ms preceding and following the partner stimulation. The statistical comparison among subgroups in Open-Not Stimulated condition, shows an increased activation of right parieto-occipital regions in empathic controls, which did not reach statistical significance.
empathic behavior in respect to controls, who could modulate the shared experience on the basis of the task peculiarity. This concept emerged also from the EEG pattern in healthy controls with higher empathic tendency, as shown by the EPS scores. Despite the comparison among different EPS subgroups did not reach a statistical significance, the prevalent activation of the right parieto-occipital cortex during the partner stimulation was evident in controls with EPS > 3. In FM patients, the desynchronization of beta rhythm over the fronto-central derivations, was clearer in the more empathic patients, thus confirming a different way of brain involvement in empathic processes in chronic pain patients and healthy controls. A recent review assessed the putative involvement of a set of brain regions activated by the sight of others in pain—that is, the anterior insula/inferior frontal gyrus (AI/ IFG), the anterior cingulate cortex, and the somatosensory and motor regions (Betti and Aglioti, 2016). Here we can suppose that the site of EEG changes due to empathic response could be modulated in dependence on the task modality in healthy controls, while patients with chronic pain manifest their empathic reaction activating the cortical networks subtending the response to relevant somatosensory stimuli. In other words, in the present experiment generic visual spatial attention EEG paradigms prevailed in controls, with a probable prominence of this pattern dependent upon the individual empathic characteristics, while FM patients anticipated the other's stimulation with the activation of somatosensory circuits with reduced involvement of visual spatial circuits. The LEPs features were similar between patients and controls, and empathic traits did not change the basic features of averaged responses and their spectral content in both groups. Previous studies described LEPs differences between FM patients and controls, but reduced habituation more than stable amplitude abnormalities of LEP waves characterized FM patients, who showed clinical and neurophysiological heterogeneity (de Tommaso, 2012; de Tommaso et al., 2014, de Tommaso et al., 2017a, b). Subjective pain sensation was higher in FM patients, with a normal pain threshold, as for a probable effect of reduced habituation across repetitive stimulations, in accord with previous results (de Tommaso et al., 2011). Here we decided to limit the analysis to LEPs amplitudes and spectral power, due to the type of experimental setting, though further studies would be dedicated to the attentional modulation of LEPs during the vision of other's pain.
stimuli, depending on the timing of the stimulus presentation (Meijer et al., 2016). It is possible that the EEG recording captured the phenomenon of visual expectancy of the other's pain, as the beta changes were quite superimposed to the time of the laser stimuli prompting. In such task, the focused attention to the partner's stimulated hand, was expressed by the beta band representation over the right parieto-occipital regions devoted to attentional concentration. This phenomenon could express the empathic anticipation of other's pain experience, as also suggested by the feature of controls with more evident empathic tracts. In this sense, the mode and site of EEG oscillation could not be unique during an empathic experience, but variable in relation to the specific task and the personal experience of pain. In fact, the expectancy of the partner's stimulation, provoked different EEG changes in FM patients. The beta rhythm was reduced on the fronto-central electrodes, in the time preceding the partner's stimulation, in a similar way as for the expectancy of own stimulation in blind situation. The desynchronization of beta rhythm is described in pain processing and during the empathic sharing of pain experience, mainly present on the somatosensory and motor areas, preparing to possible motor reaction (Hauck et al., 2015; Riečanský et al., 2015). In our experiment, subjects received a verbal warning of stimulus arriving, which could thus provoke the preparing reaction. It is conceivable that FM patients reacted to the stimulation of the partner, in a similar way as during the own stimulation in the condition of segregation from the other person, so they could concentrate and prepare themselves to own and other's pain in a quite similar modality. In an experiment exploring the facilitation of somatosensory detection under vicarious pain experience observation, the FM patients were equally accurate as controls, with few neglect errors, confirming facilitated activation of somatosensory networks during other's pain observation (Vandenbroucke et al., 2014). However, this interpretation is limited by the intrinsic procedural problems connected to the difficulty in extracting relevant changes by the not time locked laser stimuli delivered to the partner (see below). The comparison between the EEG behavior in patients and controls, stressed the different way of EEG change during the observation of the stimulated hand of the partner. What specially emerged, was the prevalence of right parieto-occipital activation in controls, absent in patients, confirming that in this peculiar task, the cortical areas subtending the visual sustained attention expressed the empathic reaction of healthy subjects. The history of chronic pain, could thus change the 51
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 9. Upper: event Related Spectral Perturbation (ERSP) computed on left fronto-central channel in FM patients grouped for empathic score. Spectral frequencies in the 3–110 Hz range are reported in Wavelet representation. Lower: two dimension maps reporting the topographic distribution of beta rhythm spectral changes in the 200–250 preceding the partner stimulation. The statistical comparison among subgroups in Open-Not Stimulated condition, shows a significant desynchronization of beta rhythm on central regions in patients with EPS score over the median value of 30.87.
5. Study limitation
6. Conclusions
There are several limitations of the results of the present study. In the interpretation of the results, we have to acknowledge that any conclusion on EEG changes related to not time locked stimuli, may be speculative. In addition, all subjects were alerted about incoming stimuli in both the blind and open conditions, so we could not evaluate the general arousal toward the own or the other's pain stimulation, but only the EEG reaction caused by the partner's hand vision. The study included 21 healthy subjects and 19 FM patients, which would be limited in the context of the experimental design and the number of statistical comparisons. The EPS groups were not balanced, though we decided to include the topographic reconstruction of statistical analysis, and to reinforce results using the linear regression analysis between EPS scores and EEG activity. A correlation with clinical features of FM patients was thus not reliable. The dyads composition was pseudo-randomized between patients and controls, but different brain responses could be present due to a partner with a history of pain. Healthy subjects selected in a hospital context, could have a different empathic characteristics as compared to general population.
The vision of a partner during a nociceptive stimulation in a known and subjectively proven modality, seemed to evoke a different EEG response, depending upon the personal experience of pain. In attendance of the partner's stimulation, FM patients desynchronize the same fronto-central regions as before own stimulation, while healthy subjects share the other's pain by activating scalp areas compatible with visual attention. The empathic reaction could thus be modulated by the task modality, at least in healthy subjects. Author contributions MD: study design, manuscript preparation, and data analysis. KR and AM: EEG recording and data analysis. MD: psychological assessment. EV: data analysis. SI: manuscript preparation and editing; cooperation for the study's design. 52
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Acknowledgment
evidence from an EEG study. Psychophysiology 54 (9), 1311–1322. https://doi.org/ 10.1111/psyp.12882. Martini, M., Valentini, E., Aglioti, S.M., 2013. Emotional conflict in a model modulates nociceptive processing in an onlooker: a laser-evoked potentials study. Exp. Brain Res. 225, 237–245. https://doi.org/10.1007/s00221-012-3365-4. Meijer, D., te Woerd, E., Praamstra, P., 2016. Timing of beta oscillatory synchronization and temporal prediction of upcoming stimuli. Neuroimage. 138, 233–241. https:// doi.org/10.1016/j.neuroimage.2016.05.071. Milic, V., Grujic, M., Barisic, J., Marinkovic-Eric, J., Duisin, D., Cirkovic, A., Damjanov, N., 2019. Personality, depression and anxiety in primary Sjogren's syndrome -association with sociodemographic factors and comorbidity. PLoS One 17, 14(1). https:// doi https://doi.org/10.1371/journal.pone.0210466. Mouraux, A., Diukova, A., Lee, M.C., Wise, R.G., Iannetti, G.D., 2011. A multisensory investigation of the functional significance of the “pain matrix”. Neuroimage 1, 2237–2249. https://doi.org/10.1016/j.neuroimage.2010.09.084. Perry, A., Bentin, S., Bartal, I.B.A., Lamm, C., Decety, J., 2010. “Feeling” the pain of those who are different from us: modulation of EEG in the mu/alpha range. Cogn. Affect. Behav. Neurosci. 10, 493–504. https://doi.org/10.3758/CABN.10.4.493. Riečanský, I., Paul, N., Kölble, S., Stieger, S., Lamm, C., 2015. Beta oscillations reveal ethnicity ingroup bias in sensorimotor resonance to pain of others. Soc. Cogn. Affect. Neurosci. 10 (7), 893–901. https://doi.org/10.1093/scan/nsu139. de Tommaso, M., 2012. Prevalence, clinical features and potential therapies for fibromyalgia in primary headaches. Expert. Rev. Neurother. 12 (3), 287–295. https:// doi.org/10.1586/ern.11.190. de Tommaso, M., Federici, A., Santostasi, R., Calabrese, R., Vecchio, E., Lapadula, G., Iannone, F., Lamberti, P., Livrea, P., 2011. Laser-evoked potentials habituation in fibromyalgia. J. Pain 12 (1), 116–124. https://doi.org/10.1016/j.jpain.2010.06.004. de Tommaso, M., Delussi, M., Vecchio, E., Sciruicchio, V., Invitto, S., Livrea, P., 2014. Sleep features and central sensitization symptoms in primary headache patients. J. Headache Pain 26 (15), 64. https://doi.org/10.1186/1129-2377-15-64. de Tommaso M, Ricci K, Montemurno A, Vecchio E., 2017a Age-related changes in laserevoked potentials following trigeminal and hand stimulation in healthy subjects. Eur J Pain. 2017 21,6,1087–1097. https://doi.org/10.1002/ejp.1010 de Tommaso, M., Sciruicchio, V., Delussi, M., Vecchio, E., Goffredo, M., Simeone, M., Barbaro, M.G.F., 2017b. Symptoms of central sensitization and comorbidity for juvenile fibromyalgia in childhood migraine: an observational study in a tertiary headache center. J. Headache Pain 18 (1), 59. https://doi.org/10.1186/s10194-0170764-8. Treede, R.D., 2003. Neurophysiological studies of pain pathways in peripheral and central nervous system disorders. J. Neurol. 250 (10), 1152–1161. https://doi.org/10.1007/ s00415-003-0237-7. Treede, R.D., Lorenz, J., Baumgärtner, U., 2003. Clinical usefulness of laser-evoked potentials. Neurophysiol. Clin. 33 (6), 303–314. https://doi.org/10.1016/j.neucli.2003. 10.009. Üçeyler, N., Zeller, J., Kewenig, S., Kittel-Schneider, S., Fallgatter, A.J., Sommer, C., 2015. Increased cortical activation upon painful stimulation in fibromyalgia syndrome. BMC Neurol. 20, 210. https://doi.org/10.1186/s12883-015-0472-4. Valeriani, M., Betti, V., Le Pera, D., De Armas, L., Miliucci, R., Restuccia, D., Avenanti, A., Aglioti, S.M., 2008. Seeing the pain of others while being in pain: a laser-evoked potentials study. Neuroimage 40, 1419–1428. https://doi.org/10.1016/j. neuroimage.2007.12.056. Vandenbroucke, S., Crombez, G., Harrar, V., Brusselmans, G., Devulder, J., Spence, C., Goubert, L., 2014. Fibromyalgia patients and controls are equally accurate in detecting tactile stimuli while observing another in pain: an experimental study. Attent, Percept, Psychophys. 76 (8), 2548–2559. https://doi.org/10.3758/s13414-0140729-9. de Vignemont, F., Singer, T., 2006. The empathic brain: how, when and why? Trends Cogn. Sci. 10, 435–441. https://doi.org/10.1016/j.tics.2006.08.008. Wolfe F, Clauw DJ, Fitzcharles MA, et al., 2010 The American College of Rheumatology preliminary diagnostic criteria for fibromyalgia and measurement of symptom severity. Arthritis Care Res (Hoboken) 62(5), 600–10. https://doi.org/10.1002/acr. 20140. Xia, X.L., Peng, W.W., Iannetti, G.D., Hu, L., 2016. Laser-evoked cortical responses in freely-moving rats reflect the activation of C-fibre afferent pathways. Neuroimage 128, 209–217. https://doi.org/10.1016/j.neuroimage.2015.12.042. Zung, W.W., 1965. As elf-rating depression scale. Arch Gen Psychiatry1965 12, 63–70. Zung WW. A rating instrument for anxiety disorders, 1971. Psychosomatics12:371–379.
The study was supported by Bari University research funds. It was awarded for the Poster Session at the Italian Congress of Psychophysiology and Cognitive Neuroscience, Milan, October 2016. Declaration of competing interest No authors declare a conflict of interest. Lower. Two dimension maps reporting the topographic distribution of beta rhythm spectral changes in the 50 ms preceding and following the partner stimulation. The maps represent the statistical comparison between the “Blind-Not-Stimulated” and “Open-Not-Stimulated” conditions in 21 controls and 19 FM patients. The results of unpaired test and two ways ANOVA, corrected for multiple comparisons, are also reported. References Balconi, M., Vanutelli, M.E., 2017. When cooperation was efficient or inefficient. Functional near-infrared spectroscopy evidence. Front. Syst. Neurosci. 11 (26). https://doi.org/10.3389/fnsys.2017.00026. Balconi, M., Gatti, L., Vanutelli, M.E., 2017. When cooperation goes wrong: brain and behavioural correlates of ineffective joint strategies in dyads. Int. J. Neurosci. 128 (2), 155–166. https://doi.org/10.1080/00207454.2017.1379519. Bernhardt, B.C., Singer, T., 2012. The neural basis of empathy. Annu. Rev. Neurosci. 35, 1–23. https://doi.org/10.1146/annurev-neuro-062111-150536. Betti, V., Aglioti, S.M., 2016. Dynamic construction of the neural networks underpinning empathy for pain. Neurosci. Biobehav. Rev. 63,191–206 https://doi.org/10.1016/j. neubiorev.2016.02.009 Bucchioni, G., Fossataro, C., Cavallo, A., Mouras, H., Neppi-Modona, M., Garbarini, F., 2016. Empathy or ownership? Evidence from corticospinal excitability modulation during pain observation. J. Cogn. Neurosci. 28, 1760–1771. https://doi.org/10. 1162/jocn_a_01003. Chien, J.H., Liu, C.C., Kim, J.H., Markman, T.M., Lenz, F.A., 2014. Painful cutaneous laser stimuli induce event-related oscillatory EEG activities that are different from those induced by nonpainful electrical stimuli. J. Neurophysiol. 112, 824–833. https://doi. org/10.1152/jn.00209.2014. Decety, J., 2009. Empathy, sympathy and the perception of pain. Pain. 145 (3), 365–366. https://doi.org/10.1016/j.pain.2009.08.006. Fallon, N., Li, X., Chiu, Y., Nurmikko, T., Stancak, A., 2015. Altered cortical processing of observed pain in patients with fibromyalgia syndrome. J. Pain 16 (8), 717–726. https://doi.org/10.1016/j.jpain.2015.04.008. Fitzgibbon, B.M., Giummarra, M.J., Georgiou-Karistianis, N., Enticott, P.G., Bradshaw, J.L., 2010. Shared pain: from empathy to synaesthesia. Neurosci. Biobehav. Rev. 34 (4), 500–512. https://doi.org/10.1016/j.neubiorev.2009.10.007. Giummarra, M.J., Fitzgibbon, B.M., Georgiou-Karistianis, N., Beukelman, M., VerdejoGarcia, A., Blumberg, Z., Chou, M., Gibson, S.J., 2015. Affective, sensory and empathic sharing of another's pain: the empathy for pain scale. Eur. J. Pain (United Kingdom) 19, 807–816. https://doi.org/10.1002/ejp.607. Hauck, M., Domnick, C., Lorenz, J., Gerloff, C., Engel, A.K., 2015. Top-down and bottomup modulation of pain-induced oscillations. Front. Hum. Neurosci. 2 (9), 375. https://doi.org/10.3389/fnhum.2015.00375. Jackson, P.L., Rainville, P., Decety, J., 2006. To what extent do we share the pain of others? Insight from the neural bases of pain empathy. Pain 125 (1–2), 5–9. https:// doi.org/10.1016/j.pain.2006.09.013. Lamm, C., Decety, J., Singer, T., 2011. Meta-analytic evidence for common and distinct neural networks associated with directly experienced pain and empathy for pain. Neuroimage 54, 2492–2502. https://doi.org/10.1016/j.neuroimage.2010.10.014. Lang, S., Yu, T., Markl, A., Müller, F., Kotchoubey, B., 2011. Hearing others' pain: neural activity related to empathy. Cogn. Affect. Behav. Neurosci. 11, 386–395. https://doi. org/10.3758/s13415-011-0035-0. Li, X., Meng, X., Li, H., Yang, J., Yuan, J., 2017. The impact of mood on empathy for pain:
53
Contents lists available at ScienceDirect
International Journal of Psychophysiology journal homepage: www.elsevier.com/locate/ijpsycho
Empathy for pain in fibromyalgia patients: An EEG study a,⁎
a
a
a
a
Marina de Tommaso , Katia Ricci , Giuliana Conca , Eleonora Vecchio , Marianna Delussi , Sara Invittob a b
T
Applied Neurophysiology and Pain Unit, TIRES Center, Bari Aldo Moro University, Italy Human Anatomy and Neuroscience Lab, DiSTeBA, University of Salento, Lecce, Italy
ARTICLE INFO
ABSTRACT
Keywords: Empathy Pain Fibromyalgia Laser evoked potentials Event related spectral perturbation
Background: Recent research in cognitive neurosciences highlights how the neural circuitries are activated during pain responses in empathic context. Aims: The present study was designed to test if healthy subjects and Fibromyalgia (FM) patients, both evaluated by Laser Evoked Potentials (LEPs) and Event-Related Spectral Perturbation (ERSP), might reveal the empathic response to the partner's nociceptive stimulation. Methods: The emphatic nociceptive paradigm was recorded through 64 channels EEG and laser stimulation of the right hand in a shared visual open setting (Open Condition) or in a blind setting (Blind condition) where the subjects didn't receive visual information about partner nociceptive condition. Twenty one healthy subjects and 19 FM patients were evaluated in pairs. All subjects were tested by the Empathy for Pain Scale (EPS). Results: The averaged LEPs were similar between patients and controls in the different conditions. In attendance of the partner's stimulation, FM patients desynchronized the same fronto-central regions as before own stimulation, while healthy subjects shared the other's pain by activating scalp areas compatible with visual attention. These EEG features were more represented in subjects with higher EPS scores. Conclusions: While empathic features of healthy subjects seemed influenced by the specific visual attentional task, patients expressed an EEG pattern compatible with somatosensory circuits activation in the expectation of own and other's pain. The visual empathic involvement in other's noxious stimulation could evoke a different EEG response depending upon the experience of chronic pain.
1. Introduction Empathy is the ability to consciously share the affective state of another individual, who is recognized as the source of that state, and generates similar perception in a ‘mirroring’ condition (de Vignemont and Singer, 2006). Recent research in the field of cognitive neuroscience has shed light on the neural circuits that are activated during responses to pain in an empathic context. Existing neuroscientific evidence suggests that empathizing with the pain of another person implies a topographic activity in regions similar to those associated with the direct experience of pain (Decety, 2009; Lamm et al., 2011). Neuroimaging studies on empathy of pain have shown that the cortical areas involved in the processing of one's own pain can also be activated during the empathic perception of another person's pain experience (Decety, 2009; Lamm et al., 2011). The empathic response is based on the same somatosensory circuits that underlie pain processing and on cortical areas involved in affective and emotional responses. It is
influenced by cortical areas included in the somatosensory system that regulates the pain elaboration, as well as by the emotional reaction. The anticipation of pain in others also evokes pain-related cortical responses, which seem to correspond with the severity of the pain expected (Jackson et al., 2006). Despite the existence of a large amount of research on empathy for pain, how the modality to share the others' pain could reflect the one's own pain elaboration is still an unresolved issue. In fact, most neuroimaging studies are based on the empathic reaction to another person's suffering, without previous personal experiences. The most commonly used paradigms are based on viewing pictures of people with limbs in painful situations; another type of paradigm is based on an interaction with a person subjected to a painful stimulation that was preceded by a visual or acoustic cue (Lamm et al., 2011). In the latter paradigm, sensory-motor processes, including mimicry, might strongly contribute to neural responses. The intrinsic modality of pain processing and affective reactions may influence the empathic correspondence with another person's suffering, with an
⁎ Corresponding author at: Applied Neurophysiology and Pain Unit, TIRES Center, Bari Aldo Moro University, Policlinico General Hospital, Giovanni XXIII Building, Via Amendola 207 A, 70124 Bari, Italy. E-mail address: [email protected] (M. de Tommaso).
https://doi.org/10.1016/j.ijpsycho.2019.09.007 Received 23 January 2019; Received in revised form 16 August 2019; Accepted 24 September 2019 Available online 21 October 2019 0167-8760/ © 2019 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
activation of the same nociceptive cortical circuits underlying personal experience. Thus, a vision of the other person's pain may evoke an empathic response, which can be reinforced by self-personal pain experience. Within the interaction paradigm (Bernhardt and Singer, 2012; Lang et al., 2011; Perry et al., 2010), we can use the concept of empathy, no longer understood as ‘looking at the other’ (Balconi and Vanutelli, 2017), but as a social system that can be generated from a context (Fitzgibbon et al., 2010). In this ‘new’ paradigm we hypothesized to construct a model of shared nociception, where the subjects became dyads (Balconi et al., 2017) and the setting allows a shared nociception. The empathic response seemed to be altered in patients with diseases that involved the nociceptive system and in disorders of affective and social relationships (Betti and Aglioti, 2016). Fibromyalgia is a chronic disabling disorder, characterized by wide spread pain, sleep disturbances, fatigue and other symptoms, which comprise the complex diagnosis (Wolfe et al., 2010). Vandenbroucke explored the spatial discrimination of tactile stimuli in FM patients and controls (Vandenbroucke et al., 2014), during the vision of videos depicting pain- and non-pain-related scene. Authors reported that the vision of others' pain facilitated spatial somatosensory detection in patients and controls. In an EEG study, FM patients showed increased activations associated with pain and non-pain pictures, suggesting that even innocuous stimuli with somatic connotations may involve the emotional state (Fallon et al., 2015). Üçeyler et al. found higher scores for personal distress and empathic concern in FM patients compared to controls, but they did not assess how these features could have impact on central pain elaboration (Üçeyler et al., 2015). The empathic response could thus make light on some aspects of pain processing in this complex disease. Laser-evoked potentials (LEPs) are a useful tool to explore the cortical activation under A-delta fiber input (Treede et al., 2003), including the psychophysiological changes due to empathic processes (Martini et al., 2013; Valeriani et al., 2008). The present study was designed to test 1) if in healthy subjects and FM patients the LEPs and the EEG rhythms might change under the vision of own or other's stimulated hand, during EEG recording session, adapted for a dyad 2) if these changes could be different between patients and controls, 3) if neurophysiological differences could depend upon the individual empathic behavior, expressed by the Empathy for Pain scale (Giummarra et al., 2015).
experimental tasks were performed in the morning. The subject's pairs were chosen in a pseudo-random way between participants who were not friends and do not know other's medical history. The study was approved by the Ethical Committee of Policlinico General Hospital of Bari (Italy). All subjects signed an informed consent. The study was conducted in accordance with the World Medical Association Declaration of Helsinki, which was published on the website of the Journal of American Medical Association. 2.2. Procedure While each subject was sitting in a comfortable chair, recordings were made using 31 scalp electrodes, (FP1,FP2,F3,Fz,F4,T3,T4,T5, T6,Cz,C3,C4,P3,P4,Pz,O1,Oz,O2,CP2,CP1,AF3,CPZ,AF4,CP6,CP5,FC5, FC1,FC2,FC6,TP6,TP7, according to the 10–20 International system) connected to a MICROMED amplifier. Four additional electrodes were positioned above and below both eyes to record the electrooculogram. The reference electrode was positioned at the nasion, and the ground electrode was positioned in the middle between the nasion and the FPZ. The EEG montage was prepared in the same room while subjects were separated by a panel screen, which avoided one person from being able to see the other person (Fig. 1). All the laser stimuli were thus triggered on the EEGs, and the tracks from both subjects were simultaneously visualized. Cutaneous heat stimuli were delivered with a CO2 laser (wavelength = 10.6 lm, 2-mm beam diameter; ELEN, Florence, Italy) on the dorsum of the right hand. The duration of stimulation was 30 ms. The laser beam light was directed to the hand during the entire stimulation session, independent of the stimuli delivering. For each subject of the dyad, the Pain Threshold (Pth) was established by increasing and decreasing the laser intensity in 0.5-watt steps. For the Pth, we considered the lowest intensity that enabled the subjects to perceive at least 50% of the stimuli as a painful pinprick. The laser intensity was two steps (1 step 1.5 W) over the Pain Threshold, in accordance with the procedure suggested by Treede (Treede, 2003; Treede et al., 2003) (Table 2). After the stimulation of the hand, we asked all subjects to rate the laser pain on a 0–100 visual analogue scale (VAS), where ‘0’ means feeling no pain and ‘100’ means the worst imaginable suffering. We stimulated one subject at a time, recording the EEG simultaneously from both participants. In the “Blind” condition, participants were separated by a panel and were invited to look at the dorsum of their right-hand during laser stimulation, while the partner was invited to look at her right non-stimulated hand. In the “Open” situation, the panel was removed, and each subject was invited to look at the other's stimulated hand, while being not stimulated, and to look at her own hand, while being stimulated (Fig. 1) We choose to limit the experiment to 30 stimuli in each stimulated situation (60 stimuli in total per subject) to avoid a time consuming task and damage to the skin. Also, to avoid fatigue or sensitization of nociceptors, we shifted the irradiated spot after each stimulus. The technician alerted both subjects just before 1 s of the stimulus delivering. The blind session preceded the open one. The dorsum of the hands was marked with a red sign to facilitate eyes fixation. Subjects were instructed to minimize eye movement and relax jaw muscles.
2. Methods 2.1. Subjects Twenty-one right-handed healthy women aged 25–40 years (32.3 ± 8.2 years) and 21 FM women, aged 26–40 years (31.8 ± 7.9 years) were evaluated during EEG recordings in pairs setting (Fig. 1). Healthy subjects were selected from the hospital and university staff. The FM patients were diagnosed at the Applied Neurophysiology and Pain Unit of Bari University, Bari Policlinico General Hospital, in accord with the ACR criteria (2010). None of the subjects reported neurological, psychiatric, or general medical diseases, except for FM diagnosis in the patients' group. All patients completed the Anxiety and Depression Zung Scales, and only patients with anxiety scores below 45 and depression scores below 50 were admitted to the study (Milic et al., 2019; Zung, 1965; Zung, 1971). No subject reported currently taking any drugs that act on the central nervous system in the last 3 months or any analgesic treatment in the last 48 h. The FM patients were admitted to the study after their first visit at the Applied Neurophysiology and Pain Unit of Bari University, and before taking the suggested treatment. The FM patients were examined before treatment prescription, and were also invited not to take analgesic drugs in the 48 h preceding the experiment, or to warn physicians about this possible cause of exclusion before the experiment started. The
2.3. Psychological tests Before the recording sessions, a psychologist administered the Empathy for Pain Scale (EPS) (Giummarra et al., 2015) to all subjects. In this scale, which was recently validated in healthy subjects, empathy for pain was measured on a 5-point scale (1 = strongly disagree; 5 = strongly agree) across 4 scenarios using 12 identical rated items. The scenarios were: (1) a person undergoing a surgical procedure (for example, as part of a television hospital drama); (2) a person who has recently had a surgical procedure (for example, they have stitches or bandaged amputation stump); (3) a person being accidentally injured 44
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 1. Study design. The technician alerted the subjects just before the delivering of laser stimuli.
(0.8 ± 0.5 in FM and 0.7 + 0.5 in control group). Laser evoked potentials (LEPs) obtained in the stimulated conditions were precomputed in the time interval 0–800 ms, using 70 Hz as low band pass filter and a baseline removing, considering the 100 ms preceding the laser stimulus. We precomputed Event Related Responses amplitude in the total considered time interval, and Statistical Probability Maps (SPMs) were obtained in the 800 ms (480 time frames at 256 Hz) following the laser stimuli. In particular, we checked the time intervals 160–190 for the N1, 210–230 for the N2 and 310–330 ms for the P2 component, taking into consideration previous normative data (de Tommaso et al., 2017a, 2017b), confirmed by visual analysis of single tracks. We also measured latencies of main components, using single tracks of Cz channel, referred to the nasion, for the N2 and P2 components, and the T3 channel, referred to Fz, for the N1 component, according to Treede et al. (2003). The EEGLAB ERSPs (Event Related Spectral Power perturbations) function was used to calculate the power spectrum changes related to laser stimulus delivering (Chien et al., 2014). In detail, Event-Related Spectral Perturbation (ERSP) and Inter-Trial Phase Coherence (ITC) images were based on 512 frames sampled at 256 Hz. Each trial contained samples from −1000 ms before to 996 ms after the time locking event, at 3 cycles at lowest frequency to 25.6 at highest. After a preliminary analysis of the relevant phenomena occurring in the total time window, two hundred time points (−443.4 to 439.5 ms) were generated and closest points for time variable were found. Time values for time/freq decomposition were not perfectly uniformly distributed, so the window size used was 285 samples (1113.28 ms) wide. The100 logspaced frequencies from 3.0 Hz to 128.0 Hz were computed. The mean baseline spectrum was calculated, and RMS was applied across channels (mask for at least 1 non-zeros values at each time/freq). The ERSP represented mean changes across the 31 scalp electrodes time locked to the laser stimulus. In the not stimulated condition, the possible brain response was obviously not perfectly time locked to the stimulus, though the analysis could be useful for the detection of possible EEG changes in approximate time intervals.
Table 1 Demographic and clinical features of Fibromyalgia (FM) patients. The Wide Pain Index here reported is included in the ACR criteria (Wolfe et al., 2010), and refers to the diffusion of pain in 19 sites of the body. N°
Age (years)
Age of illness (years)
Wide Pain Index
FM patients
19 F
Healthy subjects
21 M
M 31.5 SD 8.2 M 32.3 SD 8.2
M 5.5 SD 2.3
M 12.90 SD 4.77
(for example, in a car accident); and (4) a person being physically assaulted. The 12 response items were: distress, discomfort, disgust, fear, restlessness, sense of compassion, sense of what it feels like, a need to get help, a desire to look away, non-painful sensations, painful sensations, and visceral sensations (for example, nausea). We considered the total score evaluated across the four scenarios (Giummarra et al., 2015). Higher scores corresponded to higher levels of empathy. 2.4. Analysis of EEG data Preprocessing was performed in MATLAB, using the EEGLAB 14_1_1 tool. The data were first high-passed filtered at 1 Hz to remove slow drifts. Next, a notch filter at 50 Hz (L: 48, H: 52) and 100 Hz (L: 99, H: 101) was applied to remove power line noise artifacts. We applied the automatic channels and data segments. Artifacts components were then automatically removed, considering the components recorded on the EOG channels. Bad channels were identified by a semiautomatic method, based on visual detection and channel statistic. To precompute channels measures, spherical interpolation of missing channels and deletion of ICA artifact components pre-tagged in each dataset was performed. Channels presenting with distributions of potential values further away from a Gaussian distribution than other scalp channels, were also removed. Two cases in the FM group were not included in the analysis for an excess of bad channels. Two cases in the FM group and 1 case in the control group presented with 2 bad channels, and 2 cases in the FM and 3 in the control group with 1 bad channel. In both groups, we deleted on average 1 EEG segment for each recording session
2.5. Statistical analysis For the primary hypothesis, we conducted preliminary analysis in the single groups. In controls and patients groups, the laser evoked 45
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 2. Right: Grand average of Laser Evoked Potentials recorded in 21 healthy females in the conditions “Blind-Stimulated” and “Open-Stimulated”. Left: Event Related Spectral Perturbation (ERSP) computed on Cz channel. Spectral frequencies in the 3–110 Hz range are reported in Wavelet representation. Significant change between “Blind-Stimulated” and “Blind-Not Stimulated” conditions is present in the 200–400 ms time following the laser stimulus. In the lower part of the figure, the two dimensions maps represent the 9–11 Hz frequencies scalp distribution, and the results of parametric 2 ways ANOVA test, corrected for multiple comparisons.
Fig. 3. Right: Event Related Spectral Perturbation (ERSP) computed on O2 channel in 21 healthy controls. Spectral frequencies in the 3–110 Hz range are reported in Wavelet representation. An increase of beta activity is present in the“Open-Not Stimulated” situation, in the time close to the laser stimulus delivering to the partner. Left: the two dimensions maps represent the 15–20 Hz activity in the 0–50 ms time interval, and the results of parametric 2 ways ANOVA test, corrected for multiple comparisons. In the “Open-Not Stimulated” conditions, there was an increase of beta activity on occipital regions, during the partner's stimulation.
46
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 4. Right: Grand average of Laser Evoked Potentials recorded in 19 FM patients in the conditions “Blind-Stimulated” and “Open-Stimulated”. Left: Event Related Spectral Perturbation ERSP) computed on Cz channel. Spectral frequencies in the 3–110 Hz range are reported in Wavelet representation. Significant spectral changes are present during the 350–380 ms time following the laser stimulus, between the “Blind-Stimulated” and “Blind-Not Stimulated” conditions. The comparison of spectral components between the “Open-Stimulated” and “Open -Not Stimulated” conditions is not significant. In the lower part of the figure, the two dimensions maps represent the 2–5 Hz frequencies scalp distribution, and the results of parametric tests two ways ANOVA, corrected for multiple comparisons.
potentials and EEG spectral components were observed, and statistical analysis was performed applying the parametric statistic model of the EEGlab tool, corrected for Bonferroni multiple comparison, with a 2 × 2 ANOVA model, including the comparison of Blind-Open vs Stimulated-Not stimulated. For the comparison between groups, we applied the non-parametric statistic model with 2 × 2 ANOVA model (conditions Blind-Open vs FM patients-Controls) separately for the “Stimulated” and “Not Stimulated” sessions. Multiple comparisons bias was corrected by the Bonferroni method. Latencies of LEPs and the VAS values were compared by 2 × 2 ANOVA model with condition “Open-Stimulated” vs “BlindStimulated”- and groups -controls vs FM patients- as factors. The EPS scores were compared between groups, by the Student's ttest for unpaired data and correlated with relevant EEG and LEPs features by the linear regression analysis. For these analyses we used the SPSS software vers. 21.
with EPS below the median score, the remainder 11 in the subgroup with EPS above the median score. 3.1. Pain rating The pain threshold was similar between patients and controls. During the not-stimulated conditions, healthy subjects and FM patients did not report any sensation during the partner's laser stimulation, so these conditions were not considered in the analysis. Patients showed higher pain ratings, independent of the blind or open condition. The statistic comparison, showed an increase of pain sensation in FM patients, as an effect of group (Table 2). 3.2. Laser evoked potentials and Event Related Spectral Perturbation (ERSP) The tracks of 2 FM patients were discarded for excess of bad channels, so the final evaluation included 21 controls and 19 FM patients. In Table 1 demographic and clinical data are reported. Analysis in control group: Laser Evoked Potentials-The temporal N1 and the vertex N2P2 complex (Fig. 2 right) were similar in the condition Blind-Stimulated vs Open-Stimulated for amplitude and latency (see below). ERSP: the laser stimuli determined a power increase of the 3–12 Hz spectral components in the 200–400 post-stimulus interval. This activity was represented on the central regions. While in the Blind-stimulated situation the increase of alpha power in the 9–11 Hz range was
3. Results Empathy for Pain Scale EPS: patients seemed slightly more empathic than controls, according to the EPS scores, but the difference was not significant. (controls: 30.23 ± 9.92; FM patients: 35.48 ± 9). We subdivided the samples in subgroups, according to the median of EPS scores (30.87 in the total of cases), and we found that 14 controls were in the subgroup below the median score, 7 controls in the subgroup with EPS above the median score. Eight patients were in the subgroup 47
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 5. Two dimension maps of 19 FM patients, representing the beta activity in the frequency range 25–30 Hz, in the time interval preceding the laser stimulus. The results of parametric test two ways ANOVA, corrected for multiple comparisons, are also reported. Table 2 Laser evoked potentials amplitude and latencies, pain threshold and rating in patients (n°19) and controls (n°21). Mean and Standard Deviations (in brackets) and results of ANOVA analysis are reported.
Controls
Open Blind
FM
Open Blind
Anova
Group F P Condition F P Group × condition F P
N1 (ms)
N2 (ms)
P2 (ms)
N1 (uV)
N2P2 (uV)
Pain threshold (Watt)
VAS (0−100)
171.2 (23.4) 176.6 (18.2) 180.3 (10.1) 181.2 (12.2)
235.5 (12.2) 228.2 (13.3) 238.2 (15.5) 229.3 (13.3)
353.34 (23.4) 345.3 (30.7) 368.5 (34.3) 359.3 (29.9)
7.3 (2.5) 6.8 (3.9) 6.8 (3.3) 5.9 (2.2)
22.2 (8.99) 20.3 (11.32) 18.9 (12.2) 20.1 (10.5)
14.4 (2.32)
40.1 (21.1) 43.3 (20.2) 61.5 (29.9) 66.6 (23.3)
1.48 0.23
0.34 0.56
2.87 0.098
0.01 0.89
3.23 0.078
2.88 0.097
0.58 0.45
0.58 0.45
2.52 0.12
0.58 0.45
3.04 0.089
2.87 0.098
0.11 0.89.
0.27 0.76
0.34 0.71
0.89 0.4
2.23 0.12
2.33 0.11
relevant in respect to the Blind-not stimulated condition, the change between the Open-stimulated and Open-not stimulated condition was not significant. The interaction among conditions was not significant (Fig. 2 left). In the “Open-not stimulated” condition, there was an increase of the 3–7 Hz and 15–20 Hz spectral power in the 50 ms preceding and following the partner stimulation, compared to the “Blind-not stimulated” condition. The interaction among Stimulation vs Not Stimulation and Blind vs Open, was significant for the 15–20 Hz beta rhythm on occipital regions. (Fig. 3). Analysis in FM patients. Laser evoked potentials. In the patients group, LEP amplitude and latencies were similar between blind-stimulated and open-stimulated conditions (see below).
12.8 (4.35)
5.60 0.023
ERSP. The laser stimuli determined power increase in the 3–12 frequencies range, in the 250–380 ms time range following the stimulus. This activity had a prevalent representation over the vertex. The 3–12 Hz spectral power was significantly greater in the blind-stimulated condition, compared to the blind-not stimulated one, on central and frontal electrodes. In open condition, the significance between stimulated vs not stimulated was also present, but only on the frontal electrodes (Fig. 4). The beta band in the 25–30 Hz range decreased over the vertex in the Open-Not stimulated conditions, and also in the BlindStimulated situation, in the pre-stimulus interval. The interaction among the conditions, showed significant changes of beta power on the central regions in the 200–250 ms pre-stimulus time (Fig. 5).
48
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 6. Grand average of N2P2 vertex complex obtained in 21 controls and 19 FM patients in blind and open conditions.
paradigm was different from the majority or related studies, which were performed using functional magnetic resonance imaging (fMRI) while viewing pictures of limbs of persons in painful situations or during an interaction with a person subjected to a painful stimulation that was announced by a visual or acoustic warning (Lamm et al., 2011). Other studies have used the bio-electrical signals to describe the motor-system changes induced by empathic pain (Bucchioni et al., 2016). Few experiments have used LEPs (Valeriani et al., 2008) to show the empathic modulation induced by pictures of different pain situations. In this sense, our experiment reconstructed a “twin-type” model, which could facilitate mutual empathic participation during the presence or absence of painful stimulations. Despite the subjects received a warning of the laser stimulus, even in the “blind” situation, it seemed that the visual perception of the other person exerted an influence on the cortical pain processing, also while being not stimulated, which had different features in healthy controls and FM patients. In the following paragraphs, we will discuss in detail the results of LEPs and ERSP, and relevant general considerations. In patients and controls, the amplitude and latency of the LEPs did not change significantly due to the presence of the partner, compared to the condition of visual segregation. This means that the presence of the unstimulated partner, did not exert a relevant distractive effect from the own stimulated hand. Also, the subjective pain perception, was only slightly reduced in the open situation, confirming that both patients and controls, maintained attention toward painful stimuli. In fact the LEPs, and especially the vertex components, are reduced in amplitude in relation to the loss of relevance of the painful stimulus, a phenomenon not present in the “open” stimulation in patients and controls (Mouraux et al., 2011) The spectral compound of LEPs, mainly represented in the 2–11 Hz range, was obviously absent in not stimulated conditions (Xia et al., 2016). However, the statistical significance of the comparison between the stimulated and not stimulated situation, was absent in the “open” condition in the healthy group, and limited to the frontal electrodes in FM patients. Despite the interaction between factors stimulation vs not stimulation and open vs blind was not significant, something could occur in the spectral compound of EEG during the observation of the stimulated partner, which would be similar to the EEG changes related to the own hand stimulation. This phenomenon could be interesting, and in line with the assumption that pain empathy mechanisms generate in the same cortical areas devoted to individual pain processing
3.3. Comparison between patients and controls 3.3.1. LEPs The N1, N2 and P2 amplitudes and latencies were similar between groups in blind-stimulated and open-stimulated conditions. (Table 2; Fig. 6). 3.3.2. ERSP In the stimulated condition, the EEG power was similar between patients and controls in the 3–12 Hz frequency range. In the not stimulated condition, the occipital activity present in controls during the vision of the other's stimulated hand, was less represented in patients. The comparison between groups was significant in the time interval immediately preceding and following the partner's laser stimulation. (Fig. 7). 3.3.3. LEPs and ERSP in EPS subgroups The LEPs amplitude and latencies and the subjective pain rating were similar in EPS subgroups. In the Open-Not Stimulated condition, control subjects included in the group with higher empathy scores, displayed increased beta activity in the 28–30 Hz range over the parieto-occipital regions in the 50 ms. preceding and following the partner stimulation, as compared to other subjects. This pattern did not reach statistical significance (Fig. 8) However, the linear correlation between 28 and 30 Hz beta values averaged over the CP6 and O2 electrodes and the EPS score was significant (standardized beta coefficient: 0.48 t 2.36 p 0.021). In the Open-Not Stimulated condition, patients included in the EPS group with higher scores, displayed reduced beta rhythm on the central regions in the time preceding the other's stimulus delivering. (Fig. 9). The linear regression analysis between the average power of beta rhythm in the 25–30 Hz range computed over the fronto-central electrodes and EPS scores, was significant (standardized beta coefficient 0.72, t 3,59 p 0.0032). 4. Discussion The results of the present study seemed to indicate that seeing another person in a dyad-setting LEP recording could modulate the cortical response in an empathic way. In fact, the blind vs open condition seemed to determine some changes in EEG rhythms. Our experimental 49
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 7. Upper Event Related Spectral Perturbation (ERSP) computed on O2 channel in 21 control subjects and 19 FM patients in the Blind Not Stimulated and Open Not Stimulated conditions Spectral frequencies in the 3–110 Hz range are reported in Wavelet representation. The interaction between conditions “Blind Not Stimulated vs “Open Not Stimulated” and groups showed significant difference in beta rhythm range.
(Decety, 2009; Lamm et al., 2011). Unfortunately, it was not relevant, and probably variable in terms of individual response, time and frequency of appearance. The analysis of the EEG spectral perturbation in beta range, revealed significant changes in control subjects during the observation of the partner's hand during the stimulation. These changes, were not time locked to the stimulus delivering, as was expected, but occurred in temporal proximity of the partner's hand stimulation. In the blind not stimulated condition, control subjects exhibited a reduction of beta rhythm over the right occipito-parietal cortex, for a probable deactivation of visual attentional mechanism toward the own not stimulated hand. Numerous studies implicate a specific role of beta oscillations in pain processing (Hauck et al., 2015), and changes in beta rhythms synchronization were observed during the
observation of others' pain (Riečanský et al., 2015). However, these studies generally report that the beta rhythm desynchronization occurs over scalp derivations in proximity of somatosensory areas activation, while our control group had a relevant increase of beta power on the right parieto-occipital electrodes, as for an activation of focused visual attention. In addition, the most of researches demonstrated that the alpha–mu band is related to pain perception and plays a significant role in empathy for pain (Li et al., 2017; Perry et al., 2010), while in our experiment changes of alpha-mu power were present, but did not characterize in a substantial way the observation of the partner's pain stimulation. The task was based on visual focused attention on own or the partner's hand, in presence or in absence of the stimulation. The beta rhythm increases in relation to the expectancy of visual or auditory 50
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 8. Upper: event Related Spectral Perturbation (ERSP) computed on O2 channel in control subjects grouped for empathic score. Spectral frequencies in the 3–110 Hz range are reported in Wavelet representation. Lower: two dimension maps reporting the topographic distribution of beta rhythm spectral changes in the 50 ms preceding and following the partner stimulation. The statistical comparison among subgroups in Open-Not Stimulated condition, shows an increased activation of right parieto-occipital regions in empathic controls, which did not reach statistical significance.
empathic behavior in respect to controls, who could modulate the shared experience on the basis of the task peculiarity. This concept emerged also from the EEG pattern in healthy controls with higher empathic tendency, as shown by the EPS scores. Despite the comparison among different EPS subgroups did not reach a statistical significance, the prevalent activation of the right parieto-occipital cortex during the partner stimulation was evident in controls with EPS > 3. In FM patients, the desynchronization of beta rhythm over the fronto-central derivations, was clearer in the more empathic patients, thus confirming a different way of brain involvement in empathic processes in chronic pain patients and healthy controls. A recent review assessed the putative involvement of a set of brain regions activated by the sight of others in pain—that is, the anterior insula/inferior frontal gyrus (AI/ IFG), the anterior cingulate cortex, and the somatosensory and motor regions (Betti and Aglioti, 2016). Here we can suppose that the site of EEG changes due to empathic response could be modulated in dependence on the task modality in healthy controls, while patients with chronic pain manifest their empathic reaction activating the cortical networks subtending the response to relevant somatosensory stimuli. In other words, in the present experiment generic visual spatial attention EEG paradigms prevailed in controls, with a probable prominence of this pattern dependent upon the individual empathic characteristics, while FM patients anticipated the other's stimulation with the activation of somatosensory circuits with reduced involvement of visual spatial circuits. The LEPs features were similar between patients and controls, and empathic traits did not change the basic features of averaged responses and their spectral content in both groups. Previous studies described LEPs differences between FM patients and controls, but reduced habituation more than stable amplitude abnormalities of LEP waves characterized FM patients, who showed clinical and neurophysiological heterogeneity (de Tommaso, 2012; de Tommaso et al., 2014, de Tommaso et al., 2017a, b). Subjective pain sensation was higher in FM patients, with a normal pain threshold, as for a probable effect of reduced habituation across repetitive stimulations, in accord with previous results (de Tommaso et al., 2011). Here we decided to limit the analysis to LEPs amplitudes and spectral power, due to the type of experimental setting, though further studies would be dedicated to the attentional modulation of LEPs during the vision of other's pain.
stimuli, depending on the timing of the stimulus presentation (Meijer et al., 2016). It is possible that the EEG recording captured the phenomenon of visual expectancy of the other's pain, as the beta changes were quite superimposed to the time of the laser stimuli prompting. In such task, the focused attention to the partner's stimulated hand, was expressed by the beta band representation over the right parieto-occipital regions devoted to attentional concentration. This phenomenon could express the empathic anticipation of other's pain experience, as also suggested by the feature of controls with more evident empathic tracts. In this sense, the mode and site of EEG oscillation could not be unique during an empathic experience, but variable in relation to the specific task and the personal experience of pain. In fact, the expectancy of the partner's stimulation, provoked different EEG changes in FM patients. The beta rhythm was reduced on the fronto-central electrodes, in the time preceding the partner's stimulation, in a similar way as for the expectancy of own stimulation in blind situation. The desynchronization of beta rhythm is described in pain processing and during the empathic sharing of pain experience, mainly present on the somatosensory and motor areas, preparing to possible motor reaction (Hauck et al., 2015; Riečanský et al., 2015). In our experiment, subjects received a verbal warning of stimulus arriving, which could thus provoke the preparing reaction. It is conceivable that FM patients reacted to the stimulation of the partner, in a similar way as during the own stimulation in the condition of segregation from the other person, so they could concentrate and prepare themselves to own and other's pain in a quite similar modality. In an experiment exploring the facilitation of somatosensory detection under vicarious pain experience observation, the FM patients were equally accurate as controls, with few neglect errors, confirming facilitated activation of somatosensory networks during other's pain observation (Vandenbroucke et al., 2014). However, this interpretation is limited by the intrinsic procedural problems connected to the difficulty in extracting relevant changes by the not time locked laser stimuli delivered to the partner (see below). The comparison between the EEG behavior in patients and controls, stressed the different way of EEG change during the observation of the stimulated hand of the partner. What specially emerged, was the prevalence of right parieto-occipital activation in controls, absent in patients, confirming that in this peculiar task, the cortical areas subtending the visual sustained attention expressed the empathic reaction of healthy subjects. The history of chronic pain, could thus change the 51
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Fig. 9. Upper: event Related Spectral Perturbation (ERSP) computed on left fronto-central channel in FM patients grouped for empathic score. Spectral frequencies in the 3–110 Hz range are reported in Wavelet representation. Lower: two dimension maps reporting the topographic distribution of beta rhythm spectral changes in the 200–250 preceding the partner stimulation. The statistical comparison among subgroups in Open-Not Stimulated condition, shows a significant desynchronization of beta rhythm on central regions in patients with EPS score over the median value of 30.87.
5. Study limitation
6. Conclusions
There are several limitations of the results of the present study. In the interpretation of the results, we have to acknowledge that any conclusion on EEG changes related to not time locked stimuli, may be speculative. In addition, all subjects were alerted about incoming stimuli in both the blind and open conditions, so we could not evaluate the general arousal toward the own or the other's pain stimulation, but only the EEG reaction caused by the partner's hand vision. The study included 21 healthy subjects and 19 FM patients, which would be limited in the context of the experimental design and the number of statistical comparisons. The EPS groups were not balanced, though we decided to include the topographic reconstruction of statistical analysis, and to reinforce results using the linear regression analysis between EPS scores and EEG activity. A correlation with clinical features of FM patients was thus not reliable. The dyads composition was pseudo-randomized between patients and controls, but different brain responses could be present due to a partner with a history of pain. Healthy subjects selected in a hospital context, could have a different empathic characteristics as compared to general population.
The vision of a partner during a nociceptive stimulation in a known and subjectively proven modality, seemed to evoke a different EEG response, depending upon the personal experience of pain. In attendance of the partner's stimulation, FM patients desynchronize the same fronto-central regions as before own stimulation, while healthy subjects share the other's pain by activating scalp areas compatible with visual attention. The empathic reaction could thus be modulated by the task modality, at least in healthy subjects. Author contributions MD: study design, manuscript preparation, and data analysis. KR and AM: EEG recording and data analysis. MD: psychological assessment. EV: data analysis. SI: manuscript preparation and editing; cooperation for the study's design. 52
International Journal of Psychophysiology 146 (2019) 43–53
M. de Tommaso, et al.
Acknowledgment
evidence from an EEG study. Psychophysiology 54 (9), 1311–1322. https://doi.org/ 10.1111/psyp.12882. Martini, M., Valentini, E., Aglioti, S.M., 2013. Emotional conflict in a model modulates nociceptive processing in an onlooker: a laser-evoked potentials study. Exp. Brain Res. 225, 237–245. https://doi.org/10.1007/s00221-012-3365-4. Meijer, D., te Woerd, E., Praamstra, P., 2016. Timing of beta oscillatory synchronization and temporal prediction of upcoming stimuli. Neuroimage. 138, 233–241. https:// doi.org/10.1016/j.neuroimage.2016.05.071. Milic, V., Grujic, M., Barisic, J., Marinkovic-Eric, J., Duisin, D., Cirkovic, A., Damjanov, N., 2019. Personality, depression and anxiety in primary Sjogren's syndrome -association with sociodemographic factors and comorbidity. PLoS One 17, 14(1). https:// doi https://doi.org/10.1371/journal.pone.0210466. Mouraux, A., Diukova, A., Lee, M.C., Wise, R.G., Iannetti, G.D., 2011. A multisensory investigation of the functional significance of the “pain matrix”. Neuroimage 1, 2237–2249. https://doi.org/10.1016/j.neuroimage.2010.09.084. Perry, A., Bentin, S., Bartal, I.B.A., Lamm, C., Decety, J., 2010. “Feeling” the pain of those who are different from us: modulation of EEG in the mu/alpha range. Cogn. Affect. Behav. Neurosci. 10, 493–504. https://doi.org/10.3758/CABN.10.4.493. Riečanský, I., Paul, N., Kölble, S., Stieger, S., Lamm, C., 2015. Beta oscillations reveal ethnicity ingroup bias in sensorimotor resonance to pain of others. Soc. Cogn. Affect. Neurosci. 10 (7), 893–901. https://doi.org/10.1093/scan/nsu139. de Tommaso, M., 2012. Prevalence, clinical features and potential therapies for fibromyalgia in primary headaches. Expert. Rev. Neurother. 12 (3), 287–295. https:// doi.org/10.1586/ern.11.190. de Tommaso, M., Federici, A., Santostasi, R., Calabrese, R., Vecchio, E., Lapadula, G., Iannone, F., Lamberti, P., Livrea, P., 2011. Laser-evoked potentials habituation in fibromyalgia. J. Pain 12 (1), 116–124. https://doi.org/10.1016/j.jpain.2010.06.004. de Tommaso, M., Delussi, M., Vecchio, E., Sciruicchio, V., Invitto, S., Livrea, P., 2014. Sleep features and central sensitization symptoms in primary headache patients. J. Headache Pain 26 (15), 64. https://doi.org/10.1186/1129-2377-15-64. de Tommaso M, Ricci K, Montemurno A, Vecchio E., 2017a Age-related changes in laserevoked potentials following trigeminal and hand stimulation in healthy subjects. Eur J Pain. 2017 21,6,1087–1097. https://doi.org/10.1002/ejp.1010 de Tommaso, M., Sciruicchio, V., Delussi, M., Vecchio, E., Goffredo, M., Simeone, M., Barbaro, M.G.F., 2017b. Symptoms of central sensitization and comorbidity for juvenile fibromyalgia in childhood migraine: an observational study in a tertiary headache center. J. Headache Pain 18 (1), 59. https://doi.org/10.1186/s10194-0170764-8. Treede, R.D., 2003. Neurophysiological studies of pain pathways in peripheral and central nervous system disorders. J. Neurol. 250 (10), 1152–1161. https://doi.org/10.1007/ s00415-003-0237-7. Treede, R.D., Lorenz, J., Baumgärtner, U., 2003. Clinical usefulness of laser-evoked potentials. Neurophysiol. Clin. 33 (6), 303–314. https://doi.org/10.1016/j.neucli.2003. 10.009. Üçeyler, N., Zeller, J., Kewenig, S., Kittel-Schneider, S., Fallgatter, A.J., Sommer, C., 2015. Increased cortical activation upon painful stimulation in fibromyalgia syndrome. BMC Neurol. 20, 210. https://doi.org/10.1186/s12883-015-0472-4. Valeriani, M., Betti, V., Le Pera, D., De Armas, L., Miliucci, R., Restuccia, D., Avenanti, A., Aglioti, S.M., 2008. Seeing the pain of others while being in pain: a laser-evoked potentials study. Neuroimage 40, 1419–1428. https://doi.org/10.1016/j. neuroimage.2007.12.056. Vandenbroucke, S., Crombez, G., Harrar, V., Brusselmans, G., Devulder, J., Spence, C., Goubert, L., 2014. Fibromyalgia patients and controls are equally accurate in detecting tactile stimuli while observing another in pain: an experimental study. Attent, Percept, Psychophys. 76 (8), 2548–2559. https://doi.org/10.3758/s13414-0140729-9. de Vignemont, F., Singer, T., 2006. The empathic brain: how, when and why? Trends Cogn. Sci. 10, 435–441. https://doi.org/10.1016/j.tics.2006.08.008. Wolfe F, Clauw DJ, Fitzcharles MA, et al., 2010 The American College of Rheumatology preliminary diagnostic criteria for fibromyalgia and measurement of symptom severity. Arthritis Care Res (Hoboken) 62(5), 600–10. https://doi.org/10.1002/acr. 20140. Xia, X.L., Peng, W.W., Iannetti, G.D., Hu, L., 2016. Laser-evoked cortical responses in freely-moving rats reflect the activation of C-fibre afferent pathways. Neuroimage 128, 209–217. https://doi.org/10.1016/j.neuroimage.2015.12.042. Zung, W.W., 1965. As elf-rating depression scale. Arch Gen Psychiatry1965 12, 63–70. Zung WW. A rating instrument for anxiety disorders, 1971. Psychosomatics12:371–379.
The study was supported by Bari University research funds. It was awarded for the Poster Session at the Italian Congress of Psychophysiology and Cognitive Neuroscience, Milan, October 2016. Declaration of competing interest No authors declare a conflict of interest. Lower. Two dimension maps reporting the topographic distribution of beta rhythm spectral changes in the 50 ms preceding and following the partner stimulation. The maps represent the statistical comparison between the “Blind-Not-Stimulated” and “Open-Not-Stimulated” conditions in 21 controls and 19 FM patients. The results of unpaired test and two ways ANOVA, corrected for multiple comparisons, are also reported. References Balconi, M., Vanutelli, M.E., 2017. When cooperation was efficient or inefficient. Functional near-infrared spectroscopy evidence. Front. Syst. Neurosci. 11 (26). https://doi.org/10.3389/fnsys.2017.00026. Balconi, M., Gatti, L., Vanutelli, M.E., 2017. When cooperation goes wrong: brain and behavioural correlates of ineffective joint strategies in dyads. Int. J. Neurosci. 128 (2), 155–166. https://doi.org/10.1080/00207454.2017.1379519. Bernhardt, B.C., Singer, T., 2012. The neural basis of empathy. Annu. Rev. Neurosci. 35, 1–23. https://doi.org/10.1146/annurev-neuro-062111-150536. Betti, V., Aglioti, S.M., 2016. Dynamic construction of the neural networks underpinning empathy for pain. Neurosci. Biobehav. Rev. 63,191–206 https://doi.org/10.1016/j. neubiorev.2016.02.009 Bucchioni, G., Fossataro, C., Cavallo, A., Mouras, H., Neppi-Modona, M., Garbarini, F., 2016. Empathy or ownership? Evidence from corticospinal excitability modulation during pain observation. J. Cogn. Neurosci. 28, 1760–1771. https://doi.org/10. 1162/jocn_a_01003. Chien, J.H., Liu, C.C., Kim, J.H., Markman, T.M., Lenz, F.A., 2014. Painful cutaneous laser stimuli induce event-related oscillatory EEG activities that are different from those induced by nonpainful electrical stimuli. J. Neurophysiol. 112, 824–833. https://doi. org/10.1152/jn.00209.2014. Decety, J., 2009. Empathy, sympathy and the perception of pain. Pain. 145 (3), 365–366. https://doi.org/10.1016/j.pain.2009.08.006. Fallon, N., Li, X., Chiu, Y., Nurmikko, T., Stancak, A., 2015. Altered cortical processing of observed pain in patients with fibromyalgia syndrome. J. Pain 16 (8), 717–726. https://doi.org/10.1016/j.jpain.2015.04.008. Fitzgibbon, B.M., Giummarra, M.J., Georgiou-Karistianis, N., Enticott, P.G., Bradshaw, J.L., 2010. Shared pain: from empathy to synaesthesia. Neurosci. Biobehav. Rev. 34 (4), 500–512. https://doi.org/10.1016/j.neubiorev.2009.10.007. Giummarra, M.J., Fitzgibbon, B.M., Georgiou-Karistianis, N., Beukelman, M., VerdejoGarcia, A., Blumberg, Z., Chou, M., Gibson, S.J., 2015. Affective, sensory and empathic sharing of another's pain: the empathy for pain scale. Eur. J. Pain (United Kingdom) 19, 807–816. https://doi.org/10.1002/ejp.607. Hauck, M., Domnick, C., Lorenz, J., Gerloff, C., Engel, A.K., 2015. Top-down and bottomup modulation of pain-induced oscillations. Front. Hum. Neurosci. 2 (9), 375. https://doi.org/10.3389/fnhum.2015.00375. Jackson, P.L., Rainville, P., Decety, J., 2006. To what extent do we share the pain of others? Insight from the neural bases of pain empathy. Pain 125 (1–2), 5–9. https:// doi.org/10.1016/j.pain.2006.09.013. Lamm, C., Decety, J., Singer, T., 2011. Meta-analytic evidence for common and distinct neural networks associated with directly experienced pain and empathy for pain. Neuroimage 54, 2492–2502. https://doi.org/10.1016/j.neuroimage.2010.10.014. Lang, S., Yu, T., Markl, A., Müller, F., Kotchoubey, B., 2011. Hearing others' pain: neural activity related to empathy. Cogn. Affect. Behav. Neurosci. 11, 386–395. https://doi. org/10.3758/s13415-011-0035-0. Li, X., Meng, X., Li, H., Yang, J., Yuan, J., 2017. The impact of mood on empathy for pain:
53