Psychiatry Research: Neuroimaging 224 (2014) 341–348

Contents lists available at ScienceDirect

Psychiatry Research: Neuroimaging journal homepage: www.elsevier.com/locate/psychresns

Feeling but not caring: Empathic alteration in narcissistic men with high psychopathic traits Louis-Alexandre Marcoux a,b, Pierre-Emmanuel Michon c, Sophie Lemelin a,b, Julien A. Voisin c, Etienne Vachon-Presseau a,c, Philip L. Jackson a,b,c,n a

École de psychologie, Université Laval, Québec, Canada Centre de recherche de l’Institut universitaire en santé mentale de Québec, Québec, Canada c Centre interdisciplinaire de recherche en réadaptation et intégration sociale, Québec, Canada b

art ic l e i nf o

a b s t r a c t

Article history: Received 1 November 2013 Received in revised form 13 September 2014 Accepted 2 October 2014 Available online 13 October 2014

Psychopathy is a personality disorder characterized by specific interpersonal–affective deficits and social deviance often marked by reduced empathy and decreased affective response to the suffering of others. However, recent findings in community samples suggest that the somatosensory resonance to other's pain measured with electroencephalography (EEG) is increased by psychopathic traits. This study aimed at comparing both the response to physical pain and the observation of pain being inflicted to another person in individuals with clinically significant psychopathic traits, namely patients with severe narcissistic personality disorder (NPD, n ¼11), and community controls (CC, n ¼ 13). The gating of somatosensory responses to a tactile steady-state stimulation (25 Hz) during the observation of painevoking and non-painful visual stimuli of hands was measured using EEG. Pain thresholds were assessed with a quantitative sensory testing (QST) battery. NPD compared with CC subjects showed similar thermal pain thresholds, but significantly higher pain pressure thresholds (PPT). Significantly greater somatosensory gating (SG) during the anticipation and the observation of pain in others was observed in NPD compared with CC subjects, but this difference was not associated with differences in self-pain perception. SG to pain observation was positively correlated with the Impulsivity–Egocentricity (IE) dimension of psychopathy. These findings demonstrated a stronger somatosensory resonance in the high psychopathic trait NPD group that suggests an increased somatic representation of observed pain despite lower dispositional empathy. & 2014 Elsevier Ireland Ltd. All rights reserved.

Keywords: Pain perception Psychopathic traits Narcissistic personality disorder Somatosensory resonance Empathy

1. Introduction Psychopathy is a severe and complex personality disorder marked by affective and interpersonal alterations including a lack of guilt and empathy, a failure to form deep relationships, disregard for other people's well-being, antisocial behaviors, and weak impulse control (Hare, 1991; Lilienfeld, 1994). It has been proposed that psychopathy comprises interpersonal–affective traits (Factor 1) and antisocial deviance traits (Factor 2) (Hare, 1991, 2003; Cooke and Michie, 2001). Theoretically and empirically, interpersonal–affective deficits have received significant attention as they capture the core features of psychopathy (Cleckley, 1941; Karpman, 1941). Empathy disturbances constitute an important part of these interpersonal–affective deficits (Hare, 1991, 2003). n Correspondence to: École de psychologie, Pavillon Félix-Antoine-Savard, 2325, rue des Bibliothèques, Université Laval Québec (Québec), Canada G1V 0A6. Tel.: þ 1 418 656 2131/5151. E-mail address: [email protected] (P.L. Jackson).

http://dx.doi.org/10.1016/j.pscychresns.2014.10.002 0925-4927/& 2014 Elsevier Ireland Ltd. All rights reserved.

Empathy is largely recognized as a multilevel construct, involving the interaction of lower and higher order processes (e.g., VachonPresseau et al., 2011) that predispose individuals to emotionally identify with other feelings and to share their affective experiences (Decety and Jackson, 2004; Kernberg, 2012). One proposed component of empathy is affective resonance, or the automatic activation in an observer of the affective and sensorimotor states perceived in another person. It reflects a capacity to share the feelings of individuals with whom we interact (Decety and Jackson, 2004; Lamm et al., 2011 for a review). Several studies have reported abnormal emotional reactions to arousing stimuli in psychopathic individuals (Kiehl et al., 2001; Müller et al., 2003), particularly during the observation of fearful facial expressions (Marsh et al., 2008), providing valuable insights about their emotional deficits and more specifically their empathy deficits. More recently, observation of pain in others has been used as a paradigm to investigate the patterns of neural activity elicited by arousing stimuli in various populations (e.g., Singer et al., 2004; Avenanti et al., 2005; Bufalari et al., 2007; Lamm et al., 2007;

342

L.-A. Marcoux et al. / Psychiatry Research: Neuroimaging 224 (2014) 341–348

Fecteau et al., 2008; Voisin et al., 2011; Marcoux et al., 2013). Despite the extended literature depicting the emotional deficits observed in psychopathic individuals, little is known about the neural bases of their reactions toward the pain of others. A first transcranial magnetic stimulation study (Fecteau et al., 2008) supported the empathy deficit hypothesis by showing that in a community sample the reduction of motor-evoked potentials induced by pain observation was stronger in individuals scoring higher on the Coldheartedness (C) scale of the Psychopathic Personality Inventory (PPI; Lilienfeld and Widows, 2005). A second study, also based on a community sample, has demonstrated stronger somatosensory resonance to others' pain in high-psychopathy individuals (Marcoux et al., 2013). These studies suggest a potential link between psychopathic traits and the resonance to other people's pain. While these findings are novel and interesting, somatosensory resonance to others' pain has not yet been studied in individuals showing pathological levels of psychopathic traits. From a clinical perspective, patients with severe narcissistic personality disorder constitute an interesting population as they present significant deficits on the interpersonal–affective component that is central to psychopathy (Blair, 2005; Ritter et al., 2011). Critically, previous studies have shown that patients with narcissistic personality disorder (NPD) show empathy deficits (e.g., Watson et al., 1984; Watson and Morris, 1991; Dimaggio et al., 2006; Ritter et al., 2011) and altered psychophysiological responses to emotionally evocative stimuli (Kelsey et al., 2001). However, no study has examined the relationship among psychopathic traits, empathy traits and empathy-related brain response to observed painful situations in a sample of individuals with high psychopathic traits such as severe narcissistic personality disorder (NPD) patients. The current study thus aimed to extend the investigation of psychopathy in non-incarcerated samples by focusing on empathy deficits, which represent the core interpersonal–affective traits of psychopathy found in pathological narcissism. Because of the similarity between the cortical signatures of self-pain and of perception of pain in others, in vivo pain thresholds were also tested as the relationship between self-pain and pain in others might differ in psychopathic individuals. Severe NPD patients with high psychopathic traits and community controls (CC) were recruited to perform a pain-observation task designed to assess the somatosensory resonance component of empathy for the pain of others using a steady-state somatosensory response paradigm (Voisin et al., 2011). Based on recent findings (Marcoux et al., 2013), we predicted that severe NPD patients would show a greater reduction of somatosensory response to steady-state stimulation (somatosensory gating (SG)) during pain observation, compared with CC participants. This somatosensory gating would also be expected to be positively correlated with individual psychopathic traits. The link between the electroencephalography (EEG) measures of resonance and selfreported empathy was investigated to confirm empathy alteration in patients with a diagnosis of severe narcissistic personality disorder.

2. Methods 2.1. Sample Twenty-seven adult Caucasian men (aged 21–50) took part in this study. The clinical group comprised 13 severe NPD patients recruited from an outpatient clinic of the Institut Universitaire en Santé Mentale de Québec specialized in the treatment of severe personality disorders. Diagnostic procedures for the clinical group included the following psychiatric and psychological assessments: a clinical interview adapted from the Structural Interview for Personality

Organization (Stern et al., 2010), the Psychopathy Checklist-Revised semistructured interview (PCL-R) (Hare, 2003), and medical and criminal records. Detailed clinical data for the final sample (n ¼ 11; two patients were excluded during data pre-processing for movement-related data noise across the records) are presented in Table 1. These NPD patients obtained scores ranging from 8 to 15 on Factor 1 of the PCL-R, indicating significant psychopathic traits. Fourteen agematched men were recruited as a control group (CC) via public advertisements. Exclusion criteria for both groups included any neurological or pain-related condition. A history of psychiatric disorder was also considered an exclusion criterion for the CC group. Written informed consent was obtained from each participant. The Ethics Committees of the centers involved (Institut de Réadaptation en Déficience Physique de Québec and Institut Universitaire en Santé Mentale de Québec) approved the study, and participants received a 40$ CAD compensation.

2.2. Questionnaires The Psychopathic Personality Inventory (PPI-R) (Lilienfeld and Widows, 2005) is a 154-item self-report measure designed to assess the core personality traits of psychopathy. The total score and three main factors derived from the eight subscales were included in the analyses: Impulsivity–Egocentricity (IE), Fearlessness-Dominance (FD), and Coldheartedness (C). The PPI-R shows appropriate internal consistency among community/college samples (Cronbach's α ¼ 0.78–0.92; Ray et al., 2013) and satisfactory convergent validity with other self-report measures of psychopathy (Lilienfeld and Widows, 2005). The Interpersonal Reactivity Index (IRI; Davis, 1980) is a 28-item measure developed as a self-report assessment of trait empathy. Each item is rated on a fivepoint Likert scale from 1 (does not describe me well) to 5 (describes me very well). In this study, the three subscales that Davis tied to either affective or cognitive aspects of empathy were used: Empathic Concern (EC), Personal Distress (PD) and Perspective Taking (PT). For the affective aspects, the EC subscale measures feelings of sympathy and compassion for others in distress, while the PD subscale measures self-oriented feelings of anxiety and distress in response to tense interpersonal situations. For the cognitive aspects, the PT subscale measures the tendency to adopt the psychological point of view of others.

2.3. Sensory and pain sensitivity testing A short quantitative sensory testing (QST) battery was used to investigate the processing of somatic and nociceptive stimulation perceived by participants. Five subtests were used: cool and warm detection thresholds (respectively CDT and WDT), cold and heat pain thresholds (CPT and HPT), and pressure pain threshold (PPT). Cool and warm detection thresholds, as well as cold and heat pain detection thresholds, were obtained with a 9-cm2 MEDOC Thermode contact probe (TSA Neuro-Sensory Analyzer; Medoc Ltd. Advanced Medical System, Israel) to induce a progressive warming or cooling sensation (32 1C, ramp at 11/s) on the inner forearm of participants. Detection thresholds measured the minimal temperature change that participants could detect, while pain thresholds measured the temperature needed to induce in participants a sensation of pain associated with either warm or cold stimulation. The pain pressure test measured with an algometer the force that needed to be applied on one's thumbnail to elicit a pain sensation.

Table 1 Clinical assessment.

1 2 3 4 5 6 7 8 9 10 11

Principal Dx

Comorbid Axis-II

PCL-R total

PCL-R F1

PCL-R F2

NPD NPD NPD NPD NPD NPD NPD NPD NPD NPD NPD

Paranoïd – ASPD ASPD and BPD ASPD traits BPD traits ASPD and BPD ASPD and BPD ASPD and BPD ASPD –

18 16 24 24 28 (22) 31 21 (23) (24) 23

11 9 11 8 15 (10) 11 8 (12) (10) 9

6 4 12 16 13 (12) 18 11 (11) (14) 14

traits

traits traits traits

PCL-R: Psychopathy Checklist Revised, F1: Factor 1 and F2: Factor 2. NPD: narcissistic personality disorder, ASPD: antisocial personality disorder, BPD: borderline personality disorder. Numbers inside brackets are approximated scores using patient file information.

L.-A. Marcoux et al. / Psychiatry Research: Neuroimaging 224 (2014) 341–348

343

2500ms

No Pain

or

750ms

Pain

250ms

Pain intensity

1500ms

0

10

3000ms

Neutral

Pain intensity 0

10

Fig. 1. Schematic of the experimental design depicting one trial. Pseudo-dynamic stimuli of Pain, Nopain and Neutral conditions with timing in ms correspond to the duration of each picture. A light repetitive stimulation at a frequency of 25 Hz was continuously applied to the palm of the right hand throughout data acquisition.

2.4. Visual stimuli The stimuli used in the pain-observation task consisted of 30 color pseudodynamic image series depicting hands of Caucasian male and female adults in three different conditions: Painful, Non-Painful, and Neutral situations (for examples of stimuli, see Fig. 1). Specifically, each stimulus was composed of three static pictures presented in a rapid sequence (750 þ250 þ 1500 ¼2500 ms) to create the illusion of movement (see Fig. 1). Different types of pain (mechanical or thermal) inflicted to the hands were displayed. The Non-Painful stimuli showed hands in visually similar situations as in the Pain condition but without the painful consequence (i.e., the third frame differed; e.g., a knife on the finger [Pain] vs. a knife on the board [No Pain]). Neutral stimuli showed hands in visually different situations devoid of any nociceptive elements (e.g., a hand grasping a set of keys or a tissue). This neutral condition was used to control for the anticipation effect that could be evoked by the presence of potentially nociceptive elements in the first images in both Painful and Non-Painful conditions. All pictures were edited to show male and female hands of the same size, from approximately the same distance, and oriented with a maximum angle of 451 from the perspective of the observer. 2.5. Tactile steady-state stimulation A tactile steady-state stimulation was used to elicit a specific brain response to a somatosensory stimulation, which could then be modulated by the observation of pain. Non-painful light repetitive (25-Hz) mechanical stimulations were continuously applied to the palm and inner side of fingers of the right hand using a custom-made vibrotactile stimulator developed at our laboratory (e.g., Marcoux et al., 2013).

to control the timing of the stimuli as well as the generation of events code. The experimental session consisted of six blocks of 30 trials lasting approximately 5 min each. Each trial began with a fixation cross (2500 ms), followed by a pseudodynamic image sequence (2500 ms), and a visual rating scale (3000 ms) ranging from 0 (no pain) to 10 (worst pain possible) (see Fig. 1). After each scenario, participants used the visual rating scale to orally evaluate the level of pain depicted in the stimulus, which was manually recorded by the experimenter. Participants were asked to refrain from blinking during the presentation of fixation crosses and stimuli, and to avoid performing head and jaw movements as much as possible. Ten practice trials were run before the experiment using other pictures than those selected for the test trials. Following the EEG session, participants completed the QST and self-report questionnaires, for approximately 30 min.

2.8. Statistical analyses 2.8.1. Behavioral data Differences in observed pain ratings between each condition were computed using a mixed design analysis of variance (ANOVA) with one within-subject factor: Condition (Pain, NoPain, Neutral) and one between-subject factor: Group (CC, NPD). As the affective and cognitive subscales of self-reported empathy (IRI) were constructed as separate components, individual independent sample t-tests were performed on EC, PD and PT mean scores to assess possible group differences. Independent sample t-tests were also used to determine group differences in selfreported psychopathy (PRI-R global score and subscales IE, FD, C) and on QST measures. A step-up Bonferroni procedure (Hochberg procedure; Olejnik et al., 1997) was used to adjust the alpha value and correct for multiple tests. Finally, Pearson correlations were used to investigate the relationships between QST's detection and pain thresholds, cortical somatosensory response and psychopathy traits.

2.6. Electromyographic (EMG) and electroencephalographic (EEG) recordings Electromyographic (EMG) activity was recorded (MP150, Biopac System) using 2-cm Ag–AgCl surface electrodes over the First Dorsal Interosseus (FDI) of the right hand. EEG was acquired at a sampling rate of 500 Hz from 124 þ4 Ag/AgCl electrodes contacting the scalp surface by way of potassium chloride electrolytesoaked sponges (HCGSN, Electrical Geodesic Inc., Oregon) and amplified with the EGI GES250 system. Electrode impedances were kept below 50 kΩ. A detailed description of EEG recording methods is presented in the Supplementary materials section. 2.7. Procedure During the 60-min EEG session, participants were seated in an armchair with their right arm on the armrest while watching a 20-inch LCD monitor positioned at a distance of 85 cm in an audiometric room. Stimuli were presented with a computer running E-Prime software (Version 2.0, Psychology Software Tools, Inc.)

2.8.2. EEG The analyses were performed at the peak of somatosensory gating located in a circular region of interest (ROI) of 3-cm radius around the P3 site (in the 10–20 electrode system) based on a functional localizer performed on an independent group of participants with the same design (Marcoux et al., 2013). This peak was found in the left centro-parietal electrodes (67, 71 and 72 in the EGI system [HCGSN] close to P3–P1 in the 10–20 system). See Supplementary materials for details of the EEG pre-processing procedure. Stability of baseline during fixation cross [  200:0 ms] was assessed using a 3(Conditions: Pain vs. No Pain vs. Neutral)  2(Groups: NPD vs. CC) mixed design ANOVA. Three time windows were used for the analyses based on a previous study using the same procedure (Marcoux et al., 2013). First, the mean energy during the Gating period [600:800 ms] and the Fixation Cross Baseline [  200:0 ms] were compared for each condition using simple t-tests against H0 (i.e., absence of gating) to investigate the general gating effect. Second, pain anticipation ([Averaged Pain and No Pain] vs. Neutral) was tested by comparing ratios of mean energy (related to baseline) between the three experimental

344

L.-A. Marcoux et al. / Psychiatry Research: Neuroimaging 224 (2014) 341–348

conditions during the Gating period [600:800 ms] with a one-way repeated measures ANOVA (Conditions [3: Pain vs. No Pain vs. Neutral]). This allowed the examination of the effect of the scenario before the pain-evoking event on the somatosensory gating (SG). Third, pain-specific gating was assessed in comparison to a second picture baseline using ratios ([Second Picture Baseline–3rd Picture Pain Gating]/Second Picture Baseline). The second picture baseline was defined as the period window ranging from 800 to 1000 ms that preceded the presentation of the third image where no effect of condition was observed. The ratios for the painful and non-painful conditions were compared using a 2 (Conditions: Pain, No Pain)  2(Groups: CC, NPD) mixed design ANOVA during the maximal Gating period [1300:1500 ms] (see Marcoux et al. (2013)). All analyses were done with an alpha level set at 0.05 and corrected with the Bonferroni procedure for multiple comparisons. Descriptive statistics in the results section are presented as mean and standard deviations (M 7 SD).

Coldheartedness subscale (t(23)¼ 0.518, p¼0.609, α ¼ 0.008). This confirmed that the NPD patients showed higher psychopathic traits than controls. 3.3. Pain sensitivity Independent samples t-tests conducted on sensory testing data indicated no significant group difference on thermal detection (CDT: t (23)¼1.21, p¼0.238, α ¼ 0.025; WDT: t(23)¼0.499, p¼0.623, α ¼0.0125) or thermal pain thresholds (CPT: t(23)¼0.150, p¼ 0.882, α ¼0.01; HPT: t(23)¼ 0.727, p¼0.475, α ¼0.02), but a significantly higher pressure pain threshold for NPD compared with CC (PPT: t (23)¼5.9, p¼ 0.024, α ¼0.05) (see Table 3).

3. Results 3.4. EEG 3.1. Pain ratings in others Analyses performed on pain intensity ratings confirmed a significant main effect of Condition (Pain: 5.8 70.37; No Pain: 0.006 70.11; Neutral: 0.0137 0.23; F(1,22)¼ 245.5, po 0.001), but neither the main effect of group (CC: 5.6 71.5; NPD: 6.027 2.1; F (1,22)¼0.539, p¼ 0.417) nor the Group by Condition interaction (F(1,22) ¼0.494, p ¼0.579) was significant. Paired comparisons indicated that mean pain ratings for painful scenarios differed significantly from ratings for non-painful (p o0.001) and neutral scenarios (p o0.001) whereas the difference between the latter two was not significant (p¼ 0.183). 3.2. Self-reported empathy and psychopathy Table 2 shows group comparisons on IRI subscales, which revealed a significant between-group difference for Perspective Taking (t(23)¼ 2.91, p¼ 0.008, α ¼0.01) but not for the Empathic Concern (t(23)¼0.037, p¼0.939, α ¼0.025) or the Personal Distress subscales (t(23)¼0.991, p¼ 0.368, α ¼0.007). This suggests lower cognitive empathy in NPD patients compared with controls. Analyses performed on the PPI-R scores revealed substantial differences across groups on psychopathic traits (see Table 2). First, the mean global score was significantly higher in the NPD group than in the CC group (t(23)¼5.82, po0.001, α ¼0.05). Second, the NPD group showed significantly higher scores on the Impulsivity–Egocentricity (t(23)¼ 5.17, po0.001, α ¼0.02) and Fearlessness-Dominance subscales (t (23)¼ 5.41, po0.001, α ¼ 0.025), but no difference was found on the

CC Mean (S.D.)

21.5 (2.9) 18.2 (4.1) 9.1 (3.2)

NPD Mean (S.D.)

16.2 (3.5) 18.1 (4.4) 7.8 (3.4)

Group comparisons p-value

0.008n 0.907 0.368 n

296.5 (13.2) 344.1 (25.3) o 0.001 151.5 (9.7) 181.1 (17.9) o 0.001n 99.2 (6.8)

116.3 (8.9)

3.4.2. Pain-anticipation effect As the first and second images were all identical in Pain and No Pain conditions, we pooled them and compared them with the first two pictures of the Neutral condition to assess the possible effect of pain anticipation. Ratios of mean energy were compared during the Gating period [600:800 ms], where gating

45.8 (4.9)

46.8 (4.5)

Statistically significant corrected alphas.

CC M (S.D.)

NPD M (S.D.)

Group comparisons p-value

Detection thresholds (1C) CDT WDT

28.9 (1.4) 34.4 (0.9)

29.6 (1.7) 34.3 (0.6)

0.238 0.623

Pain thresholds (1C) CPT HPT

16.1 (7.6) 42.9 (3.7)

16.6 (8.4) 41.9 (2.7)

0.882 0.475

PPT (kg/cm2 s)

10.7 (2.4)

14.2 (4.6)

0.024n

o 0.001n 0.609

Mean and S.D. values for each group, and p-values associated with an independent sample t-test on the differences between groups. p-Values here are uncorrected. CC: community control group, NPD: narcissistic personality disorder group. For each test, degrees of freedom¼ 23. n

3.4.1. General gating effect The amplitude of the general gating (Fig. 2) was, on average, 0.22 and 0.19, corresponding to a diminution of 22% from the Cross Baseline's raw amplitude in NPD patients and 19% in CC subjects, respectively, during the [600:800 ms] period. Contrasts between the Baseline and the General Gating period were all statistically significant for both groups ([NPD]: No Pain: t(23)¼6.39, po0.001, ratio¼ 0.2270.09; Pain: t(23)¼7.06, po0.001, ratio¼0.2570.03; Neutral: t(23)¼4.91, p¼0.001, ratio¼0.1870.05; [CC]: No Pain: t(23)¼7.07, po0.001, ratio¼0.1970.06; Pain: t(23)¼ 7.01, po0.001, ratio¼ 0.2170.02; Neutral: t(23)¼ 7.34, po0.001, ratio¼0.1870.03).

Table 3 Sensory and pain sensitivity testing.

Table 2 Psychological assessment.

IRI Perspective Taking Empathic Concern Personal Distress PPI-R Total score Factor 1—Impulsivity– Egocentricity Factor 2—FearlessnessDominance Factor 3—Coldheartedness

Baseline stability was statistically confirmed during the Fixation Cross Baseline period [ 200:0 ms] before the first picture onset and during the Second Picture Baseline period [800:1000 ms] before the Third Picture onset (Painful or Non-Painful ending; see Supplementary materials). Hence, subsequent general gating, pain anticipation and pain gating effects can be considered as distinct processes that vary according to the different visual stimuli changes, and not to a merely lasting effect of previous events.

Geometric mean and S.D. values for both groups, and p-values associated with an independent sample t-test on the differences between groups. p-Values here are uncorrected. CC: community control group, NPD: narcissistic personality disorder group, CDT: cold detection threshold, WDT: warm detection threshold, CPT: cold pain threshold, HPT: heat pain threshold. For each test, degrees of freedom¼ 23. n

Statistically significant corrected alphas.

L.-A. Marcoux et al. / Psychiatry Research: Neuroimaging 224 (2014) 341–348

345

Fig. 2. Lower image: Back-top view of the subtraction map identifying the ROI electrodes (67–71–72, in the EGI system [HCGSN] corresponding to P1–P3 in the 10–20 system) in which the somatosensory gating (SG) was showing the greatest modulation during the first two pictures [0:1000 ms] in comparison to the Cross Baseline [  1000:0 ms]. Upper image: Time course of the mean energy (mA/m3) of the somatosensory steady-state response (SSSR) during the presentation of the pseudo-dynamic stimuli. Smoothing (500 ms). The mean energy of the somatosensory gating (SG) during the first two pictures (general gating [600:800 ms]) was significantly different from mean energy during the Cross Baseline [  200:0 ms] for each condition. The magnitude of the SG during the general gating [600:800 ms] was significantly greater in the Pain and NoPain conditions compared to the Neutral condition only for NPD patients. The mean energy ratios during the [1300:1500 ms] period were significantly different from those of the Second Picture Baseline [800:100 ms] in both groups and higher in NPD patients.

was present and stable. The Conditions effect (potential Pain vs. Neutral) was significant (F(1,22) ¼ 11.73, p ¼ 0.002). In NPD patients, paired comparisons for different conditions showed that the Neutral condition significantly differed from the Pain (p ¼0.016) and No Pain (p ¼0.026) conditions. The effects for the Pain and No Pain conditions did not differ significantly (p ¼1.00). No significant effects were found in the CC group as Pain and No Pain conditions did not differ from the Neutral condition (p ¼0.771). This suggests that a significant pain anticipation effect was present only in NPD patients.

(r¼0.424, p¼ 0.039), and between SG and Impulsivity–Egocentricity subscale scores (r¼ 0.488, p¼ 0.015; see Fig. 3a). Finally, the pressure pain threshold correlated only with the FearlessnessDominance subscale score (r¼0.444, p¼0.030; see Fig. 3b). This suggests that individuals displaying higher Impulsivity–Egocentricity and Fearlessness-Dominance psychopathic traits were showing respectively stronger somatosensory resonance and higher pain pressure threshold.

4. Discussion 3.4.3. Pain-observation effect Ratios of mean energy during the third picture were subsequently compared between Pain and No Pain conditions for both groups using a 2(condition: Pain vs. No Pain)  2(groups: CC vs. NPD) ANOVA. The results showed a significant Condition by Group interaction for the third picture time window [1300:1500 ms] (F (1,22)¼ 4.67, p ¼0.042). Post-hoc analyses showed a significant difference between Pain and No Pain conditions for the CC group (F(1,22) ¼3.21 p¼ 0.039) and the NPD group (F(1,22)¼12.46, p ¼0.001). The between-groups difference for the Pain conditions was significant (t(23) ¼ 2.15, p ¼0.017). These findings indicate that somatosensory resonance was stronger in NPD patients than in healthy controls. This result remain significant after using the pain pressure threshold as a covariate. 3.5. Correlations between EEG and behavioral data The correlations computed between the cortical somatosensory response and the different psychopathy dimensions when combining all participants revealed a significant positive relationship between somatosensory gating (SG) and PPI-R global scores

The goal of this study was to determine if severe NPD patients presenting psychopathic traits show alterations in somatosensory responses when observing pain in others. As expected, the results showed that NPD patients scored higher on psychopathic traits than controls. Normal detection and pain thresholds were observed in NPD patients with the exception of pain pressure thresholds that were higher. NPD patients showed more somatosensory resonance despite lower scores on the Perspective-Taking scale of the IRI. Interestingly, somatosensory resonance correlated with the Impulsivity–Egocentricity dimension of psychopathy. These findings suggest that severe NPD patients with psychopathic traits show stronger somatosensory resonance and increased attention to the somatic representation of observed pain. This may have been facilitated by weaker affective and empathic responses toward the individual in pain. 4.1. Evaluation of others' pain is equivalent between the groups The results did not reveal any group difference on the amount of pain observed in the visual stimuli. This corroborates previous

346

L.-A. Marcoux et al. / Psychiatry Research: Neuroimaging 224 (2014) 341–348

Fig. 3. Correlates of [A] Other's pain assessment using the mean energy (mA/m3) ratios of somatosensory gating to pain observation with Factor 1 (F1) Impulsivity– Egocentricity (IE ) traits of the revised version of the Psychopathic Personality Inventory (PPI-R), and [B] Self-pain assessment using the geometric means (Lb/cm2 s) of the Pain Pressure Thresholds (PPT) test with Factor 2 (F2) Fearlessness-Dominance (FD) traits of the PPI-R.

studies showing that male adolescents (Decety et al., 2009; Cheng et al., 2012) and adults (Marcoux et al., 2013) with high psychopathic traits evaluate the pain of others similarly to control participants. This suggests that the differences observed in somatosensory resonance to others' pain are not due to perceptual differences in the amount of pain depicted in the images. 4.2. Pain sensitivity was equivalent between the groups Sensory testing revealed only one significant group difference, with severe NPD patients reporting higher pain pressure thresholds than controls. This is consistent with a previous study showing that juvenile offenders, particularly those with higher callous-unemotional traits, have higher pain pressure thresholds compared with community adolescents (Cheng et al., 2012). This specific between-group difference to pain pressure may reflect a rating bias as this sub-test involved a social context, i.e., pain pressure was the only sensory test administered directly by the experimenter as opposed to a stimulation device. Overall, findings from the QST suggest that the change in somatosensory resonance to others' pain observed in NPD patients was not associated with global changes in the self-perception of pain. 4.3. Altered empathy in NPD patients Self-reported trait empathy scores revealed a significant betweengroup difference in perspective taking, a cognitive component of empathy, confirming results from a recent study with NPD patients (Ritter et al., 2011). It has been suggested that lower scores on a measure of perspective taking could reflect a lack of motivation to adopt the psychological perspective of others rather than a reduced ability to do so (Dolan and Fullam, 2004; Ritter et al., 2011). This result may also point out the likely effect of borderline personality disorder (BPD) comorbidity found in the clinical group, as BPD is associated with impaired mentalizing abilities (Kernberg, 2012). No group differences were observed in empathic concern or personal distress, suggesting no impairment in emotional empathy in NPD patients. However, a clinician-based judgment evaluation of empathy (from the PCL-R; Hare, 2003) indicated that these NPD patients were scoring, on average, 1.5 on a [0–2] scale of Insensibility/Lack of empathy, reflecting diminished emotional empathy disposition. 4.4. Somatosensory gating to pain anticipation/observation was stronger in NPD patients The EEG data revealed that somatosensory gating during the anticipation phase was stronger in severe NPD patients than in the

control participants. However, it is widely demonstrated that psychopathic individuals have blunted responses in anticipation to the threat of an aversive outcome (Patrick et al., 1993; Herpertz et al., 2001; Birbaumer et al., 2005; Rothemund et al., 2012). Findings from a recent study (Caes et al., 2012) indicated that, when anticipating pain in others, participants displaying more psychopathic traits showed a lower fear-potentiated startle reflex. Lower cognitive or reduced emotional empathy (as suggested by clinician-based ratings) reported in NPD patients may have facilitated attention to the somatic cause of the pain (as the current task required), as they were not engaged in other-oriented feelings. The absence of an anticipation effect in control individuals parallels results from Chen et al. (2012) showing an absence of significant mu rhythm suppression in community participants during pain-anticipation periods using a similar design and similar timing. However, the anticipation condition in this study is novel, and more studies should include both Neutral and Non-Pain conditions to test this further. The current results add to a growing body of literature reporting higher cerebral reactivity to pain-evoking stimuli in regions associated with sensorimotor resonance in individuals with high psychopathic traits (Fecteau et al., 2008; Decety et al., 2009; Marcoux et al., 2013). Interestingly, a recent study showed increased insula activation during the observation of dynamic stimuli of people being hurt and facial expressions of pain in high psychopathic inmates, compared to controls, together with less activation in ventromedial prefrontal cortex (vmPFC) (Decety et al., 2013). These findings suggest that recurrent emotional deficits found in psychopathy and associated vmPFC hypoactivation might be coupled with increased activity in regions involved in processing interoceptive qualities of bodily pain, indicative of a sensory/ cognitive approach in assessing others' pain. Both behavioral and EEG results add to the ongoing discussion about whether psychopathic individuals could have abnormally low spontaneous propensity for empathy but relatively normal ability to empathize when prompted to do so (Keysers and Gazzola, 2014). 4.5. Limitations Caution is needed in interpreting the current results within a continuum of psychopathic traits, as NPD patients may differ qualitatively from prototypical psychopaths in several functioning areas such as the capacity to establish some attachment relationships (Kernberg, 2012). Furthermore, the limited sample size reflects the challenges involved in recruiting individuals who have such traits (but who are not incarcerated) as they tend to chronically fail to keep appointments in the clinical framework.

L.-A. Marcoux et al. / Psychiatry Research: Neuroimaging 224 (2014) 341–348

This might also underline a slight recruitment bias, limiting the generalization potential of our results as NPD patients who have respected their commitment to the study might represent a subsample of this group. Another point of concern is the absence of a group difference on Coldheartedness as measured by the PPI-R, which suggests similar emotional reactivity in both groups, while PCL-R assessments showed significant callous/lack of empathy traits in NPD patients. Although the same instrument was used for control subjects and patients on the construct of interest, it might be useful in future studies to perform a more thorough assessment of personality in control subjects to better characterize this group.

5. Conclusion Overall, this study indicated that severe NPD patients with psychopathic traits displayed greater somatosensory resonance while anticipating pain-evoking scenarios and during pain observation compared with community controls. It also highlighted the importance for further studies to better characterize the relationship between the arousal stemming from the anticipation and observation of pain in others and the capacity to empathize and care about others in individuals with psychopathic traits.

Acknowledgments Technical support was mainly provided by the Consortium d’imagerie en neurosciences et santé mentale de Québec (CINQ) for EEG acquisition and analysis. This study was supported by funding from the Natural Sciences and Engineering Research Council of Canada (NSERC) (Grant no. RGPIN 341631 - 2007), the Canadian Foundation for Innovation (CFI) (Grant no. 15572), and salary grants to PLJ from the Fonds de la recherche du Québec – Santé (FRQS) (Grant no. 8149) and the Canadian Institutes of Health Research (CIHR) (Grant no. MSH-87688).

Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.pscychresns.2014. 10.002. References Avenanti, A., Bueti, D., Galati, G., Aglioti, S.M., 2005. Transcranial magnetic stimulation highlights the sensorimotor side of empathy for pain. Nature Neuroscience 8, 955–960. Birbaumer, N., Veit, R., Lotze, M., Erb, M., Hermann, C., Grodd, W., Flor, H., 2005. Deficient fear conditioning in psychopathy: a functional magnetic resonance imaging study. Archives of General Psychiatry 62, 799–805. Blair, R.J., 2005. Responding to the emotions of others: dissociating forms of empathy through the study of typical and psychiatric populations. Consciousness and Cognition 4, 698–718. Bufalari, I., Aprile, T., Avenanti, A., Di Russo, F., Aglioti, S.M., 2007. Empathy for pain and touch in the human somatosensory cortex. Cerebral Cortex 17, 2553–2561. Caes, L., Uzieblo, K., Crombez, G., De Ruddere, L., Vervoort, T., Goubert, L., 2012. Negative emotional responses elicited by the anticipation of pain in others: psychophysiological evidence. Journal of Pain 13, 467–476. Chen, C., Yang, C.Y., Cheng, Y., 2012. Sensorimotor resonance is an outcome but not a platform to anticipating harm to others. Social Neuroscience 7, 578–590. Cheng, Y., Hung, A.Y., Decety, J., 2012. Dissociation between affective sharing and emotion understanding in juvenile psychopaths. Developmental Psychopathology 24, 623–636. Cleckley, H., 1941. The Mask of Sanity, 1st ed. Mosby, St. Louis. Cooke, D.J., Michie, C., 2001. Refining the construct of psychopathy: toward a hierarchical model. Psychological Assessment 13, 171–188. Davis, M.H., 1980. A multidimensional approach to individual differences in empathy. JSAS Catalogue of Selected Documents in Psychology 10, 85.

347

Decety, J., Jackson, P.L., 2004. The functional architecture of human empathy. Behavioral Cognitive Neuroscience Review 3, 71–100. Decety, J., Michalska, K.J., Akitsuki, Y., Lahey, B.B., 2009. Atypical empathic responses in adolescents with aggressive conduct disorder: a functional MRI investigation. Biological Psychology 80, 203–211. Decety, J., Skelly, L.R., Kiehl, K.A., 2013. Brain response to empathy-eliciting scenarios involving pain in incarcerated individuals with psychopathy. JAMA Psychiatry 70 (6), 638–645. Dimaggio, G., Nicolo, G., Popolo, R., Semerari, A., Carcione, A., 2006. Self-regulatory dysfunctions in personality disorders: the role of poor self-monitoring and mindreading. Applied Psychology: An International Review 55, 397–407. Dolan, M., Fullam, R., 2004. Theory of mind and mentalizing ability in antisocial personality disorders with and without psychopathy. Psychological Medicine 6, 1093–1102. Fecteau, S., Pascual-Leone, A., Theoret, H., 2008. Psychopathy and the mirror neuron system: preliminary findings from a non-psychiatric sample. Psychiatry Research 160, 137–144. Hare, R.D., 1991. The Hare Psychopathy Checklist—Revised. Multi-Health Systems, Toronto. Hare, R.D., 2003. The Hare Psychopathy Checklist—Revised, 2nd ed. Multi-Health Systems, North Tonawanda, NY. Herpertz, S.C., Werth, U., Lukas, G., Qunaibi, M., Schuerkens, A., Kunert, H.J., Freese, R., Flesch, M., Mueller-Isberner, R., Osterheider, M., Sass, H., 2001. Emotion in criminal offenders with psychopathy and borderline personality disorder. Archives of General Psychiatry 58, 737–745. Karpman, B., 1941. On the need of separating psychopathy into two distinct clinical types: the symptomatic and the idiopathic. Journal of Criminal Psychopathology 3, 112–137. Kelsey, R.M., Ornduff, S.R., McCann, C.M., Reiff, S., 2001. Psychophysiological characteristics of narcissism during active and passive coping. Psychophysiology 38, 292–303. Kernberg, O., 2012. Mentalization, mindfulness, insight, empathy, and interpretation. In: Kernberg, O. (Ed.), The Inseparable Nature of Love and Aggression: Clinical and Theoretical Perspectives. American Psychiatric Publishing, Washington, DC, pp. 57–79. Keysers, C., Gazzola, V., 2014. Dissociating the ability and propensity for empathy. Trends in Cognitive Sciences 18, 163–166. Kiehl, K.A., Smith, A.M., Hare, R.D., Mendrek, A., Forster, B.B., Brink, J., Liddle, P.F., 2001. Limbic abnormalities in affective processing by criminal psychopaths as revealed by functional magnetic resonance imaging. Biological Psychiatry 50, 677–684. 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. Lamm, C., Nusbaum, H.C., Meltzoff, A.N., Decety, J., 2007. What are you feeling? Using functional magnetic resonance imaging to assess the modulation of sensory and affective responses during empathy for pain. PLoS One 12, e1292. Lilienfeld, S.O., 1994. Conceptual problems in the assessment of psychopathy. Clinical Psychology Review 14, 17–38. Lilienfeld, S.O., Widows, M.R., 2005. Psychopathic Personality Inventory—Revised: Professional Manual. Psychological Assessment Resources Inc., Odessa, FL. Marcoux, L.-A., Michon, P.-E., Voisin, J.I.A., Lemelin, S., Vachon-Presseau, E., Jackson, P.L., 2013. The modulation of somatosensory resonance by psychopathic traits and empathy. Frontiers in Human Neuroscience 7, 1–13. Marsh, A.A., Finger, E.C., Mitchell, D.G.V., Reid, M.E., Sims, C., Kosson, D.S., Towbin, K.E., Leibenluft, E., Pine, D.S., Blair, R.J.R., 2008. Reduced amygdala response to fearful expressions in children and adolescents with callous-unemotional traits and disruptive behavior disorders. American Journal of Psychiatry 165, 712–720. Müller, J.L., Sommer, M., Wagner, V., Lange, K., Taschler, H., Röder, C.H., Schuierer, G., Klein, H.E., Hajak, G., 2003. Abnormalities in emotion processing within cortical and subcortical regions in criminal psychopaths: evidence from a functional magnetic resonance imaging study using pictures with emotional content. Biological Psychiatry 54, 152–162. Olejnik, S., Li, J., Supattathum, S., Huberty, C.J., 1997. Multiple testing and statistical power with modified Bonferroni procedures. Journal of Educational and Behavioral Statistics 22, 389–406. Patrick, C.J., Bradley, M.M., Lang, P.J., 1993. Emotion in the criminal psychopath: startle reflex modulation. Journal of Abnormal Psychology 102, 82–92. Ray, J.V., Hall, J., Rivera-Hudson, N., Poythress, N.G., Lilienfeld, S.O., Morano, M., 2013. The relation between self-reported psychopathic traits and distorted response styles: a meta-analytic review. Personality Disorders: Theory, Research, and Treatment 4, 1–21. Ritter, K., Dziobek, I., Preissler, S., Rüter, A., Vater, A., Fydrich, T., Lammers, C.H., Heekeren, H.R., Roepke, S., 2011. Lack of empathy in patients with narcissistic personality disorder. Psychiatry Research 187, 241–247. Rothemund, Y., Ziegler, S., Hermann, C., Gruesser, S.M., Foell, J., Patrick, C.J., Flor, H., 2012. Fear conditioning in psychopaths: event-related potentials and peripheral measures. Biological Psychology 90, 50–59. Singer, T., Seymour, B., O’Doherty, J., Kaube, H., Dolan, R.J., Frith, C.D., 2004. Empathy for pain involves the affective but not sensory components of pain. Science 303, 1157–1162. Stern, B.L., Caligor, E., Clarkin, J.F., Critchfield, K.L., Hörz, S., MacCornack, V., Lenzenweger, M.F., Kernberg, O.F., 2010. Structured interview of personality organization (STIPO): preliminary psychometrics in a clinical sample. Journal of Personality Assessment 92, 35–44.

348

L.-A. Marcoux et al. / Psychiatry Research: Neuroimaging 224 (2014) 341–348

Vachon-Presseau, E., Martel, M.O., Roy, M., Caron, E., Jackson, P.L., Rainville, P., 2011. The multilevel organization of vicarious pain responses: effects of pain cues and empathy traits on spinal nociception and acute pain. Pain 152, 1525–1531. Voisin, J.I.A., Marcoux, L.-A., Canizales, D.L., Mercier, C., Jackson, P.L., 2011. I am touched by your pain: limb-specific modulation of the cortical response to a tactile stimulation during pain observation. Journal of Pain 12, 1182–1189.

Watson, P.J., Grisham, S.O., Trotter, M.V., Biderman, M.D., 1984. Narcissism and empathy: validity evidence for the Narcissistic Personality Inventory. Journal of Personality Assessment 48, 301–304. Watson, P.J., Morris, R.J., 1991. Narcissism, empathy, and social desirability. Personality and Individual Differences 12, 575–579.