Altered processing of visual emotional stimuli in posttraumatic stress disorder: an event-related potential study.

Psychiatry Research: Neuroimaging ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Contents lists available at ScienceDirect

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

Altered processing of visual emotional stimuli in posttraumatic stress disorder: an event-related potential study Rotem Saar-Ashkenazy a,b,c, Hadar Shalev d, Magdalena K. Kanthak e, Jonathan Guez c,f, Alon Friedman a, Jonathan E. Cohen g,n a

Department of Cognitive-Neuroscience and Zlotowski Center for Neuroscience, Ben-Gurion University of the Negev, Beer-Sheva, Israel Department of Psychology and the School of Social-work, Ashkelon Academic College, Ashkelon, Israel c Department of Psychology, Achva Academic College, Beer-Tuvia regional council, Israel d Department of Psychiatry, Soroka University Medical Center, Beer-Sheva, Israel e Department of Biological Psychology, Technical University of Dresden, Dresden, Germany f Beer-Sheva Mental Health Center, Beer-Sheva, Israel g Sharett Institute of Oncology, Hadassah Medical Organization, Kiryat-Hadassah, POB 12000, Jerusalem 91120, Israel b

art ic l e i nf o

a b s t r a c t

Article history: Received 29 May 2014 Received in revised form 6 February 2015 Accepted 27 May 2015

Patients with posttraumatic stress disorder (PTSD) display abnormal emotional processing and bias towards emotional content. Most neurophysiological studies in PTSD found higher amplitudes of eventrelated potentials (ERPs) in response to trauma-related visual content. Here we aimed to characterize brain electrical activity in PTSD subjects in response to non-trauma-related emotion-laden pictures (positive, neutral and negative). A combined behavioral-ERP study was conducted in 14 severe PTSD patients and 14 controls. Response time in PTSD patients was slower compared with that in controls, irrespective to emotional valence. In both PTSD and controls, response time to negative pictures was slower compared with that to neutral or positive pictures. Upon ranking, both control and PTSD subjects similarly discriminated between pictures with different emotional valences. ERP analysis revealed three distinctive components (at 300, 600 and 1000 ms post-stimulus onset) for emotional valence in control subjects. In contrast, PTSD patients displayed a similar brain response across all emotional categories, resembling the response of controls to negative stimuli. We interpret these findings as a braincircuit response tendency towards negative overgeneralization in PTSD. & 2015 Elsevier Ireland Ltd. All rights reserved.

Keywords: PTSD ERP Trauma Emotion Overgeneralization

1. Introduction Emotions are response-dispositions that are typically accompanied by physiological and behavioral changes, and can be coded on a dimensional scale of valence, i.e., positive to negative (Nesse and Ellsworth, 2009). Disrupted emotional reactivity and modulation are central features of post-traumatic stress disorder (PTSD), with symptoms such as intrusive memory, trauma flashbacks and avoidance of stimuli, persistent hyper-arousal and hyper-vigilance (American Psychiatric Association, 2013) reflecting impaired perceptual, attentional, and emotional-memory processes. Studies in healthy controls have been conducted to identify the chronology of normal emotional processing and neurophysiological mechanisms underlying the perception of emotional stimuli. These studies have used high temporal resolution event-related n

Corresponding author. E-mail address: [email protected] (J.E. Cohen).

potentials (ERPs) to distinguish voluntary attention (i.e., attention that is voluntarily directed towards a stimulus) from spontaneous attention (i.e., attention that is attracted by a stimulus) to emotionally negative information during different stages of perceptual identification. Specifically, studies support a modulatory effect of negative emotion on early visual perception and attention, associated with augmentation of the visual occipital P1 component of the ERP at 100 ms post-stimulus. Emotional processing has also been reported to be associated with increased early posterior negativity (EPN, at 200–300 ms post-stimulus), late positive potential (LPP, occurring at latencies larger than 300 ms post-stimulus), and a sustained positive slow wave as compared with neutral contents (see reviews by Schupp et al., 2006; Olofsson et al., 2008). ERP studies probing specifically for emotional processing abnormalities in PTSD are surprisingly limited in number and were mainly directed to test P300 abnormalities to trauma-related material. Applying a modified Stroop paradigm using personaltraumatic, personal-positive, and neutral words, Metzger et al. (1997b) tested the emotional interference effect in PTSD patients

http://dx.doi.org/10.1016/j.pscychresns.2015.05.015 0925-4927/& 2015 Elsevier Ireland Ltd. All rights reserved.

Please cite this article as: Saar-Ashkenazy, R., et al., Altered processing of visual emotional stimuli in posttraumatic stress disorder: an event-related potential study. Psychiatry Research: Neuroimaging (2015), http://dx.doi.org/10.1016/j.pscychresns.2015.05.015i

R. Saar-Ashkenazy et al. / Psychiatry Research: Neuroimaging ∎ (∎∎∎∎) ∎∎∎–∎∎∎

2

Table 1 Demographics and characteristics of PTSD subjects Subject

Sex

Age (years)

Education (years)

Trauma

Medications

Bodily physical injury

Head physical injury

1 2 3 4 5 6 7 8 9 10 11 12 13 14

M M M M M F M M M F M M F F

45 24 25 37 37 22 48 37 51 31 23 26 40 18

12 12 12 15 12 12 12 12 10 15 12 15 12 12

Work related Fall from height MVA Military related Assault MVA MVA Military related Fall from height MVA MVA Military related MVA MVA

Venlafaxine 150 mg/d None None None Paroxetine 20 mg/day None Paroxetine 20 mg/day Paroxetine 20 mg/day Bonserine 30 mg/day None Paroxetine 20 mg/day Sertraline 100 mg/day None None

No No No No Yes-mild No Yes Yes No No Yes No Yes Yes

No No No no No Lt. frontoparietal contusion No No No No No No No No

Note. Data are reported for the PTSD patients who participated in the study; MVA ¼ Motor Vehicle Accident. Lt ¼ left.

versus controls. Individuals with PTSD had a significantly reduced and delayed P300 across word types, as well as a slower response time (RT), especially for traumatic words. In the same study frontal P300 amplitudes were larger to both personal-positive and personal-traumatic words as compared with neutral words across groups. In a study that applied the oddball paradigm, P300 amplitudes to emotionally meaningful words were significantly related to PTSD symptoms, in particular avoidance and arousal (Blomhoff et al., 1998). Attias et al. (1996) applied a modified oddball paradigm in which subjects were requested to discriminate between animal pictures (targets), emotionally neutral pictures (trauma-irrelevant non-targets), and combat-related pictures (trauma-irrelevant non-target probes). The authors reported target stimuli (as compared with trauma-irrelevant non-targets) evoked accentuated P300 amplitudes in both controls and PTSD patients while non-target combat-related probes (as compared with trauma-irrelevant non-targets) elicited enhanced P300 and N100 amplitudes in the PTSD group only. Stanford et al. (2001) also used a modified oddball paradigm to compare Vietnam-war veterans with and without PTSD. The results revealed that PTSD patients demonstrated attenuated P300 response to neutral targets and increased responsiveness to trauma-relevant combat stimuli but not to trauma-irrelevant social-threat stimuli at frontal regions. The preponderance of the evidence (mostly based on the oddball paradigm or a modification of it) supports the view that PTSD patients show sensitization of the P300 response specifically to trauma-related stimuli and a diminished response to neutral stimuli. Nevertheless, there are reports of non-emotional information-processing impairments in PTSD, with studies demonstrating impaired cognitive processing at early (Gillette et al., 1997; Neylan et al., 1999; Skinner et al., 1999; Ghisolfi et al., 2004; Holstein et al., 2010; Gjini et al., 2013) as well as mid-late (Paige et al., 1990; McPherson et al., 1997; Metzger et al., 2002) temporal stages. Assessment of P300 non-emotional information-processing impairments in PTSD has mainly been conducted with an auditory or visual oddball design, with studies reporting lower amplitude and longer latency P300 components in this population (McFarlane et al., 1993; Charles et al., 1995; Metzger et al. 1997a, b; Kimble et al., 2000; Felmingham et al., 2002; Araki et al., 2005; Veltmeyer et al., 2005). These findings have also been reported when using the "Go/NoGo" paradigm, with studies reporting longer latency P300 in the NoGo trials and higher P300 amplitudes in non-target trials in PTSD populations (Shucard et al., 2008). Together, these studies indicate the presence of an altered cognitive pattern of selective attention processing in early as well as late temporal stages in PTSD in addition to a vulnerability to traumatic

reminiscences. In the current study, we aimed to test the hypothesis that abnormal emotional responsiveness in early as well as late stages of stimuli processing in PTSD is evident even for nontrauma-related context using an emotional valence choice-response task. We hypothesized that abnormal emotional responsiveness would be reflected in altered brain electrical activity, as measured via ERPs. Specifically, we hypothesized that (1) the response to negative stimuli in PTSD would be altered as compared with positive/neutral stimuli, and (2) that PTSD patients would show higher amplitudes for negative stimuli as compared with control subjects.

2. Methods 2.1. Subjects The results reported in this study are part of a larger study exploring genetic, physiological, anatomical and cognitive characteristics of PTSD patients recruited from the mental trauma clinic at Soroka University Medical Center. All patients (46 in total) were interviewed using the Clinician Administered PTSD Scale (CAPS) by a trained psychiatrist. Of those, 14 patients with extremely severe PTSD, as determined by a CAPS score Z80 (Weathers et al., 2001), participated in the current study. Fourteen control subjects with no psychiatric records and/or major traumatic experience/other trauma-related disorders were recruited for the current study. All participants (PTSD patients and controls) were recruited on a voluntary basis and were not compensated for participation. Exclusion criteria for all subjects included head trauma, preexisting neuro-psychiatric disorders, alcohol abuse or use of illicit drugs. Subject characteristics were as follows: mean age (years), 33.147 10.11 SD and 26.14 73.33 SD, mean education (years) 12.50 7 1.45 and 15.647 2.27, 10 and 9 males, for PTSD and controls, respectively (for full description of subject characteristics, see Table 1). All procedures were approved by the Soroka University Medical Center institutional review board. Written informed consent was obtained from all participants. 2.2. Emotional paradigm Pictures from the International Affective Picture System (IAPS, Center for the Study of Emotion and Attention [CSEA–NIMH], 1995, University of Florida, Gainesville, FL, USA) ranked on a valence scale [from 1 (positive) to 9 (negative)] were sorted into three groups on the basis of positive, negative, or neutral emotional content. To ensure that the hypothesized effects were not a



consequence of specific stimuli, two sets of pictures were created (version A: valence scores of 7.32 70.08 SD, 2.84 70.12 SD, and 4.98 70.05 SD; version B: 7.317 0.11 SD, 2.78 70.12 SD, and 4.97 70.08 SD; for negative, positive and neutral, respectively). No significant difference in valence was found between versions in any of the emotional categories: t(62) ¼0.72, t(62) ¼ 0.89 and t (62) ¼0.88, NS). Picture parameters such as brightness, contrast and spatial frequencies did not differ among the valence-category groups. In contrast with many studies in PTSD in which traumaspecific negative stimuli were used, in the current study we had a varied set of stimuli; thus, all subjects in the current study were exposed to a wide set of negative traumatic stimuli. The experiment consisted of four runs, each consisting of three blocks, one for each emotional valence (negative, positive, and neutral). Within blocks, eight pictures were presented for 1.5 s each, followed by a black screen for 0.5 s. Each emotional block was preceded by a fixation block in which a cross was presented eight times, for 1.5 s each, with a 0.5-s interval. Block order was randomized within runs. Subjects underwent a choice-response task in which they were instructed to maintain fixation at the center of the screen for the duration of the task and press one of two buttons using their right hand depending on the presence or absence of a person or human part in the picture, and RT was recorded. Stimulus probability distribution (i.e., human or non-human) was approximately 50% (48.43%) in each block. Stimuli were presented and responses were collected using E-prime 1.1 software. Following EEG acquisition, subjects ranked the pictures they viewed for valence by using the self-assessment manikin scale of 1 (positive) to 9 (negative) (Bradley and Lang, 1994). 2.3. EEG acquisition Subjects were seated at a distance (from the eyes) of 1.5 m from a 15-in computer screen. EEG was recorded using a 128-channel digital acquisition unit (CEEGRAPH IV, Biological Systems Corp., Mundelein, IL, USA) employing a 64-(58 EEG) electrode cap (Electro-Cap, Eaton, OH, USA). Sampling rate was 256 Hz, and online filtering was performed between 0.1 Hz and 100 Hz. All electrode impedances were maintained below 5 kΩ. Central frontopolar site was used as ground. An additional channel registered the onset of stimuli via an opto-coupler box built in-house. 2.4. ERP analysis Analysis was performed using in-house prepared Matlab scripts (2007b, MathWorks, Natick, MA, USA). EEG segments were inspected visually for gross artifacts (i.e., components exceeding 100 mV) and, in their absence, cut according to experimental blocks and sorted into the corresponding emotional categories. Electrodes were re-referenced to average, and data were band-pass-filtered off-line at 1–35 Hz using a zero-phase finite impulse response (FIR) digital filter. Baseline correction was performed on each segment subtracting the mean voltage of the 200 ms preceding stimulus onset from the 1800-ms post stimulus analyzed (corresponding to 450 time points with a 256-Hz sampling rate; for this reason the first segment in each block was discarded for lack of preceding baseline). For each subject, EEG segments were averaged to yield ERPs in each emotional category and for each electrode. To reduce the number of comparisons, we averaged neighboring electrodes into 17 regions of interest (ROIs) (electrodes in each region are depicted in Fig. 1). A two-step procedure was applied for each group (PTSD patients and controls) separately to detect emotional modulation of the ERP components (see Schupp et al. , 2003). In the first step, repeated measures analysis of variance (ANOVA) across subjects was performed on the ERPs from each region and time point with

3

Fig. 1. Regions of interest. Scalp map demonstrating the 17 regions of interest (ROIs) and corresponding electrodes. Lt. refers to left ROIs and Rt. refers to right ROIs.

the factor of emotional valence (positive, negative, and neutral) in order to identify temporal and spatial characteristics of ERP modulation by emotion. A significance threshold of p o0.01 was applied; with eight consecutive significant time points required (0.018 450 [time points] 17 [regions]¼resulting in a probability of 7.65 13 for type 1 error) with an additional requirement of at least two adjacent regions showing statistically significant results simultaneously for results to be considered meaningful. This method was chosen to preserve the high temporal resolution of EEG while offering a method for correction for multiple comparisons. Using methods such as Bonferroni or false discovery rate (FDR) corrections will assign similar significance to distinct and separate time points as to contiguous time points, which lacks a physiological reason (as the biological processes studied are not likely to be as brief as a single time sample). In this regard, requiring results to be significant in contiguous time points (as in the method of Schupp et al., 2003) adds a physiologically relevant correction for randomly generated significant p values. In the second step, regional and temporal components identified as significant in the first step were averaged and subjected to repeated measures ANOVA to conform to standard methods of presentation. 2.5. Source localization To determine the origin of each component, component means across subjects and emotions were subjected to sLORETA source localization (Pascual-Marqui, 2002). Subsequently, and using standardized tools implemented in sLORETA, statistical comparisons across subjects were performed on the sLORETA-transformed ERP components to determine the source of alterations in brain activation detected in the surface analysis. 2.6. Statistical analysis Statistical analysis was performed using Statistica 9.0 software. Repeated measures ANOVA was performed to determine the significance of both the behavioral and the ERP within-subject effects, with group (PTSD/control) as a between-subjects factor. The Mauchly test for sphericity was performed to test for equality of variances. When the assumption of sphericity was not met, the Greenhouse-Geisser correction was applied (Handy, 2005). Following this analysis, planned comparisons between groups (PTSD/



4

controls) and valences (positive, negative and neutral) were carried out to determine the factors contributing to the observed differences (p-values and effect-size are reported).

3. Results 3.1. Self-report and response time Self-reported measures of valence were collected from 14 controls and 12 patients (2 patients did not rank the pictures for valence due to technical problems). For the analysis of valence differences, two-way repeated-measures ANOVA with the factor group emotion was performed on the picture's rankings (see Fig. 2A). The two-way interaction was non-significant (F(2, 48) ¼ 1.1, p 40.05) as was the effect of group (F(1, 24)¼ 1.6, p4 0.05); the effect of emotional condition was, however, significant (F(2, 48) ¼ 80.5, p o0.01, η2p ¼0.771). Analysis of valence reports in each group determined that valence ranking was highest for negative, followed by neutral, and finally positive stimuli for both control and PTSD subjects; mean valence levels for controls and patients were as follows: controls 6.837 0.97 SD, 4.81 70.32 SD, 3.33 70.74 SD, F(2, 26) ¼71.7, p o0.01, η2p ¼0.847; all pair-wise comparisons were significant at p o0.01, and PTSD 6.04 71.46 SD, 4.40 71.53 SD, 3.28 71.21 SD, F(2, 22) ¼ 22.6, po 0.01, η2p ¼0.673; except for the neutral-positive comparison, which was not significant (p 4 0.05), all other pair-wise comparisons were significant at po 0.01. Thus, both control and PTSD subjects reported differential valence ranking of the emotional conditions, with no significant differences between groups. RT (collected "online", i.e., during EEG recordings) was available for 14 PTSD patients and 14 controls and analyzed by a two-way repeated measures ANOVA with the factor group emotion (see Fig. 2B). The two-way interaction was non-significant (F(2, 52) ¼ 0.603, p 40.05, η2p ¼0.022). There was a significant main effect for emotion (F(2, 52) ¼24.011, p o0.01, η2p ¼0.480), with longer RT for negative pictures (mean ¼949.76 ms, SD ¼130.484) as compared with positive and neutral ones (mean¼ 823.399 ms, SD ¼138.418). A trend for a main effect for group revealed that PTSD subjects responded more slowly across all emotional conditions, 909.854 768.11 SD ms, compared with 821.190 755.91 SD ms in controls (F(1, 26) ¼3.962, p ¼ 0.05, η2p¼ 0.132). Notably, RT for neutral pictures in the PTSD group (850.344 ms 7102.488 SD) did not significantly differ from that of controls to negative pictures (894.332 ms 7148.160 SD) (F(1, 26) ¼ 0.775, p 40.05, η2p ¼0.028).

3.2. ERP results Grand averaged representative ERPs over left frontal and occipito-temporal regions are shown in Fig. 3A and B, for control and PTSD subjects, and for positive, negative, and neutral emotional stimuli. Repeated measures ANOVA per time point and electrode was performed on ERPs in response to positive, negative, and neutral stimuli, following a two-step method proposed by Schupp et al. (2003) in order to identify ERP components differing by emotional valence. In healthy controls (Fig. 3C), three components were identified. First, a frontal positivity and corresponding parieto-occipital negativity significantly differentiated emotional stimuli at 310–350 ms post-stimulus and corresponding to the P300. This component, averaged across stimulus categories, localized to Brodmann area 11 in the frontal lobe (Fig. 3E). A second component was identified around 600 ms post-stimulus, with left frontocentral positivity, and corresponding right parieto-central negativity. This component localized to Brodmann area 7 in the parietal lobe. Finally, a late frontal positivity and a corresponding parietooccipital negativity were found around 1000 ms post-stimulus, localizing again to Brodmann area 7. PTSD subjects exhibited similar gross morphology of the grand averaged ERP; however, repeated measures ANOVA in this population failed to identify components differing by emotional valence (Fig. 3D). All three components correspond to clearly visible peaks in the ERPs seen in Fig. 3A–B. In conjunction with the Schupp method, the second step of analysis involved averaging the data over electrodes in regions identified in the first step of analysis and corresponding time windows. The averaged data were then subjected to repeated measures ANOVA with emotional valence and scalp regions as within-subject factors, and group (control vs. PTSD) as a betweensubject factor. 3.2.1. The P300 component Analysis was performed in the time window of 310–350 ms for the six regions identified in the first step of analysis (frontal left: Fp1, AF3, F5; frontal right: Fp2, AF4, F6; frontal midline: Fpz, Fz; parieto-occipital left: P1, P3, P5, PO7, O1; parieto-occipital right: P2, P4, P6, PO8, O2; parieto-occipital midline: Pz, Poz, Oz). A three-way repeated measures ANOVA was performed with the following factors: group (2 levels: control vs. PTSD) by region (6 levels) by valence (3 levels: positive, negative, neutral), as presented in Fig. 4. A Mauchly sphericity test was significant (p o0.01); thus Greenhouse-Geisser correction was applied. The triple interaction group region valence was significant (F(10, 260)¼ 4.9, ε ¼0.24, p o0.01, η2p¼ 0.158). The simple two-way interaction for region emotion in each group was significant in controls (F(10, 130) ¼10.303, po 0.01, η2p ¼0.442), but not in PTSD

Fig. 2. Behavioral results. Valence ranking (A) and reaction-time (B) to positive, negative, and neutral stimuli. No significant differences were found between groups in any of the compared variables.



5

Left Right Midline Front Midline Back Frontal Fronto-central Central Centro-parietal Parietal Temporal Parieto-occipital Occipital Current density scale Fig. 3. ERP analysis of emotional response in control and PTSD subjects. Grand averaged ERPs over left frontal and occipital electrodes for control (A) and PTSD (B) in response to positive, negative, and neutral stimuli. (C) and (D) represent the results of repeated measures ANOVA per time frame across all electrodes; black rectangles indicate a significant result at p o0.01 in at least 8 consecutive time frames. (E) Source localization of ERP components identified as significantly differing according to emotional valence in control subjects.

patients (F(10, 130) ¼0.233, p 40.05). To determine emotional modulation of the P300 component in healthy controls, a comparison was performed between ERPs by valence across all regions. For negative versus neutral (F(1, 26) ¼ 7.57, p o0.01) and negative versus positive valence (F(1, 26) ¼8.36, p o0.01), there was a significant difference with higher amplitude for negative valence stimuli. No significant differences were observed for positive versus neutral stimuli (F(1, 26) ¼0.02, p 40.05). Further region-wise analysis for negative versus neutral stimuli revealed significant differences across all regions [frontal left (F(1, 26) ¼13.2, po 0.01), frontal right (F(1, 26) ¼10.4, p o0.01), frontalmidline (F(1, 26) ¼9.84, p o0.01), parieto-occipital left (F(1, 26) ¼ 8.88, p o0.01), parieto-occipital right (F(1, 26) ¼13.2, p o0.01), and parieto-occipital midline (F(1, 26) ¼ 9.05, p o0.01)]. In the analysis of negative vs. positive stimuli, there were again significant differences across all regions [frontal left (F(1, 26) ¼11.46, p o0.01), frontal right (F(1, 26) ¼8.85, p o0.01), frontal-midline (F(1, 26) ¼ 8.12, p o0.01), parieto-occipital left (F(1, 26) ¼5.16, p o0.05), parieto-occipital right (F(1, 26) ¼11.43, p o0.01), and parieto-occipital midline (F(1, 26) ¼10.29, p o0.01)]. For source-localization via sLORETA of the statistically significant differences in controls, the positive and the neutral responses were averaged and compared with the negative response. This comparison indicated that the increase in negative-induced activity in the healthy control group is in left Brodmann area (BA 40) of the parietal lobe (for the sLORETA mapping of t-values for the negative vs. the average of positive and neutral stimuli, see Fig. 4B). Thus, negative emotional stimuli led to increased P300 amplitude in the left parietal region in this group. Finally, we compared the results between controls and PTSD patients. Comparisons across all six regions revealed no significant differences between ERPs in response to each of the three categories of emotional stimuli.

3.2.2. The P600 component The second component identified was analyzed in the 590- to 630-ms post-stimulus window in a single region—fronto-central left (including electrodes F1, FC1, Fz), as this was the only region meeting the first step criteria (see above). A two-way repeated measures ANOVA was performed with the following factors: group (2 levels: control vs. PTSD) by valence (3 levels: positive, negative, neutral; see Fig. 5A). The Mauchly sphericity test was not significant, so no correction was required. The group valence interaction was significant (F(2, 52) ¼7.6, p o0.01, η2p ¼0.22). Planned comparisons were carried out between valences in control subjects to determine emotion modulation of the second component. Negative versus neutral (F(1, 26) ¼18.9, p o0.01) and positive versus neutral comparisons (F(1, 26) ¼10.4, po 0.01) were significant, reflecting a larger mean component voltage in response to neutral pictures, with a trend towards a difference in positive versus negative pictures (F(1, 26) ¼3.9, p ¼0.05) reflecting a larger response to positive pictures. Source localization of the difference in components (Fig. 5B) localized the decrease in emotional stimuli activation extensively to regions across all lobes including occipital visual regions (BA 17, 18, 19), frontal regions (BA 5, 6, 8), the parietal lobe (BA 5, 7, 40), and the temporal lobe (BA 13, 22, 39). Thus, in controls, the mid-latency second component in response to emotional stimuli is diminished in widespread brain regions when compared with the response to neutral stimuli. Comparison across emotions in the PTSD group revealed no significant effect for valence (Fig. 5A). Comparison of PTSD patients with controls by valences revealed a larger response in control subjects for neutral stimuli (F(1, 26)¼ 5.6, po 0.05), but no difference was found for negative or positive ones. Thus, control subjects show a larger response to neutral stimuli compared with PTSD subjects, while the response to negative and positive stimuli



6

The P300 component Frontal Lt

Frontal Rt

Frontal mid

Parieto-occipital Lt

Parieto-occipital Rt

Parieto-occipital mid

Control: negative vs. positive & neutral Lt

t-value

Rt

p-value .05 = t-value 1.65 Fig. 4. (A) Early component comparison across regions, valance, and group. Comparisons marked with * indicate statistically significant difference at po 0.05. (B) Source localization of increased negative valance induced activity.

is unaltered. 3.2.3. The P1000 component The late positive potential (LPP) component was analyzed in the 980–1035-ms time window in four regions (fronto-central left: F5, F7, FC3; parieto-occipital left: P1, Po1, PO3, O1; parieto-occipital right: P2, Po2, Po4, O2; parieto-occipital midline: Pz, Poz, Oz). A three-way repeated measures ANOVA was performed with the following factors: group (2 levels: control vs. PTSD) by region (4

levels) by valence (3 levels: positive, negative, neutral; see Fig. 6A). The Mauchly sphericity test was significant (po 0.01); thus, the Greenhouse-Geisser correction was applied. The triple group region valence interaction was significant after correction (F(6, 156) ¼4.9, ε ¼0.33, p o0.01, η2p ¼ 0.159). The simple two-way interaction for region emotion in each group was significant in controls (F(6, 78) ¼12.838, p o0.01, η2p ¼0.496), but not in PTSD patients (F(6, 78) ¼0.610, p 40.05). To determine emotion modulation of this component in

The P600 component Control: positive & negative vs. neutral

t-value

Fronto-central Lt

p-value .05 = t-value -2.06

Fig. 5. (A) Mid-latency component comparison across emotions and group. Comparisons marked with * indicate statistically significant difference at p o 0.05. (B) Source localization of decreased emotional stimuli induced activity.



7

The P1000 component Fronto-central Lt

Parieto-occipital Lt

Parieto-occipital Rt

Parieto-occipital mid

Negative

Positive - Neutral

Fig. 6. (A) Late component comparison across regions, emotions, and group. Comparisons marked with * indicate a statistically significant difference at po 0.05. (B) Localization of maximal activation in negative and positive-neutral mean.

healthy controls, we performed a planned comparison across emotional categories collapsed over regions. For negative versus neutral stimuli, there was a significant difference (F(1, 26)¼ 18.4, p o0.01) and there was a trend toward significance in negative versus positive stimuli (F(1, 26) ¼4.0, p ¼ 0.05); in both cases negative stimuli led to higher LPP amplitude. No significant difference was found between response to positive versus neutral stimuli (F(1, 26) ¼2.2, p 40.05). Region-wise analysis for negative versus neutral stimuli revealed significant differences for all regions, reflecting a larger response to negative stimuli [frontocentral left (F(1, 26) ¼23.0, p o0.01), parieto-occipital left (F(1, 26) ¼20.2, p o0.01), parieto-occipital right (F(1, 26) ¼23.6, p o0.01), and parieto-occipital midline (F(1, 26) ¼ 23.7, po 0.01)]. Region-wise analysis for negative versus positive stimuli revealed significant differences again for all regions, reflecting a larger response to negative stimuli [fronto-central left (F(1, 26) ¼14.8, p o0.01), parieto-occipital left (F(1, 26) ¼7.9, p o0.01), parietooccipital right (F(1, 26) ¼ 5.4, p o0.05), and parieto-occipital midline (F(1, 26) ¼10.0, p o0.01)]. Source localization of the difference between negative and the average of positive and neutral stimuli was not significant (precluding identification of the source of late negative emotion-induced brain activation). However, maximal activation in both negative and the mean of positive and neutral stimuli was in the parietal lobe (BA 7), with higher current density in the negative condition, suggesting excess negative emotion induced LPP localized to this region (Fig. 6B). In PTSD subjects, there was no main effect of valence collapsed across regions. Comparison of this LPP component between PTSD and control subjects per emotional category, collapsed across regions, did not identify a significant difference. However regionwise analysis revealed significant differences between groups for positive stimuli in the fronto-central left region (F(1, 26) ¼8.5, p o0.01) and for neutral stimuli in the arieto-occipital left region (F(1, 26)¼ 4.5, p o0.05). The main direction of difference was toward higher amplitudes in the LPP component in the PTSD group, decreasing the gap between ERPs to the negative valence stimuli and ERPs to the positive/neutral ERP.

4. Discussion In this study, we computed ERPs in response to visual, nontrauma-related, emotional stimuli in PTSD patients and healthy controls. Self-report ranking measured following EEG recordings indicated that both controls and PTSD patients perceived the pictures similarly on the scale of valence. RT measured during EEG recordings showed a slower response to negative stimuli in both groups; RT in PTSD patients was significantly slower, irrespective of emotional content. Notably, RT in PTSD patients to neutral stimuli resembled that of controls to negative stimuli. ERP analysis in controls revealed three distinctive components, all between 300 and 1000 ms after picture presentation, found to distinguish between pictures with different emotional valences. The first component, 310–350 ms following stimulus onset, was associated with alterations in left parietal regions (BA 40) and is consistent with the previously described P300 increase in amplitude to negative valence (Delplanque et al., 2006; Johnson et al., 2013; also, see reviews by Radilova, 1982; Johnston and Wang, 1991; Olofsson et al., 2008). The second component, identified at 600 ms, was reduced in amplitude during emotional (negative and positive) stimuli. Alterations in activity between the emotional and nonemotional conditions in this component were localized diffusely through large regions of the cerebral cortex. The third component at 1000 ms, previously associated with higher amplitudes for both positive and negative emotional pictures as compared with neutral pictures (Schupp et al., 2000), was larger during negative stimuli and localized to the parietal lobe, consistent with the reported late positive potential (Pastor et al., 2008). ERPs in the PTSD group resembled those recorded in controls during negative stimuli. Importantly, no measurable differences in brain activity were found in response to pictures with different valences in patients. P300 is known to be associated with processes involving attention, working memory, and executive control (Kok, 2001). Our ERP results may suggest that PTSD patients allocate equally increased attentional resources to all valences, and not only to negative trauma-related material as was suggested by other studies (Attias et al., 1996; Metzger et al. 1997a, b; Stanford et al., 2001).


8


Cognitively, this could represent a general “negative-signature”, i.e., the processing of positive/neutral stimuli is more likely to imitate a negative stimulus even when perceived as positive (as seems to be the case upon “off-line” ranking). These results are in line with ERP studies showing similar P200 components across all emotional categories in high anxiety individuals (Eldar et al., 2010). Similarly, EPN to positive pictures was increased in amplitude under stressful manipulation in healthy controls (Bublatzky et al., 2010). This may suggest that increased anxiety or lower threat/gain threshold (Mathews and MacLeod, 2005), often characterizing patients with PTSD, underlies the altered response found in our study. An alternative/complementary hypothesis is a possible failure of brain modulatory mechanisms to differentiate between emotional valences that stem in anomalous functional connectivity between emotionally driven limbic (e.g. the amygdala) and regulative regions (e.g., medial and ventro-medial pre-frontal cortices). Recent functional magnetic resonance imaging (fMRI) studies support this view, showing rapid/robust activation of the amygdala in response to emotional stimuli, and diminished prefrontal responsiveness, resultant in disrupted amygdala-prefrontal cortex functional circuitry (Shin et al., 2005; Stevens et al., 2013). An additional support for this view comes from our recent eventrelated synchronization study showing highly correlated, emotionrelated activity between frontal-theta to parietal-beta bands in controls, but not in PTSD patients (Cohen et al., 2013). Covey et al. (2013) showed that trauma-exposed police officers exhibited greater P300 amplitude for the Go/No-Go and non-target trials when compared with controls. Additionally, PTSD symptom severity in this group was associated with fronto-central (but not posterior) NoGo P300 amplitude. These findings provide evidence of heightened attention and/or arousal in trauma-exposed individuals, as indicated by the generally greater P300 amplitude, during a task requiring sustained attention and inhibitory control. Thus, greater symptom severity in trauma-exposed individuals may affect frontal cognitive control systems related to response inhibition. Structural and functional changes in white-matter structures that link neocortical with subcortical and limbic networks may indeed result in loss of inhibitory control (see review by Williamson et al., 2013). Applied to our findings, loss of inhibitory control can cause changes in response threshold, resultant in impaired brain response to all stimulus categories. Indeed, numerous studies reported changes in white-matter tracts in individuals with PTSD (Fani et al., 2012; Sanjuan et al., 2013; Sun et al., 2013; Williamson et al., 2013; Yeh et al., 2013; Saar-Ashkenazy et al., 2014). Recently, animal studies have provided insight into the possible cellular and molecular basis of white-matter alterations, reporting cortisolmediated increased oligodendrogenesis (and decreased neurogenesis) in the dentate gyrus of the adult rat hippocampus (Chetty et al., 2014). Together, these findings suggest a novel model in which acute stress promotes specific alterations in white matter structure and function, as the basis of impaired processing of negative stimuli in PTSD (see also Shin et al., 2005; Wessa and Flor, 2007; Ashley et al., 2013). It is the second component we describe ( 600 ms post-stimulus onset), previously associated with syntactic/semantic anomalies in language (van Herten et al., 2005), which is of special interest with regard to healthy controls, showing a decrease in cortical activation to emotional (negative and positive as compared with neutral) stimuli (Kemp et al., 2004). The relatively wide spatial distribution of emotion-related deactivation is suggestive of the activation of a diffuse modulatory neurotransmitter system —cholinergic, adrenergic or serotoninergic. For example, increased activation of the serotoninergic or adrenergic systems (using reuptake inhibitors) reduced the processing of negative emotional

input in healthy controls (Harmer et al., 2004). In this respect, it is interesting to note that changes in diffuse modulatory systems have been suggested in patients with PTSD (Southwick et al., 1993; Kaufer et al., 1998; Liberzon et al., 1999; see also reviews by Southwick et al., 1999; Hageman et al., 2001; Vermetten and Bremner, 2002; Strawn and Geracioti, 2008). A more detailed analysis is possible in animal models, indeed showing that cortical brain circuits are “over-sensitized” to ACh 1 month following severe stress (e.g., Pavlovsky et al., 2012). Such a change, if it exists in patients, may explain the lack of differential response across valences we report in PTSD patients. The clinical significance of similar cortical responses across valence categories is not known. It is tempting to hypothesize that such overgeneralization may underlie attentional deficits often reported in PTSD patients, as even “insignificant” stimuli ( i.e., neutral stimuli) affect cortical response, thus also affecting the processing of subsequent stimuli. This view is consistent with the P500, suggested to be involved in updating working memory representations (Javanbakht et al., 2011), reported to be impaired in patients with PTSD (Zhang et al., 2013). Several limitations of this study must be acknowledged. First, our sample was small and heterogeneous in the nature of traumatic events; therefore, we may have failed to detect other differences between the groups that may require a larger sample size. Another limitation lies in the “within-subject” analysis across valences. While block order was randomized across subjects, it is still possible that viewing of negative valence pictures served as a “negative primer” affecting the processing of subsequent nonnegative pictures. This could have potentially been overcome by a between-subject design in which each subject viewed only a single valence category; however, this type of analysis has its obvious drawbacks as well. Another issue to be considered here is that patients were taking psychotropic medications; therefore, we cannot rule out medication effects in our findings. Also it is known that cognitive alterations, and especially attentional deficits, are observed in PTSD patients; the time devoted to each measurement in the current study was relatively short, which also explains the low trial count in the emotional paradigm. More importantly, emotional processing, reactivity and regulation deficits in PTSD have been linked to functional brain alteration in both cortical (e.g., the ventrolateral and medial prefrontal cortex and the anterior cingulate cortex), as well as subcortical (e.g., the amygdala and hippocampus) structures (e.g., see reviews by Nutt and Malizia, 2004; Shin et al., 2006; Etkin and Wager, 2007; Liberzon and Sripada, 2008; Brown and Morey, 2012). Modulation of cortical sensory and associative activity by emotional valence is thought to depend on an initial identification of valence taking place in the amygdala. While most studies have used fMRI paradigms to identify the activation of amygdala in response to emotional stimuli (e.g., Hendler et al., 2003; Shin et al., 2005; Stevens et al., 2013), the poor time resolution of this method limits the possibility to identify alterations in the timing of amygdala activation. The use of EEG, as in the present study, allows a better time resolution compared with that of fMRI, but is limited in detection of activation in the amygdala due to the amygdala’s spherical structure and deep location. Future studies should use combined methodologies (e.g., EEG-fMRI) to achieve a more precise understanding of the neural correlates of abnormal emotional responsiveness seen in PTSD. To summarize, we report the absence of differential brain response to varying emotionally laden, non-trauma-related stimuli in PTSD patients. While the underlying brain pathology is not yet known, the data suggest cortical malfunction in discriminating emotional versus non-emotional stimuli, and “overgeneralization” of environmental stimuli as negative ones. This sensitization may reflect a reduced capacity to discriminate between non-threat and



generalized threat stimuli in PTSD (Felmingham et al., 2003), and may serve as a basis underlying hypervigilance, attentional deficits and cognitive difficulties - all known as core symptoms in PTSD.

5. Author contributions Rotem Saar-Ashkenazy, Dr. Jonathan E. Cohen and Prof. Alon Friedman had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Dr. Jonathan E. Cohen, Prof. Alon Friedman and Dr. Hadar Shalev. Recruitment of patients: Dr. Hadar Shalev. Acquisition of data: Dr. Jonathan Cohen. Analysis and interpretation of data: Rotem Saar-Ashkenazy, Dr. Jonathan E. Cohen, Dr. Jonathan Guez and Prof. Alon Friedman. Drafting of the manuscript: Rotem Saar-Ashkenazy, Dr. Jonathan E. Cohen, Magdalena K. Kanthak and Prof. Alon Friedman.

Acknowledgements This study was supported by the Israel Science Foundation (BIKURA Program, to Prof. Alon Friedman and Prof. Talma Hendler), the DFG (DFG-Trilateral Program) and the Ministry of Health (MOH, to Prof. Alon Friedman and Dr. Hadar Shalev).

References American Psychiatric Association, 2013. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Arlington, VA. Araki, T., Kasai, K., Yamasue, H., Kato, N., Kudo, N., Ohtani, T., Nakagome, K., Kirihara, K., Yamada, H., Abe, O., Iwanami, A., 2005. Association between lower P300 amplitude and smaller anterior cingulate cortex volume in patients with posttraumatic stress disorder: a study of victims of Tokyo subway sarin attack. Neuroimage 25, 43–50. http://dx.doi.org/10.1016/j.neuroimage.2004.11.039. Ashley, V., Honzel, N., Larsen, J., Justus, T., Swick, D., 2013. Attentional bias for trauma-related words: exaggerated emotional Stroop effect in Afghanistan and Iraq war veterans with PTSD. BMC Psychiatry 13, 13–86. http://dx.doi.org/ 10.1186/1471-244X-13-86. Attias, J., Bleich, A., Furman, V., Zinger, Y., 1996. Event-related potentials in posttraumatic stress disorder of combat origin. Biol. Psychiatry 40, 373–381. http: //dx.doi.org/10.1016/0006-3223(95)00419-X. Blomhoff, S., Reinvang, I., Malt, U.F., 1998. Event-related potentials to stimuli with emotional impact in posttraumatic stress patients. Biol. Psychiatry 44, 1045–1053. http://dx.doi.org/10.1016/S0006-3223(98)00058-4. Bradley, M.M., Lang, P.J., 1994. Measuring emotion: the self-assessment manikin and the semantic differential. J. Behav. Ther. Exp. Psychiatry 25, 49–59. http: //dx.doi.org/10.1016/0005-7916(94)90063-9. Brown, V.M., Morey, R.A., 2012. Neural systems for cognitive and emotional processing in posttraumatic stress disorder. Front. Psychol. 3. http://dx.doi.org/ 10.3389/fpsyg.2012.00449 449-449. Bublatzky, F., Flaisch, T., Stockburger, J., Schmalzle, R., Schupp, H.T., 2010. The interaction of anticipatory anxiety and emotional picture processing: an eventrelated brain potential study. Psychophysiology 47, 687–696. http://dx.doi.org/ 10.1111/j.1469-8986.2010.00966.x. Charles, G., Hansenne, M., Ansseau, M., Pitchot, W., Machowski, R., Schittecatte, M., Wilmotte, J., 1995. P300 in posttraumatic stress disorder. Neuropsychobiology 32, 72–74. http://dx.doi.org/10.1159/000119216. Chetty, S., Friedman, A.R., Taravosh-Lahn, K., Kirby, E.D., Mirescu, C., Guo, F., Krupik, D., Nicholas, A., Geraghty, C., Krishamurthy, A., Tsai, M.K., Covarrubias, D., Wong, A.T., Francis, D.D., Sapolsky, R.M., Palmer, T.D., Pleasure, D., Kaufer, D., 2014. Stress and glucocorticoids promote oligodendrogenesis in the adult hippocampus. Mol. Psychiatry 11, 1–9. http://dx.doi.org/10.1038/mp.2013.190. Cohen, J.E., Shalev, H., Admon, R., Hefetz, S., Gasho, C.J., Shachar, L.J., Admon, R., Hefetz, S., Gasho, C.J., Shachar, L.J., Shelef, I., Hendler, T., Friedman, A., 2013. Emotional brain rhythms and their impairment in post-traumatic patients. Hum. Brain Mapp. 34, 1344–1356. http://dx.doi.org/10.1002/hbm.21516. Covey, T.J., Shucard, J.L., Violanti, J.M., Lee, J., Shucard, D.W., 2013. The effects of exposure to traumatic stressors on inhibitory control in police officers: a dense electrode array study using a Go/NoGo continuous performance task. Int. J. Psychophysiol. 87, 363–375 0.1016/j.ijpsycho.2013.03.009. Delplanque, S., Silvert, L., Hot, P., Rigoulot, S., Sequeira, H., 2006. Arousal and valence effects on event-related P3a and P3b during emotional categorization. Int.

9

J. Psychophysiol. 60, 315–322. http://dx.doi.org/10.1016/j.ijpsycho.2005.06.006. Eldar, S., Yankelevitch, R., Lamy, D., Bar-Haim, Y., 2010. Enhanced neural reactivity and selective attention to threat in anxiety. Biol. Psychol. 85, 252–257. http: //dx.doi.org/10.1016/j.biopsycho.2010.07.010. Etkin, A., Wager, T.D., 2007. Functional neuroimaging of anxiety: a meta-analysis of emotional processing in PTSD, social anxiety disorder, and specific phobia. Am. J. Psychiatry 164, 1476–1488. http://dx.doi.org/10.1176/appi.ajp.2007.07030504. Fani, N., King, T.Z., Jovanovic, T., Glover, E.M., Bradley, B., Choi, K., Ely, T., Gutman, D. A., Ressler, K.J., 2012. White matter integrity in highly traumatized adults with and without post-traumatic stress disorder. Neuropsychopharmacology 37, 2740–2746. http://dx.doi.org/10.1038/npp.2012.146. Felmingham, K.L., Bryant, R.A., Gordon, E., 2003. Processing angry and neutral faces in post-traumatic stress disorder: an event-related potentials study. Neuroreport 14, 777–780. http://dx.doi.org/10.1097/00001756-200304150-00024. Felmingham, K.L., Bryant, R.A., Kendall, C., Gordon, E., 2002. Event-related potential dysfunction in posttraumatic stress disorder: the role of numbing. Psychiatry Res. 109, 171–179. http://dx.doi.org/10.1016/S0165-1781(02)00003-3. Ghisolfi, E.S., Margis, R., Becker, J., Zanardo, A.P., Strimitzer, I.M., Lara, D.R., 2004. Impaired P50 sensory gating in post-traumatic stress disorder secondary to urban violence. Int. J. Psychophysiol. 51, 209–214. http://dx.doi.org/10.1016/j. ijpsycho.2003.09.002. Gillette, G.M., Skinner, R.D., Rasco, L.M., Fielstein, E.M., Davis, D.H., Pawelak, J.E., Freeman, T.W., Karson, C.N., Boop, F.A., Garcia-Rill, E., 1997. Combat veterans with posttraumatic stress disorder exhibit decreased habituation of the P1 midlatency auditory evoked potential. Life Sci. 61, 1421–1434. http://dx.doi.org/ 10.1016/S0024-3205(97)00688-7. Gjini, K., Boutros, N.N., Haddad, L., Aikins, D., Javanbakht, A., Amirsadri, A., Tancer, M.E., 2013. Evoked potential correlates of post-traumatic stress disorder in refugees with history of exposure to torture. J. Psychiatr. Res. 47, 1492–1498. http://dx.doi.org/10.1016/j.jpsychires.2013.06.007. Hageman, I., Andersen, H.S., Jorgensen, M.B., 2001. Post-traumatic stress disorder: a review of psychobiology and pharmacotherapy. Acta Psychiatr. Scand. 104, 411–422. http://dx.doi.org/10.1034/j.1600-0447.2001.00237.x. Handy, T.C., 2005. Event-related Potentials: A Methods Handbook. MIT Press, MA. Harmer, C.J., Shelley, N.C., Cowen, P.J., Goodwin, G.M., 2004. Increased positive versus negative affective perception and memory in healthy volunteers following selective serotonin and norepinephrine reuptake inhibition. Am. J. Psychiatry 161, 1256–1263. http://dx.doi.org/10.1176/appi.ajp.161.7.1256. Hendler, T., Rotshtein, P., Yeshurun, Y., Weizmann, T., Kahn, I., Ben-Bashat, D., Malach, R., Bleich, A., 2003. Sensing the invisible: differential sensitivity of visual cortex and amygdala to traumatic context. Neuroimage 19, 587–600. http://dx. doi.org/10.1016/S1053-8119(03)00141-1. Holstein, D.H., Vollenweider, F.X., Jancke, L., Schopper, C., Csomor, P.A., 2010. P50 suppression, prepulse inhibition, and startle reactivity in the same patient cohort suffering from posttraumatic stress disorder. J. Affect. Disord. 126, 188–197. http://dx.doi.org/10.1016/j.jad.2010.02.122. Javanbakht, A., Liberzon, I., Amirsadri, A., Gjini, K., Boutros, N.N., 2011. Event-related potential studies of post-traumatic stress disorder: a critical review and synthesis. Biol. Mood Anxiety Disord. 1, 2045–5380. http://dx.doi.org/10.1186/ 2045-5380-1-5. Johnson, J.D., Allana, T.N., Medlin, M.D., Harris, E.W., Karl, A., 2013. Meta-analytic review of P3 components in posttraumatic stress disorder and their clinical utility. Clin. EEG Neurosci. 44, 112–134. http://dx.doi.org/10.1177/ 1550059412469742. Johnston, V.S., Wang, X.T., 1991. The relationship between menstrual phase and the P3 component of ERPs. Psychophysiology 28, 400–409. http://dx.doi.org/ 10.1111/j.1469-8986.1991.tb00723.x. Kaufer, D., Friedman, A., Seidman, S., Soreq, H., 1998. Acute stress facilitates longlasting changes in cholinergic gene expression. Nature 393, 373–377. http://dx. doi.org/10.1038/30741. Kemp, A.H., Gray, M.A., Silberstein, R.B., Armstrong, S.M., Nathan, P.J., 2004. Augmentation of serotonin enhances pleasant and suppresses unpleasant cortical electrophysiological responses to visual emotional stimuli in humans. Neuroimage 22, 1084–1096. http://dx.doi.org/10.1016/j.neuroimage.2004.03.022. Kimble, M., Kaloupek, D., Kaufman, M., Deldin, P., 2000. Stimulus novelty differentially affects attentional allocation in PTSD. Biol. Psychiatry 47, 880–890. http://dx.doi.org/10.1016/S0006-3223(99)00258-9. Kok, A., 2001. On the utility of the P3 amplitude as a measure of processing capacity. Psychophysiology 38, 557–577. http://dx.doi.org/10.1017/ S0048577201990559. Liberzon, I., Abelson, J.L., Flagel, S.B., Raz, J., Young, E.A., 1999. Neuroendocrine and psychophysiologic responses in PTSD: a symptom provocation study. Neuropsychopharmacology 21, 40–50. http://dx.doi.org/10.1016/S0893-133X(98) 00128-6. Liberzon, I., Sripada, C.S., 2008. The functional neuroanatomy of PTSD: a critical review. Prog. Brain Res. 167, 151–169. http://dx.doi.org/10.1016/S0079-6123 (07)67011-3 10.1016/S0079-6123(07)67011-3.. Mathews, A., MacLeod, C., 2005. Cognitive vulnerability to emotional disorders. Annu. Rev. Clin. Psychol. 1, 167–195. http://dx.doi.org/10.1146/annurev. clinpsy.1.102803.143916. McFarlane, A.C., Weber, D.L., Clark, C.R., 1993. Abnormal stimulus processing in posttraumatic stress disorder. Biol. Psychiatry 34, 311–320. http://dx.doi.org/ 10.1016/0006-3223(93)90088-U. McPherson, W.B., Newton, J.E., Ackerman, P., Oglesby, D.M., Dykman, R.A., 1997. An event-related brain potential investigation of PTSD and PTSD symptoms in abused children. Integr. Physiol. Behav. Sci. 32, 31–42. http://dx.doi.org/


10


10.1007/BF02688611. Metzger, L.J., Carson, M.A., Paulus, L.A., Lasko, N.B., Paige, S.R., Pitman, R.K., Orr, S.P., 2002. Event-related potentials to auditory stimuli in female Vietnam nurse veterans with posttraumatic stress disorder. Psychophysiology 391, 49–63. http://dx.doi.org/10.1111/1469-8986.3910049. Metzger, L.J., Orr, S.P., Lasko, N.B., Berry, N.J., Pitman, R.K., 1997a. Evidence for diminished P3 amplitudes in PTSD. Ann. N. Y. Acad. Sci. 821, 499–503. http://dx. doi.org/10.1111/j.1749-6632.1997.tb48315.x. Metzger, L.J., Orr, S.P., Lasko, N.B., McNally, R.J., Pitman, R.K., 1997b. Seeking the source of emotional Stroop interference effects in PTSD: a study of P3s to traumatic words. Integr. Physiol. Behav. Sci. 32, 43–51. http://dx.doi.org/ 10.1007/BF02688612. Nesse, R.M., Ellsworth, P.C., 2009. Evolution, emotions, and emotional disorders. Am. Psychol. 64, 129–139. http://dx.doi.org/10.1037/a0013503. Neylan, T.C., Fletcher, D.J., Lenoci, M., McCallin, K., Weiss, D.S., Schoenfeld, F.B., Marmar, C.R., Fein, G., 1999. Sensory gating in chronic posttraumatic stress disorder: reduced auditory P50 suppression in combat veterans. Biol. Psychiatry 46, 1656–1664. http://dx.doi.org/10.1016/S0006-3223(99)00047-5. Nutt, D.J., Malizia, A.L., 2004. Structural and functional brain changes in posttraumatic stress disorder. J. Clin. Psychiatry 65, 11–17. Olofsson, J.K., Nordin, S., Sequeira, H., Polich, J., 2008. Affective picture processing: an integrative review of ERP findings. Biol. Psychol. 77, 247–265. http://dx.doi. org/10.1016/j.biopsycho.2007.11.006. Paige, S.R., Reid, G.M., Allen, M.G., Newton, J.E., 1990. Psychophysiological correlates of posttraumatic stress disorder in Vietnam veterans. Biol. Psychiatry 27, 419–430. http://dx.doi.org/10.1016/0006-3223(90)90552-D. Pascual-Marqui, R.D., 2002. Standardized low-resolution brain electromagnetic tomography (sLORETA): technical details. Methods Find. Exp. Clin. Pharmacol. 24, 5–12, Epub 2003/02/11. Pastor, M.C., Bradley, M.M., Low, A., Versace, F., Molto, J., Lang, P.J., 2008. Affective picture perception: emotion, context, and the late positive potential. Brain Res. 16, 145–151. http://dx.doi.org/10.1016/j.brainres.2007.10.072. Pavlovsky, L., Bitan, Y., Shalev, H., Serlin, Y., Friedman, A., 2012. Stress-induced altered cholinergic-glutamatergic interactions in the mouse hippocampus. Brain Res. 7, 99–106. http://dx.doi.org/10.1016/j.brainres.2012.05.057. Radilova, J., 1982. The late positive component of visual evoked response sensitive to emotional factors. Act. Nerv. Super. 3, 334–337. Saar-Ashkenazy, R., Cohen, J., Guez, J., Gasho, C., Shelef, I., Friedman, A., Shalev, H., 2014. Reduced corpus-callosum volume in posttraumatic stress disorder highlights the importance of interhemispheric connectivity for associative memory. J. Trauma. Stress 27, 18–26. http://dx.doi.org/10.1002/jts.21887. Sanjuan, P.M., Thoma, R., Claus, E.D., Mays, N., Caprihan, A., 2013. Reduced white matter integrity in the cingulum and anterior corona radiata in posttraumatic stress disorder in male combat veterans: a diffusion tensor imaging study. Psychiatry Res.: Neuroimaging 214, 260–268. http://dx.doi.org/10.1016/j. pscychresns.2013.09.002. Schupp, H.T., Cuthbert, B.N., Bradley, M., Cacioppo, J.T., Ito, T., Lang, P.J., 2000. Affective picture processing: the late positive potential is modulated by motivational relevance. Psychophysiology 37, 257–261. http://dx.doi.org/10.1111/ 1469-8986.3720257. Schupp, H.T., Junghofer, M., Weike, A.I., Hamm, A.O., 2003. Emotional facilitation of sensory processing in the visual cortex. Psychol. Sci. 14, 7–13. http://dx.doi.org/ 10.1111/1467-9280.01411. Schupp, H.T., Stockburger, J., Codispoti, M., Junghofer, M., Weike, A.I., Hamm, A.O., 2006. Stimulus novelty and emotion perception: the near absence of habituation in the visual cortex. Neuroreport 17, 365–369. http://dx.doi.org/10.1016/ S0079-6123(06)56002-9. Shin, L.M., Rauch, S.L., Pitman, R.K., 2006. Amygdala, medial prefrontal cortex, and hippocampal function in PTSD. Ann. N. Y. Acad. Sci. 1071, 67–79. http://dx.doi. org/10.1196/annals.1364.007. Shin, L.M., Wright, C.I., Cannistraro, P.A., Wedig, M.M., McMullin, K., Martis, B., Macklin, M.L., Lasko, N.B., Cavanagh, S.R., Krangel, T.S., Orr, S.P., Pitman, R.K., Whalen, P.J., Rauch, S.L., 2005. A functional magnetic resonance imaging study of amygdala and medial prefrontal cortex responses to overtly presented fearful

faces in posttraumatic stress disorder. Arch. Gen. Psychiatry 62, 273–281. http: //dx.doi.org/10.1146/annurev.clinpsy.1.102803.143916. Shucard, J.L., McCabe, D.C., Szymanski, H., 2008. An event-related potential study of attention deficits in posttraumatic stress disorder during auditory and visual Go/NoGo continuous performance tasks. Biol. Psychol. 79, 223–233. http://dx. doi.org/10.1016/j.biopsycho.2008.05.005. Skinner, R.D., Rasco, L.M., Fitzgerald, J., Karson, C.N., Matthew, M., Williams, D.K., Garcia-Rill, E., 1999. Reduced sensory gating of the P1 potential in rape victims and combat veterans with posttraumatic stress disorder. Depress. Anxiety 9, 122–130. http://dx.doi.org/10.1016/S0006-3223(99)00047-5. Southwick, S.M., Bremner, J.D., Rasmusson, A., Morgan, C.A., Arnsten, A., Charney, D. S., 1999. Role of norepinephrine in the pathophysiology and treatment of posttraumatic stress disorder. Biol. Psychiatry 46, 1192–1204. http://dx.doi.org/ 10.1016/S0006-3223(99)00219-X. Southwick, S.M., Krystal, J.H., Morgan, C.A., Johnson, D., Nagy, L.M., Nicolaou, A., Heininger, G.R., Charney, D.S., 1993. Abnormal noradrenergic function in posttraumatic stress disorder. Arch. Gen. Psychiatry 50, 266–274. http://dx.doi. org/10.1001/archpsyc.1993.01820160036003. Stanford, M.S., Vasterling, J.J., Mathias, C.W., Constans, J.I., Houston, R.J., 2001. Impact of threat relevance on P3 event-related potentials in combat-related posttraumatic stress disorder. Psychiatry Res. 102, 125–137. http://dx.doi.org/ 10.1016/S0165-1781(01)00236-0. Stevens, J.S., Jovanovic, T., Fani, N., Ely, T.D., Glover, E.M., Bradley, B., Ressler, K.J., 2013. Disrupted amygdala-prefrontal functional connectivity in civilian women with posttraumatic stress disorder. J. Psychiatric Res. 47, 1469–1478. http://dx. doi.org/10.1016/j.jpsychires.2013.05.031. Strawn, J.R., Geracioti Jr., T.D., 2008. Noradrenergic dysfunction and the psychopharmacology of posttraumatic stress disorder. Depress. Anxiety 25, 260–271. http://dx.doi.org/10.1002/da.20292. Sun, Y., Wang, Z., Ding, W., Wan, J., Zhuang, Z., Zhang, Y., Liu, Y., Zhou, Y., Xu, J., 2013. Alterations in white matter microstructure as vulnerability factors and acquired signs of traffic accident-induced PTSD. PLoS One 8, e83473. http://dx.doi.org/ 10.1371/journal.pone.0083473. van Herten, M., Kolk, H.H.J., Chwilla, D.J., 2005. An ERP study of P600 effects elicited by semantic anomalies. Cognit. Brain Res. 22, 241–255. http://dx.doi.org/ 10.1016/j.cogbrainres.2004.09.002. Veltmeyer, M.D., Clark, C.R., McFarlane, A.C., Felmingham, K.L., Bryant, R.A., Gordon, E., 2005. Integrative assessment of brain and cognitive function in post-traumatic stress disorder. J. Integr. Neurosci. 4, 145–159. http://dx.doi.org/10.1142/ S0219635205000719. Vermetten, E., Bremner, J.D., 2002. Circuits and systems in stress. II. Applications to neurobiology and treatment in posttraumatic stress disorder. Depress. Anxiety 16, 14–38. Weathers, F.W., Keane, T.M., Davidson, J.R., 2001. Clinician-administered PTSD scale: a review of the first ten years of research. Depress. Anxiety 13, 132–156. http://dx.doi.org/10.1002/da.1029. Wessa, M., Flor, H., 2007. Failure of extinction of fear responses in posttraumatic stress disorder: evidence from second-order conditioning. Am. J. Psychiatry 164, 1684–1692. http://dx.doi.org/10.1176/appi.ajp.2007.07030525. Williamson, J.B., Heilman, K.M., Porges, E.C., Lamb, D.G., Porges, S.W., 2013. A possible mechanism for PTSD symptoms in patients with traumatic brain injury: central autonomic network disruption. Front. Neuroengin. 6, 13. http://dx. doi.org/10.3389/fneng.2013.00013. Yeh, P.H., Wang, B., Oakes, T.R., French, L.M., Pan, H., Graner, J., Liu, W., Riedy, G., 2013. Postconcussional disorder and PTSD symptoms of military-related traumatic brain injury associated with compromised neurocircuitry. Hum. Brain Mapp. 13, 22358. http://dx.doi.org/10.1002/hbm.22358. Zhang, J.N., Xiong, K.L., Qiu, M.G., Zhang, Y., Xie, B., Wang, J., Li, M., Chen, H., Zhang, J.J., 2013. Negative emotional distraction on neural circuits for working memory in patients with posttraumatic stress disorder. Brain Res. 19, 94–101. http://dx. doi.org/10.1016/j.brainres.2013.07.042.


Atypical visual processing in posttraumatic stress disorder.

Emotional Processing in Individuals with Substance Use Disorder and Posttraumatic Stress Disorder.

Aberrant Neural Connectivity during Emotional Processing Associated with Posttraumatic Stress.

Altered one-carbon metabolism in posttraumatic stress disorder.

Impaired allocentric spatial processing in posttraumatic stress disorder.

Naloxone-reversible analgesic response to combat-related stimuli in posttraumatic stress disorder. A pilot study.

Diagnostic potential of multimodal neuroimaging in posttraumatic stress disorder.

Levels of processing for visual stimuli in an "extinguished" field.

Patients with posttraumatic stress disorder exhibit an altered phenotype of regulatory T cells.

Processing of emotional faces in congenital amusia: An emotional music priming event-related potential study.

Altered spontaneous neuronal activity of visual cortex and medial anterior cingulate cortex in treatment-naïve posttraumatic stress disorder.

Posttraumatic stress disorder.

Altered functional connectivity in posttraumatic stress disorder with versus without comorbid major depressive disorder: a resting state fMRI study.

Pharmacotherapy of posttraumatic stress disorder.

Alterations in attention capture to auditory emotional stimuli in job burnout: an event-related potential study.

Age As Moderator of Emotional Stroop Task Performance in Posttraumatic Stress Disorder (PTSD).

Sensory Symptoms and Processing of Nonverbal Auditory and Visual Stimuli in Children with Autism Spectrum Disorder.

hyperactivity disorder: an event related potential-study.

Enhanced association between perceptual stimuli and trauma-related information in individuals with posttraumatic stress disorder symptoms.

Posttraumatic Stress Disorder: The Misappropriation of Military Suicide Causation and Medication Treatment of Posttraumatic Stress Disorder.

Post-Decision Wagering Affects Metacognitive Awareness of Emotional Stimuli: An Event Related Potential Study.

Neural Processing of Emotional Musical and Nonmusical Stimuli in Depression.

Ketamine for posttraumatic stress disorder.

Psychosis-induced posttraumatic stress disorder.