Introduction
People with schizophrenia often demonstrate diminished emotional expression and engagement in pleasurable activities on clinical assessments of negative symptoms (Blanchard et al. Reference Blanchard, Kring, Horan and Gur2011). However, laboratory studies using affective science methods have revealed that several components of emotion seem remarkably intact in schizophrenia. On tasks involving exposure to emotionally evocative stimuli (e.g. pictures, films, foods), people with schizophrenia report feeling pleasant emotions as strongly as unaffected people (Cohen & Minor, Reference Cohen and Minor2008). Similarly, they report feeling unpleasant emotions as strongly as, or sometimes more strongly than, people without schizophrenia. These intact responses extend beyond self-reported emotions as patients show normal emotion-modulated cardiovascular, electrodermal and startle eyeblink responses to evocative stimuli (Kring & Moran, Reference Kring and Moran2008).
People with schizophrenia also show largely normal event-related potentials (ERPs) during even the initial stages of processing emotional stimuli. A series of studies has focused on emotional modulation of the late positive potential (LPP), an index of motivated attention that is thought to reflect increased allocation of attentional resources to, and sustained attentional processing of, motivationally relevant stimuli (Hajcak et al. Reference Hajcak, MacNamara and Olvet2010). In healthy adults, emotional modulation of the LPP begins about 300 ms following stimulus onset at midline central–parietal sites. The LPP can persist for several seconds during picture presentation and demonstrates somewhat different characteristics during early and later time windows. For example, the P300 and the early portion of the LPP seem fairly similar in terms their temporal and spatial characteristics, and also their sensitivity to emotional stimuli. In contrast to the transient P300, the LPP is sustained at central sites and is thought to reflect more elaborated processing related to the significance and meaning of stimuli (Hajcak et al. Reference Hajcak, Weinberg, MacNamara, Foti, Luck and Kappenman2012). People with schizophrenia have generally shown normal patterns of LPP enhancement to emotional versus neutral pictures; this was true under experimental conditions in which subjects were explicitly instructed to attend to the pictures in a passive viewing paradigm and when the emotional pictures where incidental to task demands (Horan et al. Reference Horan, Wynn, Kring, Simons and Green2010, Reference Horan, Foti, Hajcak, Wynn and Green2012).
Although emotional responses to evocative stimuli seem to be largely intact in schizophrenia, adaptive functioning in daily life also requires the capacity to regulate one's emotions to meet goals and manage arousal. Emotion regulation is the process by which people influence which emotion will be experienced, when it will be experienced and how it will be experienced and expressed (Gross & Thompson, Reference Gross, Thompson and Gross2007).
Gross and colleagues distinguish between regulation strategies that are antecedent focused, that is before emotion response tendencies have become fully activated to change behavioral and physiological responding, and response focused, that is after the response tendencies have been generated (Gross, Reference Gross1998; Gross & John, Reference Gross and John2003). Cognitive reappraisal, an extensively studied antecedent-focused strategy, is a form of cognitive change that involves construing a potentially emotion-eliciting situation in a way that changes its emotional impact (Lazarus, Reference Lazarus1991; Gross, Reference Gross2002; Ochsner & Gross, Reference Ochsner and Gross2005; Hajcak & Nieuwenhuis, Reference Hajcak and Nieuwenhuis2006). For example, during a stressful job interview, an individual might down-regulate negative emotions by viewing the experience as an opportunity to find out how much they are suited to a particular work environment rather than as a test of their worth. Reappraisal may be a particularly effective emotion regulation strategy because it serves to change the fundamental meaning and personal significance of an emotional stimulus prior to an unpleasant emotional response being elicited. Contrasted with less optimal response-focused strategies, such as suppression of emotional expression (e.g. maintaining a poker face), changing the personal significance of an unpleasant stimulus is efficient and effective, with few physiological or cognitive costs (Gross, Reference Gross2002).
Several studies have investigated the neural correlates of cognitive reappraisal in healthy adults. For example, functional magnetic resonance imaging (fMRI) studies examining the impact of reappraisal instructions (e.g. ‘Reappraise’ versus ‘Attend’) on emotional responses after unpleasant stimuli are presented have identified a key neural circuit that shows increased activity in prefrontal brain areas involved in cognitive control coupled with decreased activity in the amygdala and other brain areas involved in emotional responses (Ochsner & Gross, Reference Ochsner and Gross2005). In addition, ERP paradigms have begun to shed light on the time course of appraisal manipulations.
In a series of studies, Hajcak and colleagues demonstrated that the magnitude of the LPP to unpleasant picture stimuli is sensitive to manipulations that impact how the stimuli are interpreted (Foti & Hajcak, Reference Foti and Hajcak2008; MacNamara et al. Reference MacNamara, Foti and Hajcak2009, Reference MacNamara, Ochsner and Hajcak2011). These studies involved preappraisal in which auditory descriptions were presented before unpleasant and neutral pictures that framed the images in either a more negative or a more neutral manner (Hajcak et al., in press). For example, a picture of a fierce-looking snake might be preceded by either ‘This is a poisonous snake that is very dangerous’ or ‘This is a snake that is completely harmless – it doesn't even have teeth’; the latter description is designed to down-regulate the emotional response elicited by the unpleasant image. Preappraisal with verbal descriptions removes the potential confound of task difficulty (i.e. greater effort involved in ‘Reappraise’ versus ‘Attend’ conditions in many fMRI studies), such that each condition simply involves listening to descriptions and viewing pictures. Unpleasant images preceded by descriptions that framed them in a more neutral light elicited reliably smaller LPPs than unpleasant images preceded by more negative descriptions; LPPs to unpleasant images in both conditions were larger than those found for neutral pictures preceded by neutral descriptions. Modulation of the LPP began within 300 ms after stimulus onset and was sustained for at least several hundred milliseconds during picture presentation, indicating that shifts in the meaning due to preappraisal exert an early, sustained influence on emotion processing in the healthy adult brain. These findings have been extended to young children and applied to understanding emotion regulation processes associated with vulnerability to depression and anxiety disorders (Dennis & Hajcak, Reference Dennis and Hajcak2009; Mocaiber et al. Reference Mocaiber, Pereira, Erthal, Figueira, Machado-Pinheiro, Cagy, Volchan and de Oliveira2009; DeCicco et al. Reference DeCicco, Solomon and Dennis2012). Thus, it is possible to modulate early electrocortical responses to unpleasant pictures by preappraisal and this paradigm is well suited for use in clinical populations.
The primary aim of the current study was to examine whether people with schizophrenia can modulate their electrophysiological responses (i.e. LPP) to unpleasant stimuli through preappraisal-based manipulations of stimulus meaning. We are unaware of prior studies that have used ERP paradigms to study emotion regulation in schizophrenia, although the few relevant studies using alternative methods suggest impaired emotion regulation in this population. In the most directly relevant work, a recent fMRI study found that people with schizophrenia (n = 12) showed ventrolateral prefrontal cortex (VLPFC) hypoactivation, along with decreased PFC–amygdala coupling, during down-regulation of unpleasant emotion through cognitive reappraisal (Morris et al. Reference Morris, Sparks, Mitchell, Weickert and Green2012). At a broader level, several studies indicate that people with schizophrenia demonstrate lower scores on the ‘Managing Emotions’ branch of the Mayer–Salovey–Caruso Emotional Intelligence Test (MSCEIT; Mayer et al. Reference Mayer, Salovey, Caruso and Sitarenios2003), which assesses, in part, the ability to adaptively regulate one's own emotions (e.g. Eack et al. Reference Eack, Greeno, Pogue-Geile, Newhill, Hogarty and Keshavan2008; Kee et al. Reference Kee, Horan, Salovey, Kern, Sergi, Fiske, Lee, Subotnik, Nuechterlein, Sugar and Green2009). There have also been a few studies using self-report emotion regulation questionnaires in schizophrenia, although the findings are mixed. Some studies found that patients reported lower habitual use of cognitive reappraisal (Livingstone et al. Reference Livingstone, Harper and Gillanders2009; van der Meer et al. Reference van der Meer, van't Wout and Aleman2009) or greater use of suppression (van der Meer et al. Reference van der Meer, van't Wout and Aleman2009) than healthy controls but others found no group differences on these measures (Henry et al. Reference Henry, Rendell, Green, McDonald and O'Donnell2008; Badcock et al. Reference Badcock, Paulik and Maybery2011; Perry et al. Reference Perry, Henry and Grisham2011). The available evidence led us to predict that schizophrenia patients would show diminished modulation of the LPP to unpleasant pictures preceded by more neutral than more negative descriptions as compared to matched healthy controls.
A secondary aim was to explore relationships among ERP indices from the emotion regulation task, scores on a self-report emotion regulation questionnaire assessing habitual use of cognitive reappraisal and suppression, and clinical symptoms within the patient group. The few prior studies that have examined relationships between indices of emotion regulation and symptoms are inconsistent, reporting relationships to negative symptoms (Henry et al. Reference Henry, Green, de Lucia, Restuccia, McDonald and O'Donnell2007; Kee et al. Reference Kee, Horan, Salovey, Kern, Sergi, Fiske, Lee, Subotnik, Nuechterlein, Sugar and Green2009; Perry et al. Reference Perry, Henry and Grisham2011), positive symptoms (Badcock et al. Reference Badcock, Paulik and Maybery2011) or no relationship to symptoms (Henry et al. Reference Henry, Rendell, Green, McDonald and O'Donnell2008). We therefore did not have specific directional hypotheses for these analyses.
Method
Participants
Thirty-one out-patients with schizophrenia and 27 healthy control subjects participated in this study. The schizophrenia patients were recruited from out-patient treatment clinics at the Veterans Affairs Greater Los Angeles Healthcare System (VAGLAHS) and through presentations in the community. Patients met criteria for schizophrenia based on the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I; First et al. Reference First, Gibbon, Spitzer and Williams1996). None of the patients were in a depressed or manic episode at the time of testing. Additional exclusion criteria for patients included: substance abuse or dependence in the past 6 months; IQ <70 based on chart reviews; a history of loss of consciousness for more than 1 h; an identifiable neurological disorder; or insufficient fluency in English. All patients were clinically stable as defined by: no hospitalizations in the past 3 months, no changes in living situation in the past 2 months, and no medication changes in the past 6 weeks. Patients were medicated at clinically determined dosages, with 26 receiving atypical antipsychotics, four receiving typical antipsychotics and one receiving both types of antipsychotic medication. The average antipsychotic medication dosage in chlorpromazine (CPZ) equivalent units (Andreasen et al. Reference Andreasen, Pressler, Nopoulos, Miller and Ho2010) was 269.59 (s.d. = 160.07).
Healthy controls were recruited through flyers posted in local newspapers, websites and posted advertisements. An initial screening interview excluded potential control participants with identifiable neurological disorder or head injury, those whose first-degree relatives had schizophrenia or other psychotic disorder, or participants not sufficiently fluent in English. Potential controls were screened with the SCID-I and excluded for history of schizophrenia or other psychotic disorder, bipolar disorder, recurrent depression, lifetime history of substance dependence or any substance abuse in the past 6 months. Potential controls were also administered portions of the Structured Clinical Interview for DSM-IV Axis II Disorders (SCID-II; First et al. Reference First, Spitzer, Gibbon, Williams and Benjamin1994) and excluded if they had avoidant, paranoid, schizoid or schizotypal personality disorder.
All SCID interviewers were trained through the Treatment Unit of the Department of Veterans Affairs VISN 22 Mental Illness Research, Education, and Clinical Center (MIRECC) to a minimum κ of 0.75 for key psychotic and mood items (Ventura et al. Reference Ventura, Liberman, Green and Shaner1998). All participants had the capacity to give informed consent and provided written informed consent after all procedures were fully explained in accordance with procedures approved by the Institutional Review Boards at the University of California, Los Angeles (UCLA) and the VAGLAHS.
Demographic information for both groups and clinical data for the schizophrenia group are presented in Table 1. The groups did not significantly differ in sex, age or ethnicity. The patients had lower personal education levels than controls but the groups did not differ in parental education. We intended to match the groups for parental education as a control for family socio-economic status because personal education can be influenced by the illness itself. Patients in the schizophrenia group had a typical age of onset, were chronically ill, and showed mild to moderate levels of clinical symptoms at the time of testing.
Table 1. Demographic and clinical data
BPRS, Brief Psychiatric Rating Scale; s.d., standard deviation.
* p < 0.005.
Symptom ratings
For all patients, psychiatric symptoms during the previous 2 weeks were rated using the expanded 24-item UCLA version of the Brief Psychiatric Rating Scale (BPRS; Overall & Gorham, Reference Overall and Gorham1962; Lukoff et al. Reference Lukoff, Nuechterlein and Ventura1986) by a trained rater (Ventura et al. Reference Ventura, Green, Shaner and Liberman1993). Ratings from the positive and negative symptom subscales, along with total scores, were examined (Kopelowicz et al. Reference Kopelowicz, Ventura, Liberman and Mintz2008).
ERP paradigm
Stimulus materials
The paradigm was modeled on prior ERP studies of preappraisal (Foti & Hajcak, Reference Foti and Hajcak2008; Dennis & Hajcak, Reference Dennis and Hajcak2009; MacNamara et al. Reference MacNamara, Foti and Hajcak2009). A total of 84 pictures were selected from the International Affective Picture System (IAPS; Lang et al. Reference Lang, Bradley and Cuthbert1999); of these, 28 depicted neutral scenes and 56 depicted unpleasant scenes. The two categories differed on normative [1 (unpleasant) to 9 (pleasant)] ratings of valence (mean = 5.05, s.d. = 1.21, for neutral pictures; mean = 2.82, s.d. = 1.64, for unpleasant pictures); in addition, the emotional pictures were reliably higher on normative [1 (low) to 9 (high)] arousal ratings (mean = 5.71, s.d. = 2.16, for unpleasant pictures; and mean = 2.91, s.d. = 1.93, for neutral pictures). People with schizophrenia have consistently been found to report emotional responses to IAPS stimuli that are comparable to healthy control subjects (see Cohen & Minor, Reference Cohen and Minor2008 for a meta-analysis).
For the preappraisal manipulation, a brief auditory description (using the established descriptions of Foti & Hajcak, Reference Foti and Hajcak2008; Dennis & Hajcak, Reference Dennis and Hajcak2009; MacNamara et al. Reference MacNamara, Foti and Hajcak2009) of the upcoming picture was played through speakers, prior to the presentation of the picture. Every participant heard the same neutral description for the 28 neutral images (‘Neutral’ condition). For each of the 56 unpleasant images, participants heard one of two descriptions: for 28 of the unpleasant pictures, a neutral description preceded the image (‘Preappraised neutral’ condition) whereas a negative description preceded the other 28 (‘Preappraised negative’ condition). A complete list of the IAPS image numbers and associated descriptions is presented in the Appendix.
Procedure
After a brief description of the experiment, electroencephalography (EEG) electrodes were attached. Participants were told that they would be viewing pictures and that they would hear a brief description played over the speakers prior to each picture that would describe the upcoming picture. Prior to each picture, a white fixation cross was presented on a black screen for 6 s. During this period, a brief (3–5 s) description of the upcoming picture was presented. Subjects then viewed the pictures passively; the pictures were in color, displayed for 6 s, and occupied the entire monitor (48.26 cm). The viewing distance was approximately 1 m. After each picture, there was a variable 5- to 7-s delay until the next trial. To ensure that subjects understood the task, six practice trials were administered to all subjects and were repeated with additional instruction if necessary. As in previous studies using a similar paradigm in children (Dennis & Hajcak, Reference Dennis and Hajcak2009; DeCicco et al. Reference DeCicco, Solomon and Dennis2012), self-reported emotional responses were not collected in this initial ERP study of emotion regulation in schizophrenia. After the practice trials, all participants performed the 84-task trials with breaks after every 14 trials. The order of trials and the descriptions were determined randomly for each subject. The task was administered using Eprime software (PST Technologies, USA) to control the presentation and timing of all stimuli.
EEG recording and processing
Participants had their EEG activity recorded continuously from 64 electrodes based on the 10/20 system placed in an electrode cap (Cortech Solutions, USA) and the ActiveTwo BioSemi system (BioSemi, The Netherlands). The signal was preamplified at the electrode with a gain of one; the EEG was digitized at 24-bit resolution with a sampling rate of 512 Hz with a bandpass of 0–100 Hz using a low-pass fifth-order sinc filter with a half-power cut-off of 102.4 Hz. Recordings were taken from the 64 electrodes, and also from two electrodes placed on the left and right mastoids. The electro-oculogram was recorded from four facial electrodes: two 1 cm above and below the left eye, one 1 cm to the left of the left eye, and one 1 cm to the right of the right eye. Each electrode was measured online with respect to a common mode sense electrode that formed a monopolar channel.
Off-line analysis was performed using Brain Vision Analyzer software (Brain Products, Germany). All EEG data were re-referenced to the average of all electrodes and band-pass filtered with cut-offs of 0.1 and 30 Hz. The EEG was segmented for each image beginning 200 ms before each stimulus and continuing for 2000 ms post-stimulus onset (total of 2200 ms). Each EEG segment was corrected for blinks and eye movements using the method developed by Gratton et al. (Reference Gratton, Coles and Donchin1983) . Specific channels were rejected in each trial using a semi-automated procedure, with physiological artifacts identified by the following criteria: a step of more than 50 μV between sample points, a maximum difference of less than 0.5 μV within 100-ms intervals, and an amplitude that exceeded ±75 μV. Two patients and two controls were excluded because the quality of the ERP data was poor (less than 50% artifact free trials) and one patient was excluded because of ERP values that differed by more than 3 s.d. from the patient group means. The final sample in this report consisted of 58 individuals (31 patients, 27 controls).
Based on the topographical maps from this sample, we selected a set of six electrodes (C1, C2, Cz, CPz, CP1, CP2) where the LPP was maximal for analyses. Based on prior studies (Foti & Hajcak, Reference Foti and Hajcak2008; Dennis & Hajcak, Reference Dennis and Hajcak2009) and the morphology of the waveforms from the current study, we evaluated the LPP across four time epochs: (1) 300–600 ms, (2) 600–1000 ms, (3) 1000–1500 ms and (4) 1500–2000 ms. ERPs were constructed by separately averaging segments of the three experimental conditions (Neutral, Preappraised neutral, Preappraised negative). In each case, the average activity in the 200-ms window prior to picture onset served as the baseline. The LPP was quantified as the mean activity during each epoch across the six electrode sites for each participant.
Emotion Regulation Questionnaire (ERQ)
All participants completed the ERQ (Gross & John, Reference Gross and John2003). On this 10-item questionnaire, six items focus on cognitive reappraisal (e.g. ‘When I want to feel more positive emotion, I change the way I'm thinking about the situation’) and four focus on suppression (e.g. ‘I keep my emotions to myself’). Subjects were asked to indicate to what extent the statements apply to them, scoring from 1 (strongly disagree) to 7 (strongly agree). Higher scores reflect the more frequent use of a particular strategy. The ERQ has been shown to be a reliable and valid measure of emotion regulation strategies in healthy subjects (Gross & John, Reference Gross and John2003) and in people with schizophrenia (e.g. Henry et al. Reference Henry, Rendell, Green, McDonald and O'Donnell2008; van der Meer et al. Reference van der Meer, van't Wout and Aleman2009; Perry et al. Reference Perry, Henry and Grisham2011).
Data analyses
For demographic data, group differences for continuous variables were evaluated with t tests and for categorical variables with χ2 tests. Preliminary analyses of the distributional properties of the demographic, clinical, ERP and self-report measures indicated that parametric statistical tests were appropriate. Primary analyses were conducted in three stages. First, for the ERP data, separate 3 (condition: Neutral, Preappraised neutral, Preappraised negative) × 2 (group: patient, control) repeated-measures analyses of variance (ANOVAs) were conducted for each of the four time epochs. Greenhouse–Geisser ε corrections were used for repeated-measures analyses with more than one degree of freedom. In these cases, we report the uncorrected degrees of freedom and the corrected p values. Second, group differences on the self-report emotion regulation scales were evaluated with t tests. Third, within the schizophrenia group, Pearson correlation coefficients examined associations among ERP, symptom and self-report emotion regulation variables. For the ERPs in these correlational analyses, LPP difference wave scores during each epoch for the Preappraised negative minus Preappraised neutral conditions were used as the electrophysiological index of emotion regulation.
Results
LPP results
Grand average ERPs for the control and schizophrenia groups are presented in Fig. 1Footnote †. Table 2 displays results of the repeated-measures ANOVAs along with corresponding follow-up within-group comparisons during each LPP epoch.
Fig. 1. Late positive potential (LPP) grand averages at pooled electrodes C1, Cz, C2, CP1, CPz and CP2 in Preappraised negative, Preappraised neutral, and Neutral conditions for (a) control and (b) schizophrenia groups. Stimulus onset occurred at 0 ms. Positive is plotted upward. The four epoch numbers are given in parentheses.
Table 2. Group comparisons on the late positive potential (LPP)
Scz, Schizophrenia; s.d., standard deviation.
† p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.005.
LPP during epoch 1 (300–600 ms)
There were significant condition and interaction effects during the earliest LPP epoch. For controls, preappraisal effects were evident: the LPP was significantly larger during the Preappraised negative condition compared to both the Preappraised neutral and Neutral conditions, which did not differ significantly from each other. By contrast, for patients the LPP was significantly larger during both the Preappraised negative and Preappraised neutral conditions compared to the Neutral condition, and the two Preappraise conditions did not differ from each other. Thus, the patients showed larger LPPs to unpleasant pictures than neutral pictures but no dampening of the LPP for the Preappraised neutral condition during this initial epoch.
LPP during epoch 2 (600–1000 ms)
There were again significant condition and interaction effects in this epoch. Controls continued to show a significant preappraisal effect evidenced by significant differences among the three conditions. The LPP amplitude showed a linear decrease across conditions, with the LPP largest during the Preappraised negative, intermediate during the Preappraised neutral, and smallest during the Neutral condition. Patients showed a strikingly different pattern. Their LPP was significantly larger during the Preappraised neutral condition than during both the Preappraised negative and Neutral conditions, which did not differ from each other. The patients therefore demonstrated an atypical pattern of relatively enhanced LPP for the Preappraised neutral condition.
LPP during epoch 3 (1000–1500 ms)
During the third epoch, a trend-level condition effect was accompanied by a significant interaction. The controls continued to show separation among the conditions in line with the expected preappraisal influence. The LPP was larger in the Preappraised negative versus Neutral condition whereas there was no difference between the Preappraised neutral and Neutral conditions; there was also a trend for larger LPP during the Preappraised negative versus Preappraised neutral conditions. For patients, the pattern was again different. Their LPP during the Preappraised neutral condition was significantly larger during than the Preappraised negative condition and different from the Neutral condition at a trend level. There was no difference between the Preappraised negative and Neutral conditions, again indicating an enhanced LPP during the Preappraised neutral condition.
LPP during epoch 4 (1500–2000 ms)
During the final epoch there was a significant condition effect but the interaction effect was not significant. Across the two groups, the LPP was larger for the Preappraised negative versus Neutral conditions (t 56 = − 2.17, p < 0.05) and there was a trend for larger LPP in the Preappraised negative versus Preappraised neutral conditions (t 56 = − 1.96, p = 0.09). For consistency, within-group t tests are also presented in Table 2. The control group continued to show a pattern similar to the previous epoch (the LPP was larger in the Preappraised negative versus Neutral condition and no difference between the Preappraised neutral and Neutral conditions; there was also a trend for larger LPP during the Preappraised negative versus Preappraised neutral conditions). There were no significant differences across condition within the patient group.
Group comparisons on self-reported emotion regulation strategies
The groups differed on both types of self-reported habitual emotion regulation strategies. Cognitive reappraisal scores were significantly higher in controls (mean = 5.3, s.d. = 1.0) than patients (mean = 4.5, s.d. = 1.3, t 56 = − 2.45, p < 0.05). By contrast, suppression scores were significantly lower in controls (mean = 3.2, s.d. = 0.9) than patients (mean = 3.9, s.d. = 1.6, t 56 = 2.27, p < 0.05).
Exploratory correlations within the schizophrenia group
First, we examined whether the four LPP difference wave scores (Preappraised negative minus Preappraised neutral) correlated significantly with self-reported habitual emotion regulation strategies (reappraisal, suppression) and clinical symptoms (positive, negative, total). Of the 40 correlations, only one was significant and the direction of the correlation was counterintuitive. Higher difference scores (i.e. greater preappraisal effect) during the 300–600-ms epoch correlated with higher positive symptoms (r 31 = 0.37, p < 0.05). There were also no significant correlations with CPZ equivalent units. (There were no significant correlations between ERPs and self-reported emotion regulation strategies in the control group.) Second, we examined correlations between the self-reported emotion regulation and clinical symptom variables. Higher levels of suppression correlated significantly with higher positive (r 31 = 0.36, p < 0.05), negative (r 31 = 0.45, p < 0.05) and total (r 31 = 0.41, p < 0.05) symptom levels. Reappraisal levels did not correlate significantly with symptoms.
Discussion
This ERP study evaluated whether people with schizophrenia alter their neurophysiological responses to unpleasant stimuli following appraisal-based manipulations of stimulus meaning. Consistent with prior studies (Foti & Hajcak, Reference Foti and Hajcak2008; Dennis & Hajcak, Reference Dennis and Hajcak2009; MacNamara et al. Reference MacNamara, Foti and Hajcak2009; DeCicco et al. Reference DeCicco, Solomon and Dennis2012), healthy controls showed a pattern in which the LPP was significantly reduced when unpleasant pictures were preappraised in less negative terms. Specifically, the LPP in the Preappraised neutral condition did not differ from the Neutral condition during the earliest epoch (300–600 ms) and fell between the Neutral and the Preappraised negative conditions during the middle two epochs (600–1500 ms). By contrast, patients failed to down-regulate their LPP responses, instead showing (a) larger LPP in response to both preappraise conditions versus Neutral conditions in the initial epoch and (b) the opposite pattern from controls in the middle periods with larger LPP to Preappraised neutral versus Preappraised negative and Neutral conditions. The patients also reported lower habitual use of cognitive reappraisal and higher expressive suppression than controls, although there were no significant correlations between these self-reports and LPP variables. This study provides the first electrophysiological evidence of impaired sensitivity to preappraisal manipulations of emotional stimulus meaning in schizophrenia.
The patients' absence of LPP preappraisal effects converges with the results of a few prior studies of cognitive reappraisal in schizophrenia using self-report, behavioral and fMRI paradigms (e.g. Eack et al. Reference Eack, Greeno, Pogue-Geile, Newhill, Hogarty and Keshavan2008; Kee et al. Reference Kee, Horan, Salovey, Kern, Sergi, Fiske, Lee, Subotnik, Nuechterlein, Sugar and Green2009; Livingstone et al. Reference Livingstone, Harper and Gillanders2009; van der Meer et al. Reference van der Meer, van't Wout and Aleman2009). A likely explanation for this impairment is a disturbance in top-down cognitive control of emotional responses. Disturbances in cognitive control and associated prefrontal cortical circuitry are well documented in schizophrenia (Barch & Braver, Reference Barch, Braver, Engle, Sedek, von Hecker and McIntosh2005; Lesh et al. Reference Lesh, Niendam, Minzenberg and Carter2011) and cognitive control plays a central role in cognitive reappraisal (Ochsner & Gross, Reference Ochsner and Gross2005). This explanation converges with a recent fMRI study in schizophrenia in which subjects were instructed to voluntarily decrease, maintain or increase their emotional responses to unpleasant pictures (Morris et al. Reference Morris, Sparks, Mitchell, Weickert and Green2012). When attempting to decrease their responses, patients displayed hypoactivation of prefrontal regions and significantly less corticolimbic coupling than healthy controls, consistent with impaired cognitive control. A key aspect of the fMRI study that affects interpretation was that subjects had to generate their own reinterpretations in the decrease condition. This leaves open the possibility that the patients' impairment merely reflected increased cognitive demands associated with generating alternative interpretations after unpleasant stimuli are presented, rather than an inability to benefit from a shift in meaning of the stimuli. The current ERP paradigm eliminated these differences in cognitive demands across conditions by presenting subjects with descriptions before each type of trial. In this way, the shifts in meaning occurred in a more implicit manner (for further discussion, see Foti & Hajcak, Reference Foti and Hajcak2008; MacNamara et al. Reference MacNamara, Foti and Hajcak2009). Even under these task conditions, schizophrenia patients showed impaired modulation of the neurophysiological responses.
We previously found that patients and controls showed comparable increases in LPP amplitudes for unpleasant versus neutral pictures across tasks in a passive viewing paradigm (Horan et al. Reference Horan, Wynn, Kring, Simons and Green2010) and when emotional pictures were incidental to task demands (Horan et al. Reference Horan, Foti, Hajcak, Wynn and Green2012). By contrast, patients in the current study showed a smaller LPP difference between the Preappraised negative and Neutral conditions than controls; this dampening was true even during the first epoch (t 56 = 2.80, p < 0.005). Because the recruitment process and sample characteristics were similar across studies, the most likely explanation for this discrepancy is differences in the experimental tasks. An important difference is that the current study provided descriptive sentences before each trial whereas the previous studies did not provide any such information. The sentence frames essentially provide a context in which the images can be perceived. Thus, the patients’ atypical LPP pattern in the current preappraisal manipulation may be associated with impaired context processing, an aspect of cognitive control that has also received considerable investigation in schizophrenia (Park et al. Reference Park, Lee, Folley and Kim2003; Hemsley, Reference Hemsley2005; MacDonald et al. Reference MacDonald, Carter, Kerns, Ursu, Barch, Holmes, Stenger and Cohen2005).
Although preappraisal modulated the LPP in the control group, it did not affect the patients in a similar way. In the initial epoch, patients showed a larger LPP to unpleasant pictures in both the Preappraised negative and Preappraised neutral conditions compared to the Neutral condition. Although this indicates some early sensitivity to the emotional significance of unpleasant pictures, no preappraisal modulation was evident. During the two middle epochs, the sentence frames seemed to have a paradoxical effect in the patients: they showed larger LPP amplitude in the Preappraised neutral condition versus Preappraised negative and Neutral conditions, suggesting that the Preappraised neutral condition in fact resulted in enhanced, sustained neural processing. Perhaps the relative incongruity between the context provided by the more neutral frames and the evocative unpleasant pictures was more difficult or effortful for patients to process than the other conditions in which the sentence and picture pairings were congruous. One might consider whether this effect is impacted by N400, a centro-posteriorly distributed negative-going waveform found to be larger (more negative) to words that are semantically incongruous (versus congruous) with preceding words or sentences (van Petten & Luka, Reference van Petten and Luka2006). If N400 accounted for the patients' unusual pattern, we would expect a more negative LPP in the Preappraised neutral condition than in the Preappraised negative condition in this group. However, the patients showed the opposite pattern. Furthermore, prior research in healthy subjects suggests that LPP preappraisal modulation in this paradigm is not driven by N400 (MacNamara et al. Reference MacNamara, Foti and Hajcak2009). Thus, patients failed to show an adaptive benefit from the meaning shift typically associated with the contextual frames. Context is a complex, multi-faceted construct (Park et al. Reference Park, Lee, Folley and Kim2003; Hemsley, Reference Hemsley2005; MacDonald et al. Reference MacDonald, Carter, Kerns, Ursu, Barch, Holmes, Stenger and Cohen2005) and behavioral studies that examined socio-emotional context processing in schizophrenia have found both impaired (Green et al. Reference Green, Waldron and Coltheart2007; Huang et al. Reference Huang, Chan, Lu and Tong2009) and intact (Chung & Barch, Reference Chung and Barch2011; Lee et al. Reference Lee, Kern, Harvey, Horan, Kee, Ochsner, Penn and Green2012) context effects. Further investigations will help to clarify the factors that influence how contextual information impacts emotional responses in schizophrenia.
This study also examined self-reported habitual use of emotion regulation strategies. In addition to an atypical LPP pattern, the schizophrenia group reported less frequent use of cognitive reappraisal and more frequent expressive suppression than healthy controls. This pattern is consistent with some (Livingstone et al. Reference Livingstone, Harper and Gillanders2009; van der Meer et al. Reference van der Meer, van't Wout and Aleman2009) but not all (Henry et al. Reference Henry, Rendell, Green, McDonald and O'Donnell2008; Badcock et al. Reference Badcock, Paulik and Maybery2011; Perry et al. Reference Perry, Henry and Grisham2011) prior studies. Although there were no significant correlations between self-reported emotion regulation strategies and LPPs, higher levels of suppression were associated with higher levels of positive, negative, and overall symptoms. Again, some prior studies have found suppression to be related to higher negative or positive symptoms (Henry et al. Reference Henry, Green, de Lucia, Restuccia, McDonald and O'Donnell2007; Kee et al. Reference Kee, Horan, Salovey, Kern, Sergi, Fiske, Lee, Subotnik, Nuechterlein, Sugar and Green2009; Badcock et al. Reference Badcock, Paulik and Maybery2011), and to worse social functioning (Perry et al. Reference Perry, Henry and Grisham2011), but others did not (Henry et al. Reference Henry, Rendell, Green, McDonald and O'Donnell2008). It is unclear why the results have been inconsistent across studies, although differences in sample characteristics (e.g. some included mixed samples of out-patients and in-patients, and in one study patients were selected for high levels of hallucinations) and symptom assessment measures may be relevant. The current findings are consistent with the notion that expressive suppression is associated with greater levels of psychopathology (Werner & Gross, Reference Werner, Gross, Kring and Sloan2010), even among people with schizophrenia.
Some limitations of this study should be considered. First, this study focused on an ERP index of emotional regulation and did not include self-reported levels of valence or arousal after each trial. Future studies could examine the coherence between ERP's and self-reported emotions, along with other aspects of emotional responding such as facial electromyography (Wu et al. Reference Wu, Winkler, Andreatta, Hajcak and Pauli2012). Second, this study only examined the preappraisal of unpleasant images. We focused on this emotion regulation process to eliminate differences in difficulty across task conditions and focus on contextual modulation of the LPP. It remains an open question whether similar LPP results would be seen for other types of regulation strategies (e.g. cognitive reappraisal, expressive suppression) or for pleasant stimuli. Third, patients were taking various antipsychotic medications at clinically determined dosages and the impact of these medications on the LPP are unknown. Although there were no significant correlations with CPZ equivalents and atypical antipsychotics have in fact been reported to improve some aspects of emotional processing (Juckel et al. Reference Juckel, Schlagenhauf, Koslowski, Filonov, Wustenberg, Villringer, Knutson, Kienast, Gallinat, Wrase and Heinz2006; Schlagenhauf et al. Reference Schlagenhauf, Juckel, Koslowski, Kahnt, Knutson, Dembler, Kienast, Gallinat, Wrase and Heinz2008), further research on antipsychotic medication effects and in unmedicated samples is needed. Fourth, the patients were clinically stable, predominantly male and chronically ill, which may limit the generalizability of these findings. Research in other types of samples, such as patients in the early phases of schizophrenia, can help to address these limitations.
Although recent research has highlighted several areas of intact emotional responding in schizophrenia, the current findings contribute to emerging evidence of disturbances in the regulation of emotion in this disorder. Impairments in emotion regulation may show linkages to clinical symptoms and could also contribute to social cognitive impairments associated with poor functional outcome (Green & Horan, Reference Green and Horan2010). For example, emotion regulation has been described as a key component of empathy (Decety & Lamm, Reference Decety and Lamm2006), a core social cognitive processes that seems to be disturbed in schizophrenia (e.g. Smith et al. Reference Smith, Horan, Karpouzian, Abram, Cobia and Csernansky2012). Difficulties in emotion regulation have been associated with various forms of psychopathology and empirically supported interventions have been developed to treat them (Aldao et al. Reference Aldao, Nolen-Hoeksema and Schweizer2010). The current results support recent suggestions that emotion regulation interventions may also be useful in treating people with schizophrenia (Bassam & LeCompte, Reference Bassam and LeCompte2012).
Acknowledgments
Support for this study came from a Veteran Affairs (VA) Career Development Award (PI: W. P. Horan) and National Institute of Mental Health (NIMH) Grants MH065707 and MH43292 (PI: M. F. Green). We thank J. Lee, A. Bender, M. Dolinsky, C. Gibson, C. Tripp and K. Weiner for assistance in data collection.
Declaration of Interest
M. F. Green reports having received consulting fees from Abbott Laboratories, Amgen, Cypress, Lundbeck and Teva, and also speaking fees from Otsuka and Sunovion. The rest of the authors report no biomedical financial interests or potential conflicts of interest.
Appendix 1
Descriptions for International Affective Picture System (IAPS) pictures
