The multilevel organization of vicarious pain responses: Effects of pain cues and empathy traits on spinal nociception and acute pain

PAIN�

xxx (2011) xxx–xxx

w w w . e l s e v i e r . c o m / l o c a t e / p a i n

The multilevel organization of vicarious pain responses: Effects of pain cuesand empathy traits on spinal nociception and acute pain

Etienne Vachon-Presseau a,b,c,⇑, Marc O. Martel d, Mathieu Roy e, Etienne Caron b,c,f, Philip L. Jackson g,Pierre Rainville b,c,f,h

a Département de Psychologie, Université de Montréal, Montreal, QC, Canadab Centre de recherche de l’Institut universitaire de gériatrie de Montréal (CRIUGM), Montreal, Quebec, Canadac Centre de recherche en neuropsychologie et cognition (CERNEC), Université de Montréal, Montreal, QC, Canadad Department of Psychology, McGill University, Montreal, QC, Canadae Department of Psychology, Columbia University, New York, NY, USAf Department of Stomatology, Faculty of Dentistry, Université de Montréal, Montreal, QC, Canadag École de psychologie and CIRRIS and CRULRG, Université Laval, Quebec, QC, Canadah Groupe de recherche sur le système nerveux central (GRSNC), Université de Montréal, Montreal, QC, Canada

Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.

a r t i c l e i n f o

Article history:Received 27 July 2010Received in revised form 1 February 2011Accepted 16 February 2011Available online xxxx

Keywords:EmpathyNociceptive flexion reflexPain modulationSelf-regulation

0304-3959/$36.00 � 2011 International Associationdoi:10.1016/j.pain.2011.02.039

⇑ Corresponding author. Tel.: +1 514 340 3540 x41E-mail address: etienne.vachon-presseau@umontr

Please cite this article in press as: Vachon-Presseon spinal nociception and acute pain. PAIN

�(20

a b s t r a c t

The shared-representation model of empathy suggests that vicarious pain processes rely partly on theactivation of brain systems underlying self-pain in the observer. Here, we tested the hypothesis thatself-pain may be facilitated by the vicarious priming of neural systems underlying pain perception.Pictures illustrating painful agents applied to the hand or the foot (sensory information), or painfulfacial expressions (emotional information) were shown to 43 participants to test the effects of vicariouspain on the nociceptive flexion reflex (NFR) of the lower limb and pain intensity and unpleasantnessproduced by transcutaneous electrical stimulation applied over the sural nerve. Results confirmedthe expected priming effects of vicarious pain on spinal and perceptual processes. However, for compa-rable pain intensity and arousal evoked by the pain pictures, the facilitation of the NFR and the self-painunpleasantness measurements was more robust in response to pictures depicting pain sensory com-pared to emotional information. Furthermore, the facilitation of the NFR by pain pictures was positivelycorrelated with the empathy trait of the observer. In contrast, the change in perceived shock-pain inten-sity was negatively correlated with empathic traits. This dissociation implies that low-level vicariouspriming processes underlying pain facilitation may be downregulated at higher pain-processing stagesin individuals reporting higher levels of empathy. We speculate that this process contributes to reduc-ing self–other assimilation and is necessary to adopt higher-order empathic responses and altruisticbehaviors.

� 2011 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.

1. Introduction

Altruism typically involves overcoming self-oriented responsesfor the benefit of other individuals and is thought, at least in high-er-order species, to rely on the empathic ability to represent thepain and suffering of others. In agreement with the Perception-Action Model of empathy proposed by Preston and de Waal [28],single-cell recording [16], neuroimaging [18,25,35], event-relatedpotentials [5], and transcranial magnetic stimulation [1] studiessuggest that the perception of pain in others relies, at least inpart, on a shared-representation system involving neural circuitssubserving the first-person experience of pain. More precisely,

for the Study of Pain. Published by

35.eal.ca (E. Vachon-Presseau).

au E et al. The multilevel organ11), doi:10.1016/j.pain.2011.02

witnessing body limbs threatened or stimulated by a painful agent[18], observing facial expressions of pain [4,33,34], or simply view-ing visual cues signaling pain in others [35] activates parts of thecortical signature normally associated with pain perception and re-ferred to as the ‘‘pain matrix.’’ Beyond these automatic responses,second-order attributions permit the inference of intentions andfeelings specific to others and may prevent the assimilation of selfto others’ states promoted through lower-level spontaneous reso-nance or contagion effects [11,17]. Accordingly, a functional mag-netic resonance imaging study demonstrated that the implicitshared representation and the explicit mental states attributionare regulated by 2 sets of brain regions simultaneously activatedwhen one is accurately evaluating the emotional state of someoneelse [39]. Furthermore, additional factors including context, per-sonality traits, knowledge, interpersonal variables, and empathic

Elsevier B.V. All rights reserved.

ization of vicarious pain responses: Effects of pain cues and empathy traits.039

2 E. Vachon-Presseau et al. / PAIN�

xxx (2011) xxx–xxx

concern may affect different aspects of the complex socioaffectiveresponses to vicarious pain [12,14,21,22].

While most studies examined the effect of resonance and empa-thy mechanisms on brain activity, none has yet distinguished theirinfluence on self-pain processing across the different levels of theneuraxis. To the extent that perceiving pain in others may evokea negative emotion in the observer, one would predict a descend-ing facilitation of spinal sensorimotor processes [31,32] as well asincreased pain [9,13,24,30]. However, resonance for pain has alsobeen shown to affect the corticospinal motor systems in a somato-topically organized manner [1]. Thus, one may expect strongervicarious facilitation of low-level sensorimotor processes whensensory information is provided about the noxious stimulation it-self, and especially when the same somatotopic area is stimulated,as opposed to viewing only emotional-communicative informationconveyed by the facial expression of pain. In this study, images pre-senting nociceptive stimuli applied to the hand or foot, as well aspictures depicting facial expressions of pain, were presented tothe participants immediately before the application of an electricalstimulation eliciting a nociceptive flexion reflex (NFR) and acutepain. We expected that perceiving pain in others would activatecerebrospinal mechanisms facilitating spinal nociception andincreasing self-pain. Secondly, consistent with sensorimotor reso-nance processes, we posited that pictures illustrating noxiousstimulation to the foot and possibly the hand would elicit strongerfacilitation of the lower-limb NFR. Furthermore, we hypothesizedthat individual differences in the regulation of vicarious effects ob-served at different levels of the pain processing pathways wouldreflect the individual’s empathic ability.

2. Method

2.1. Participants

Fifty-nine healthy volunteers with no history of chronic pain,psychiatric disorders, or neurological problems participated in pre-testing sessions. Forty-eight of them were invited to take part in theexperiment based on the reliability of their NFR response and theirtolerance to the pain induced at the target stimulus intensity, as de-scribed below (24 males and 24 females, mean age = 24 ± 6 years).Out of these, 5 participants were excluded from data analyses: onebecause one electrode fell off during the experiment, another be-cause more than half of the trials failed to show a reflex, suggestingan important habituation to the stimulation; and 3 more becausethey did not perceive a painful expression in the images of the facecategory (mean scores of 0, 0, and 1.1 on the pain perception scaleranging from 0 to 100). A separate group of 15 participants assessedthe emotional valence and the arousal of the 90 selected pictures ona computerized adaptation of the Self-Assessment Manikin [20]. Allparticipants provided signed informed consent, and the study wasapproved by the Research Ethics Committee of the Institut univers-itaire de gériatrie de Montréal.

2.2. Apparatus

A computer running the E-Prime 1.1 software (Psychology Soft-ware Tools, Inc, Sharpsburg, PA, USA) controlled the presentationof the pictures, the administration of the electrical stimulations,and the visual analogue scales used by the participants for ratings.The screen was positioned at approximately 0.7 m from the partic-ipant. Electrical stimulations were produced by a Grass stimulator(Model S48 with optical isolation; Grass Technologies, WestWarwick, RI) and delivered through a pair of 1-cm2 bipolar stimu-lating electrodes. An optical isolation unit was used to prevent theelectrical stimulus from exceeding a safe intensity. Physiological

Please cite this article in press as: Vachon-Presseau E et al. The multilevel organon spinal nociception and acute pain. PAIN

�(2011), doi:10.1016/j.pain.2011.02

signals were amplified, filtered, and sampled at 1000 Hz with aBIOPAC MP150 amplifier; data were acquired and stored for off-line analysis on a computer running AcqKnowledge 4.0. (BIOPACSystems, Inc, Goleta, CA, USA).

The NFR was elicited by stimulating the sural nerve with a bipo-lar surface electrode placed at its retromalleolar site. The stimula-tions consisted of a volley of 10 1-ms square-wave pulsesadministered during 30 ms. Muscle reflex activity was recordedfrom the brevis head of the left biceps femoris muscle above thepopliteal fossa using 2 recording electrodes placed on the previ-ously cleaned, shaved (if necessary), and abraded skin in order toreach an impedance of <10 kX. A third electrode was used as aground and placed by the medial side of the tibial tuberosity.

2.3. Visual stimuli displaying nociceptive cues and pain expression

Vicarious pain stimuli consisted of 3 categories of static picturesdisplaying nociceptive stimuli applied to the hand (n = 15) or thefoot (n = 15) (ie, providing sensory information about the noxiousstimulus; taken and adapted from [18]) and facial expressions ofpain by actors (n = 15) (ie, providing emotional information aboutthe affective pain response; taken and adapted from [34]). Partici-pants of this study confirmed that the levels of pain displayed inthe images were comparable across categories (see below). Match-ing neutral pictures of a hand (n = 15) and foot (n = 15) displaying asimilar situation without nociception and neutral facial expres-sions of the same actors (n = 15) were also included as controlconditions.

The selected stimuli were further described in terms of theiremotional dimensions of valence and arousal in a separate groupof participants using a computerized adaptation of the Self-Assessment Manikin, which includes 5 pictograms illustrating suc-cessive levels of valence or arousal. The pictograms and each ofthe 4 intermediate steps between successive pictograms providedthe 9 possible rating levels on each scale [20]. The participants wereexplained the notion of valence and arousal and were asked to rateeach stimulus by selecting one of 9 levels of valence (1 = mostunpleasant to 9 = most pleasant) and arousal (1 = most relaxing to9 = most arousing). As in the main experiment, each picture waspresented for 1 second, immediately followed by the valence andarousal scales displayed on the computer screen. This confirmedthat all 3 categories of pain-evoking pictures generated negativevalence (Mean ± SD: 3.04 ± .85 for hand, 3.00 ± 1.12 for foot, and3.62 ± 1.10 for facial expression), with comparable increases inarousal relative to the corresponding neutral condition (pain-evoking pictures � Neutral: 2.88 ± 1.51 for hand, 2.95 ± 2.18 forfoot, and 2.05 ± 1.49 for facial expression; F(1,62) = 2.43, P = 0.12).

2.4. Procedure

Participants were first given an overview of the experiment andwritten informed consent was obtained. After completing ques-tionnaires (see below), the electrodes were attached while the par-ticipants were sitting comfortably in the reclined chair. The reflexthreshold was determined for each participant using the staircasemethod involving several series of increasing and decreasing stim-ulations [37]. Stimulations were delivered to the sural nerve atintervals of 6 seconds, while the voltage was increased in 0.5-Vsteps until a reflex was detected, with a minimum of 4 stimuli ap-plied at each intensity. The NFR threshold was defined as the low-est stimulus intensity evoking a stable response (ie, a clearlydetectable response in at least 80% of the trials). The stimulusintensity was then increased to 135% of the NFR threshold or untilparticipants rated the shock-pain as 60 on the 0–100 pain intensityscale. This intensity was referred to as supra-threshold intensity(HIGH) and corresponded to a mean shock intensity of 22.18 mA

ization of vicarious pain responses: Effects of pain cues and empathy traits.039

E. Vachon-Presseau et al. / PAIN�

xxx (2011) xxx–xxx 3

(± 3.51). A second intensity corresponding to 50% of the supra-threshold intensity (LOW) was also used in half of the trials tointroduce some uncertainty regarding the occurrence of painfulstimulation and in order to reduce the risk of habituation. Thisstimulus did not produce reliable NFR responses, but participantstypically reported mild pain at that intensity level.

Each trial started with a picture presented for 1000 ms andimmediately followed by an electrical stimulus varying pseudo-randomly between the HIGH and LOW intensity. Participants werenot told that only 2 electrical intensities were used and were given14 seconds to rate each stimulus successively on 2 different pain-rating scales as described below [29]. The overall length of eachtrial varied between 20 and 23 seconds, with each scale presentedfor 7 seconds, and the intertrial interval varied randomly from 5 to8 seconds.

The pictures were shown in one of 2 different orders, counter-balanced across participants (2 inverted sequences; ie, startingby the first or the last image of a predetermined pseudo-randomorder). For each of those sequences, the association between a spe-cific picture and the intensity of the electric stimulation (HIGH/LOW) was inverted in half of the participants. At the end of theexperimental phase, the participants were shown the same 90images and were asked to evaluate the pain induced by the situa-tion displayed in the picture or the pain expression perceived inthe face (scale ranging from 0 to 100).

2.5. Measures

2.5.1. Nociceptive flexion reflex (NFR)The magnitude of the NFR was assessed using the integral of the

rectified electromyogram signal between 90 and 180 ms poststim-ulation. The NFR was standardized within-individual using z-scoretransformation to account for individual differences in the absoluteNFR amplitude, gain, and variability (see supplementary materialfor unstandardized scores). The LOW intensity stimulations werenot intended to elicit an NFR and were not included in this analysis.The NFR z-scores were averaged across trials within each categoryof pictures and pain level (Category: hand, foot, or face; Level:painful or neutral) for each participant. The skin conductance re-sponse was also recorded, but the data are not presented becausethis experimental paradigm did not allow for a clear separation be-tween responses evoked by the shock from those possibly inducedby the pain pictures.

2.5.2. Shock-pain intensity and unpleasantnessParticipants were explained the difference between pain inten-

sity and unpleasantness before the experiment [29] and wereshown visual-numerical (0–100) pain-rating scales displayedhorizontally on the computer screen with verbal anchors at the left(‘‘0 – no pain/not at all unpleasant’’) and right extremities (‘‘100 –extremely intense/unpleasant’’). After each electrical stimulus,subjects provided their ratings by moving a cursor smoothly alongthe scales using 2 digital response keys. The ratings of shock-painintensity and unpleasantness were standardized within-individualusing z-scores transformation applied separately for HIGH- andLOW-intensity stimuli. The standardized intensity and unpleasant-ness ratings were then averaged across each category of pictures(hand, foot, or face) with the same pain-evoking level (painful orneutral).

2.5.3. Assessment of psychological factorsThe Empathy Quotient (EQ; ([3]) and the Interpersonal Reactivity

Index (IRI; [8]) were used to assess trait empathy. The EQ is a self-report questionnaire of 60 items that are summed to an overallscore, while the IRI is a 28-item questionnaire measuring 4 factors.For the purpose of the present study, only the empathic concern

Please cite this article in press as: Vachon-Presseau E et al. The multilevel organon spinal nociception and acute pain. PAIN

�(2011), doi:10.1016/j.pain.2011.02

(EC), the perspective-taking (PT), and the personal distress (PD)factors from the IRI were included in our analyses. Because the 4scores of interest (EQ, EC, PT, and PD) showed high correlation be-tween them, a principal component analysis was used for datareduction (as in [2]). The 4 scores loaded onto one factor, referredto as the global ‘‘empathy score,’’ which explained 62% of the vari-ance and positively correlated with the EQ (r = .86), the EC (r = .89),and the PT (r = .89), and negatively correlated with the PD(r = �.42).

2.6. Statistical analysis

Shock-pain ratings and NFR were analyzed using repeated-measures analyses of variance (ANOVAs) with the followingwithin-subject factors: picture Category (hand, foot, and face),Pain-evoking Picture (painful and neutral) and Stimulation Inten-sity (HIGH and LOW; except for NFR where only the HIGH condi-tion was analyzed). To avoid biasing statistical estimates whenthe assumption of sphericity was not met, the degrees of freedomwere corrected using the Greenhouse-Geisser e. A P 6 0.05 wasconsidered statistically significant and Bonferroni correction wasapplied to correct for multiple comparisons in post hoc contrasts.A pain modulation index was calculated for each dependent vari-able (shock-pain ratings and NFR) by subtracting, for each partici-pant, the average shock-evoked response following the pain-related images from the response to the neutral images. Correla-tion analyses (Pearson r) and a multiple regression were then per-formed specifically to test the relation across pain modulationindices and between pain modulation indices and the empathy.All analyses were performed with SPSS 16.0 (SPSS Inc, Chicago,IL, USA).

3. Results

3.1. Pain perceived in the images

The mean scores (M ± SD) of pain perceived by the 43 partici-pants were 62.4 ± 20.4 for the hand, 63.0 ± 20.8 for the foot, and58.0 ± 22.1 for the face categories in the pain-evoking picturesand 5.9 ± 9.0 for the hand, 5.9 ± 9.3 for the foot, and 7.8 ± 16.2 forthe face categories in the neutral pictures. An ANOVA on the scoresof the pain perceived revealed significant differences between neu-tral and pain-evoking pictures, with pain pictures displaying mod-erate to strong pain [F(1,41) = 333.45, P < 0.001]. There was nosignificant effect of category or interaction between pain-evokingpictures and category (P’s > 0.2).

3.2. Modulation of shock-pain intensity and unpleasantness

The mean shock-pain intensity and unpleasantness ratings forHIGH- and LOW-intensity stimuli administered after a neutral orpain-evoking picture of a hand, foot, or face are reported in Table 1.The corresponding standardized mean of the pain intensity andpain unpleasantness ratings are illustrated in Fig. 1. The within-subject ANOVA on standardized shock-pain intensity ratings re-vealed that witnessing painful images of hand, foot, or face en-hanced shock-pain intensity at both HIGH and LOW intensities(main effect of pain-evoking picture [F(1,42) = 8.18, P = 0.007]).No other effect or interaction reached significance. Changes inshock-pain were more apparent in the absolute ratings ofunpleasantness and were confirmed by the ANOVA performed onstandardized ratings, which revealed a significant interaction ofPain-evoking Picture with Category [F(1.88,79.02) = 5.55, P =0.006]. The results were further examined using the change scorecorresponding to the difference between the pain-related andneutral picture within each category. Paired t test revealed that

ization of vicarious pain responses: Effects of pain cues and empathy traits.039

Table 1Mean (SD) shock-pain intensity and unpleasantness ratings for HIGH- and LOW-intensity shocks applied after images of a hand, a foot, or a face in the neutral or painful condition.

Shock intensity Pain rating Picture of hand Picture of foot Picture of face

Neutral Pain Neutral Pain Neutral Pain

HIGH Pain intensity 55.4 (± 20.0) 56.0 (± 19.9) 55.1 (± 19.5) 56.2 (± 19.9) 54.2 (± 20.1) 55.5 (± 20.4)Unpleasantness 49.1 (± 18.5) 55.6 (± 19.1) 48.0 (± 18.7) 55.2 (± 19.6) 48.3 (± 20.4) 51.9 (± 19.6)

LOW Pain intensity 32.2 (± 17.0) 33.6 (± 17.0) 30.8 (± 16.4) 33.7 (± 16.8) 31.1 (± 16.7) 33.3 (± 17.1)Unpleasantness 30.0 (± 16.3) 37.0 (± 18.2) 28.4 (± 16.0) 36.9 (± 18.8) 27.7 (± 16.2) 33.5 (± 18.3)

Fig. 1. Standardized shock-pain intensity (A, B) and unpleasantness (C, D) ratings to HIGH (A, C) or LOW (B, D) painful electrical stimulations administered after thepresentation of images depicting hand, foot, or face in pain-evoking or neutral situations. ⁄P 6 0.05; ⁄⁄⁄P < 0.001.

4 E. Vachon-Presseau et al. / PAIN�

xxx (2011) xxx–xxx

Please cite this article in press as: Vachon-Presseau E et al. The multilevel organization of vicarious pain responses: Effects of pain cues and empathy traitson spinal nociception and acute pain. PAIN

�(2011), doi:10.1016/j.pain.2011.02.039

E. Vachon-Presseau et al. / PAIN�

xxx (2011) xxx–xxx 5

the mean standardized change scores in pain unpleasantness in-duced by the pain-evoking pictures (Vs neutral) were significantlylarger for pictures of foot (M ± SD = +0.51 ± 0.51) and hand(M ± SD = +0.46 ± 0.54) than pictures of face (M ± SD = +0.35 ±0.56); P-corrected = 0.028 and 0.048, respectively.

A third ANOVA including the intensity and unpleasantnesschange scores revealed a main effect of Pain dimension [F(1,42) =10.09, P = 0.001], reflecting a larger modulation of the unpleasant-ness dimension, and a marginally significant interaction ofCategory � Pain dimension [F(1.85,77.60) = 3.08, P = 0.054], reflect-ing larger differences between categories in the unpleasantnessdimension.

3.3. Modulation of the nociceptive flexion reflex (NFR)

Fig. 2 displays the standardized NFR following the administra-tion of HIGH stimulations. The results show an interaction ofPain-evoking Picture with Category, [F(1.91,80.02) = 4.18,P = 0.02]. Tests of simple effect of Pain-evoking Picture in each cat-egory revealed that pictures of a foot showed a highly significantfacilitation of the NFR (P < 0.001), pictures of a hand showed a mar-ginally significant effect (P = 0.08), while the face failed to showany consistent modulation (P = 0.90). Paired t tests on the stan-dardized NFR change scores indicated that the presentation of pic-tures of the foot category produced stronger facilitation than facialpain expression (corrected P = 0.006), but failed to differ signifi-cantly from pictures of the hand category (corrected P = 0.32).

3.4. The relation between vicarious pain modulation and trait empathy

The influence of trait empathy on vicarious pain modulationwas examined using Pearson correlations. As illustrated in Fig. 3,individual differences in the global empathy score positively corre-lated with the modulation of the NFR change score (r = .41;P = 0.007) and negatively correlated with the shock-pain intensity

Fig. 2. Standardized nociceptive flexion reflex (NFR) when HIGH painful electricalstimulations were administered after the presentation of images depicting hand,foot, or face in pain-evoking or neutral situations. The NFR was facilitated onlywhen images of a foot (⁄⁄⁄P < 0.001) or a hand (borderline P = 0.08) were presented.

Please cite this article in press as: Vachon-Presseau E et al. The multilevel organon spinal nociception and acute pain. PAIN

�(2011), doi:10.1016/j.pain.2011.02

change score at HIGH-intensity stimulations (r = �.29; borderlineP = 0.056). No significant correlation was found between the empa-thy score and the shock-pain unpleasantness. The influence ofempathy on vicarious pain modulation was further investigatedby computing a stepwise multiple regression to examine the jointpredictive value of the NFR facilitation and of the decrease inshock-pain intensity on the empathy score. The NFR change scorecontributed significant variance in the prediction of the empathyscore [b = +0.42, R2 = 0.17, F(1,41) = 8.65, P = 0.007]. The shock-painintensity change score at HIGH intensity also contributed signifi-cantly to the predicted empathy score [b = �0.31, R2

change = .10,F(1,41) = 5.55, P = 0.027]. These findings indicate that the shock-pain intensity change score at HIGH intensity explained 10% of un-ique variance, adding to the 17% already predicted by the NFRchange score of the participants’ empathy score. These resultsdemonstrate that participants with vicarious facilitation in low-level pain processing (NFR) combined with a downregulation ofself-pain perception scored higher on empathy scales.

3.5. Relation between vicarious changes in NFR, shock-pain intensity,and unpleasantness

No significant correlations were observed between shock-painintensity, unpleasantness, and the NFR when testing those correla-tions between pain measures within each category of picture (allP’s > 0.05). This implies that vicarious effects observed on the sep-arate self-pain-related measures were largely independent and dri-ven by separate mechanisms. This finding is not surprising, sinceother studies underpinned the multiplicity of central mechanismsassociated with the modulation of spinal nociception and acutepain perception by emotions [32] and counterirritation [26], orwith baseline individual differences and spontaneous fluctuationsin motor, autonomic, and perceptual responses [27].

4. Discussion

The present study demonstrates a vicarious priming of thenociceptive system consistent with self-other resonance pro-cesses proposed in the shared-representation model of empathy[18,25,28,35]. This facilitation of pain responses is found at multi-ple levels of the neuraxis but is influenced by the nature of theinformation provided about the pain of others. More robust vicar-ious facilitation was found in response to images of noxious stim-ulation to the foot and hand compared to facial expression of pain.The results further reveal that the direction and magnitude of thevicarious modulation of self-pain are determined by the individ-ual’s empathic trait. Modulation of spinal and supraspinal pain re-sponses are discussed in relation to the shared-representationmodel of pain empathy and their up- or downregulation by high-er-order empathic processes.

4.1. Vicarious pain affects the multiple stages of self-pain processing

Starting with the higher-order pain processing systems, theresults of the current study show that witnessing others’ painincreases the intensity and unpleasantness of a nociceptive stimu-lation (shock-pain) received by the observer. Our results suggestthat witnessing others’ pain most strongly modulates the unpleas-antness dimension of the participants’ shock-pain. This distinctionhas also been established by previous studies reporting that in-duced emotional changes selectively modulate the pain unpleas-antness of the participants [23]. In addition, the results also showthat vicarious pain modulates nociceptive responses at lowerlevels of the neuraxis. Hence, cues signaling pain in others donot solely involve cortical circuits, but also activate descending

ization of vicarious pain responses: Effects of pain cues and empathy traits.039

Fig. 3. Relationship between the empathy score and the (A) shock-pain intensity (r = �.29), and (B) NFR (r = .41) change scores (change score = pain evoking picturecondition � neutral picture condition).

6 E. Vachon-Presseau et al. / PAIN�

xxx (2011) xxx–xxx

facilitation processes that primed the nociceptive responses at thespinal level. From a functional point of view, these findings suggestthat perceiving pain in others activates defensive brain systemspromoting stronger escape or avoidance responses.

4.2. Vicarious pain responses depend on the sensory and affective cuessignaling pain in others

In this study, the visual cues signaling pain in others includedstimuli providing direct information about the nature, location,and context of the noxious event (sensory information) or informa-tion about the other’s affective reaction to pain (affective informa-tion). Results demonstrated that this factor was critical indetermining the magnitude of the vicarious modulation of self-pain. We observed that witnessing body limb threatened by pain-ful agents elicited facilitation in spinal nociception. This effect wasspecific to pictures depicting sensory features of the pain and wasnot observed when facial pain expressions were presented. Thissupports the view of an organization including a somatic reso-nance for others’ pain that facilitates the withdrawal response toprevent potential tissue damage to the self. These results are inaccordance with the results of a transcranial magnetic stimulationstudy by Avenanti et al. [1] demonstrating that the sensory aspectsof someone else’s pain are topographically mapped on one’s ownbody representation, resulting in a localized modulation of thecerebrospinal motor excitability. Our results, however, do notshow specific facilitation to images of a foot (the limb stimulatedto elicit the NFR), but rather support descending facilitation gener-alized to pictures depicting the pain sensory information on thehand or foot.

Pain-evoking situations depicting sensory information also elic-ited robust increases in shock-pain unpleasantness. These differ-ences may be attributed to facial expressions generating arelatively weaker threat-related response rather than a specific re-sponse to the origin of the pain that can be mirrored by its sensoryfeatures. While the nociceptive defensive response (ie, NFR) wasfacilitated more clearly by images explicitly depicting pain-evoking sensory information, the unpleasantness of acute self-painwas increased by all pain-related images and may be more suscep-tible to weaker and less specific emotion-related processes orhigher-order pain-priming mechanisms triggered by witnessingpain-threatening situations from either sensory or emotionalperspectives.

Please cite this article in press as: Vachon-Presseau E et al. The multilevel organon spinal nociception and acute pain. PAIN

�(2011), doi:10.1016/j.pain.2011.02

4.3. Vicarious pain responses are regulated by empathy

The present results demonstrate a low-level sensorimotor reso-nance (NFR) that is automatic and may contribute to the affectivecoupling between the self and other based on shared representa-tions [11]. This sensorimotor coupling was stronger in participantsscoring higher on the global empathy score, as demonstrated bythe correlation analyses. This is consistent with a bottom-up con-tribution of low-level resonance to empathy. These results arecomplementing the observations of several recent studies lookingat the effects of empathy on brain activity. For instance, onetranscranial magnetic stimulation study showed that the perspec-tive-taking abilities and the personal distress of the participantsobserving needles penetrating a model’s hand were positively cor-related with the modulation of corticospinal activity [2]. Along thisline, patients with congenital insensitivity to pain (no experienceof self-pain) showed deficits in assessing pain in others only whenperceiving sensory information (body parts), while their capacitiesto infer pain from facial expressions remained preserved [7]. Thespecificity of the deficit seems to be generated by abnormallylow activity in brain regions involved in the perspective-takingabilities necessary to map the basic sensory aspect of someoneelse’s pain [6]. It is therefore possible, as previously proposed, thatempathy traits may foster the linkage of the mirror neuron systemwith the emotional response contributing to the effective shapingof sensorimotor mirroring [19] and facilitate low-level vicariouspain responses.

The present study also suggests that higher-level vicarious pro-cesses measured by changes in self-pain intensity perception werenegatively correlated with empathy score. This relation betweenhigher empathy scores and the reduced vicarious hyperalgesiamay reflect the downregulation of automatic, self-protective re-sponses in highly empathic individuals. This finding explains theoverall very small increases found in the shock-pain intensity (Ta-ble 1) since hypoalgesia was observed in several highly empathicparticipants. This is consistent with hierarchical models of empa-thy [11,28] in which the lower levels of empathy reflect the auto-matic sharing of emotional states that needs to be self-regulated toallow for the expression of empathic concern toward others.Accordingly, recent evidence suggests that unconscious processingof affective value does not facilitate attention toward a prosocialempathic response but rather facilitates attention toward thethreat value of pain [38]. These results challenge the conception

ization of vicarious pain responses: Effects of pain cues and empathy traits.039

E. Vachon-Presseau et al. / PAIN�

xxx (2011) xxx–xxx 7

that witnessing pain spontaneously generates an approach re-sponse and concern for others. As such, understanding someoneelse’s pain would require the tuning-down of defensive responsesand negative emotions before an empathic response could occur[10,36]. Hence, this approach/avoidance paradox [15] seems tobe overcome by higher-order processes that suppress self-painfacilitation at higher-order stages of processing to preventemotional contagion or personal distress. Altogether, our resultssupport the coexistence of at least 2 mechanisms in which a sen-sory resonance primes low level of defensive responses (ie, NFR)along with a higher level of self-regulated responses that wouldbe necessary to suppress avoidance responses to signs of painand suffering in others.

This study demonstrates an effect of vicarious pain on severallevels of pain processing that are differently influenced by individ-ual differences in empathy. These novel findings strongly supportthe view that empathy for pain is a complex process involvingseveral levels of facilitatory and inhibitory self-regulation mecha-nisms affecting the multiple stages of nociceptive processing inthe central nervous system. Correlation findings on the effect ofempathy should, of course, be replicated in future studies, possiblyby contrasting individuals preselected based on their high vs lowempathy level. Further research should also test the hypothesisthat highly empathic individuals demonstrate higher self-regulation abilities while witnessing the pain of others. Monitoringboth the self-pain and the perception of others’ pain on a trial-by-trial basis would also permit researchers to determine if thehypothesized downregulation of self-pain in highly empathic indi-viduals improves empathic accuracy. This research may providesome insight on pain communication processes and help improvepain assessment methods while, it is hoped, reducing the poten-tially deleterious impact on health care professionals of being ex-posed chronically to cues signaling pain in others.

Conflict of interest statement

Etienne Vachon-Presseau, Marc-Olivier Martel, Mathieu Roy,Etienne Caron, Philip L. Jackson, and Pierre Rainville have no finan-cial or other relationships that might lead to a conflict of interest.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.pain.2011.02.039.

References

[1] Avenanti A, Bueti D, Galati G, Aglioti SM. Transcranial magnetic stimulationhighlights the sensorimotor side of empathy for pain. Nat Neurosci2005;8:955–60.

[2] Avenanti A, Minio-Paluello I, Bufalari I, Aglioti SM. The pain of a model in thepersonality of an onlooker: influence of state-reactivity and personality traitson embodied empathy for pain. Neuroimage 2009;44:275–83.

[3] Baron-Cohen S, Wheelwright S. The empathy quotient: an investigation ofadults with Asperger syndrome or high functioning autism, and normal sexdifferences. J Autism Dev Disord 2004;34:163–75.

[4] Botvinick M, Jha AP, Bylsma LM, Fabian SA, Solomon PE, Prkachin KM. Viewingfacial expressions of pain engages cortical areas involved in the directexperience of pain. Neuroimage 2005;25:312–9.

[5] Bufalari I, Aprile T, Avenanti A, Di Russo F, Aglioti SM. Empathy for pain andtouch in the human somatosensory cortex. Cereb Cortex 2007;17:2553–61.

[6] Danziger N, Faillenot I, Peyron R. Can we share a pain we never felt? Neuralcorrelates of empathy in patients with congenital insensitivity to pain. Neuron2009;61:203–12.

[7] Danziger N, Prkachin KM, Willer JC. Is pain the price of empathy? Theperception of others’ pain in patients with congenital insensitivity to pain.Brain 2006;129:2494–507.

Please cite this article in press as: Vachon-Presseau E et al. The multilevel organon spinal nociception and acute pain. PAIN

�(2011), doi:10.1016/j.pain.2011.02

[8] Davis M. A multidimensional approach to individual differences in empathy.JSAS Catalog Selected Documents Psychol 1980;10:85.

[9] de Wied M, Verbaten MN. Affective pictures processing, attention, and paintolerance. Pain 2001;90:163–72.

[10] Decety J. Empathy, sympathy and the perception of pain. Pain2009;145:365–6.

[11] Decety J, Jackson PL. The functional architecture of human empathy. BehavCogn Neurosci Rev 2004;3:71–100.

[12] Decety J, Yang CY, Cheng Y. Physicians down-regulate their pain empathyresponse: an event-related brain potential study. Neuroimage 2010;50:1676–82.

[13] Godinho F, Magnin M, Frot M, Perchet C, Garcia-Larrea L. Emotionalmodulation of pain: is it the sensation or what we recall? J Neurosci2006;26:11454–61.

[14] Goubert L, Craig KD, Vervoort T, Morley S, Sullivan MJ, de C Williams AC, CanoA, Crombez G. Facing others in pain: the effects of empathy. Pain2005;118:285–8.

[15] Goubert L, Vervoort T, Crombez G. Pain demands attention from others: theapproach/avoidance paradox. Pain 2009;143:5–6.

[16] Hutchison WD, Davis KD, Lozano AM, Tasker RR, Dostrovsky JO. Pain-relatedneurons in the human cingulate cortex. Nat Neurosci 1999;2:403–5.

[17] Jackson PL, Brunet E, Meltzoff AN, Decety J. Empathy examined through theneural mechanisms involved in imagining how I feel versus how you feel pain.Neuropsychologia 2006;44:752–61.

[18] Jackson PL, Meltzoff AN, Decety J. How do we perceive the pain of others? Awindow into the neural processes involved in empathy. Neuroimage2005;24:771–9.

[19] Keysers C, Gazzola V. Towards a unifying neural theory of social cognition.Prog Brain Res 2006;156:379–401.

[20] Lang P. Behavioral treatment and bio-behavioral assessment. In: Sidowski JB,Williams TA, editors. Technology in mental health care delivery systems. NewJersey: Norwood; 1980. 119–37.

[21] Langford DJ, Crager SE, Shehzad Z, Smith SB, Sotocinal SG, Levenstadt JS,Chanda ML, Levitin DJ, Mogil JS. Social modulation of pain as evidence forempathy in mice. Science 2006;312:1967–70.

[22] Loggia ML, Mogil JS, Bushnell MC. Empathy hurts: compassion for anotherincreases both sensory and affective components of pain perception. Pain2008;136:168–76.

[23] Loggia ML, Mogil JS, Bushnell MC. Experimentally induced mood changespreferentially affect pain unpleasantness. J Pain 2008;9:784–91.

[24] Meagher MW, Arnau RC, Rhudy JL. Pain and emotion: effects of affectivepicture modulation. Psychosom Med 2001;63:79–90.

[25] Morrison I, Lloyd D, di Pellegrino G, Roberts N. Vicarious responses to pain inanterior cingulate cortex: is empathy a multisensory issue? Cogn Affect BehavNeurosci 2004;4:270–8.

[26] Piche M, Arsenault M, Rainville P. Cerebral and cerebrospinal processesunderlying counterirritation analgesia. J Neurosci 2009;29:14236–46.

[27] Piche M, Arsenault M, Rainville P. Dissection of perceptual, motor andautonomic components of brain activity evoked by noxious stimulation. Pain2010;149:453–62.

[28] Preston SD, de Waal FB. Empathy: its ultimate and proximate bases. BehavBrain Sci 2002;25:1–20 [discussion 20–71].

[29] Rainville P, Feine JS, Bushnell MC, Duncan GH. A psychophysical comparison ofsensory and affective responses to four modalities of experimental pain.Somatosens Mot Res 1992;9:265–77.

[30] Rhudy JL, Bartley EJ, Williams AE, McCabe KM, Chandler MC, Russell JL, Kerr KL.Are there sex differences in affective modulation of spinal nociception andpain? J Pain 2010;11:1429–41.

[31] Rhudy JL, Williams AE, McCabe KM, Nguyen MA, Rambo P. Affectivemodulation of nociception at spinal and supraspinal levels. Psycho-physiology 2005;42:579–87.

[32] Roy M, Piche M, Chen JI, Peretz I, Rainville P. Cerebral and spinal modulation ofpain by emotions. Proc Natl Acad Sci USA 2009;106:20900–5.

[33] Saarela MV, Hlushchuk Y, Williams AC, Schurmann M, Kalso E, Hari R. Thecompassionate brain: humans detect intensity of pain from another’s face.Cereb Cortex 2007;17:230–7.

[34] Simon D, Craig KD, Miltner WH, Rainville P. Brain responses to dynamic facialexpressions of pain. Pain 2006;126:309–18.

[35] Singer T, Seymour B, O’Doherty J, Kaube H, Dolan RJ, Frith CD. Empathy for paininvolves the affective but not sensory components of pain. Science2004;303:1157–62.

[36] van Rysewyk S. The integration of emotion and reason in caregiver painassessment. J Pain 2010;11:804–5.

[37] Willer JC. Comparative study of perceived pain and nociceptive flexion reflexin man. Pain 1977;3:69–80.

[38] Yamada M, Decety J. Unconscious affective processing and empathy: aninvestigation of subliminal priming on the detection of painful facialexpressions. Pain 2009;143:71–5.

[39] Zaki J, Weber J, Bolger N, Ochsner K. The neural bases of empathic accuracy.Proc Natl Acad Sci USA 2009;106:11382–7.

ization of vicarious pain responses: Effects of pain cues and empathy traits.039