Neural activation during emotion regulation in people with ultra-high risk for psychosis

Abstract

Disturbances in emotion processing and regulation are a hallmark characteristic of schizophrenia, and have been shown to be present in at-risk populations. We investigated behavioral and functional Magnetic Resonance Imaging (fMRI) differences during emotion processing and reappraisal of negative images in people at ultra-high risk (UHR) for psychosis compared to healthy controls (HC). We hypothesized that during emotion processing UHR participants will show increased activation of the ventral stream compared to HC. During reappraisal UHR were expected to show less activation of the dorsal stream compared to HC. Twelve UHR and eleven HC participated in this fMRI study in which they performed an Emotion Regulation Task. In addition, participants were asked to complete several relevant questionnaires. Brain activation during processing of negative versus neutral images did not yield significant differences between the UHR and HC, but showed a trend for increased activation in UHR participants: HC showed activation in the middle and inferior temporal gyrus, whereas the UHR group showed activation in the bilateral DLPPFC, and bilateral frontopolar cortex, the superior and inferior occipital gyrus,the cingulate gyrus, the insula and the basal ganglia. During reappraisal differences in brain activation supported our hypothesis: the UHR group showed decreased prefrontal and temporal lobe activation compared to HC. Results from the behavioral data showed that both groups were able to successfully down-regulate negative emotion. We concluded that the at-risk group showed abnormal brain activity during reappraisal that was in line with expectations, however the outcome of the behavioral results remains puzzling, and should be subjected to further study.

Keywords: Schizophrenia, at-risk, emotion processing, reappraisal, fMRI

Introduction

Schizophrenia is a severe and frequently lifelong disorder marked by psychotic symptoms. Research findings have enabled us to identify individuals who are at high-risk of developing schizophrenia, yet we do not know how to predict whether an individual at high-risk will make the transition towards a psychosis (McGlashan, 2003). This is crucial for starting treatment as soon as possible since this is related to better long-term outcomes (McGlashahn, 2003).

Psychotic refers to a state in which a person is suffering from delusions, hallucinations, disorganized speech or behavior. The course of schizophrenia often begins with a prodromal phase in which symptoms become gradually noticeable (American Psychiatric Association, 2000). Negative symptoms often emerge first; examples are apathy (lack of motivation), alogia (poverty of speech) and withdrawal from social interactions (American Psychiatric Association, 2000). The progression of the disorder is highly variable, and a full remission is rare (American Psychiatric Association, 2000). The disorder involves cognitive and emotional dysfunctions, which make a normal life difficult for affected individuals. These disturbances impact on the patients’ social, educational, and occupational lives. Many individuals with schizophrenia are unable to hold their previous jobs, or finish school (American Psychiatric Association, 2000).

Emotion Processing

Disturbances in emotional processing and regulation are a hallmark feature of schizophrenia. At the onset of the disorder people with schizophrenia appear to be hypersensitive to emotional events, while showing less overt expression of emotion (Phillips and Seidman, 2008). This is linked to alexithymia, a personality trait that is characterized by difficulties in identifying and conveying emotions verbally and through body language (Vorst & Bermond, 2001). Abnormalities extend to perception, processing and regulation of emotion, and manifest in negative symptoms such as affective flattening, an inability to experience pleasure (anhedonia), depression, anxiety and anger (Phillips and Seidman, 2008; American Psychiatric Association, 2000). It appears that abnormal emotion processing, social cognition and executive functions are closely intertwined in schizophrenia (Green & Malhi, 2006), and these deficiencies are thought to account for a number of symptoms such as flat affect, anxiety, and even be the precursor for delusions (Green & Malhi, 2006). Similarly, poor social skills along with misinterpreting social cues in interactions seem to precipitate persecutory delusions, as they deal with one’s role within society (Phillips, Drevets, Rauch, & Lane, 2003b).

Emotion Regulation

Emotion regulation is defined as a conscious or unconscious process that attempts to modify the emotional experience and the subsequent expression of it (Gross, 1998). Gross (1998) has distinguished two types of emotion regulation strategies: antecedent-focused and response-focused strategies. Antecedent-focused strategies are applied early, i.e. before an emotional response has been generated. On the other hand, response-focused strategies are exerted at a later stage, after the emotional response has been produced (Gross, 1998). Reappraisal is an antecedent strategy that has deserved much attention in research because it is used in everyday life, and is applicable to research because it can be manipulated and assessed in several ways (John & Gross, 2004).

Reappraisal is a relatively cost-efficient way of dealing with emotion for it requires few resources, so that more resources are available to deal with social interaction appropriately (Gross, 2002). Suppression on the other hand, requires more resources without reducing the negativity of the emotion. Thus, the quality of interaction suffers resulting from the discrepancy of the true feelings one has and the feelings one shows (John & Gross, 2004). Indeed, some of the benefits of using reappraisal as an emotion regulation strategy have been shown to be related to higher levels of personal growth, self-acceptance, a clearer purpose in life and better relationships with other people (Gross, 2002; John & Gross, 2004).

Henry and colleagues (2008) have investigated self-reported levels of emotion regulation using the Emotion Regulation Questionnaire (ERQ; Gross & John, 2003) in schizophrenia patients and healthy controls (HC). It was hypothesized that individuals with schizophrenia would report lower levels of reappraisal, and higher levels of suppression compared to the healthy controls. However, they found that patients and controls did not differ with regards to self-reported reappraisal and suppression (Henry, Rendell, Green, McDonald, & O’Donnell, 2008). The authors speculated that the absence of a difference was due to either the specific sample they used or to the nature of the scale. The study also looked at the level of social functioning (Social Functioning Scale, SFS; Birchwood, Smith, Cochrane, Wetton, & Copestake, 1990), which was found to be significantly lower in the patient group. Lower use of reappraisal was associated with worse social functioning (Henry et al.,2008). Further, the use of reappraisal was negatively correlated with depression levels(Henry et al., 2008).

Brain-Imaging Studies

Brain-imaging techniques can reveal what happens in the brain during emotion regulation, and thereby add valuable information beyond what questionnaires supply. In this setting reappraisal of affective stimuli is typically studied by instructing participants to reinterpret the meaning of a negatively valenced image so that the negative emotion is diminished (Ochsner, 2008). Behavioral data is collected at the same time, and refers to participant’s rating of the extent to which they experience negative emotion after being exposed to a stimulus.

Phillips and colleagues (2003a) proposed a model with two distinct neural systems; one for emotion processing, and another that is activated during emotion regulation. The former, referred to as the ventral stream, involves the amygdala, insula and the ventral regions of the anterior cingulate cortex (ACC) and prefrontal cortex (PFC). The latter system, known as the dorsal stream, includes the dorsal regions of the ACC, PFC and the hippocampus (Phillips, Drevets, Rauch, & Lane, 2003a). Frontal regions are generally thought to be important for language, attention, memory, response selection and reference to the self, all processes that are necessary for a meaningful reinterpretation of valenced stimuli (Ochsner, 2008). Thus, one can expect frontal regions to be implicated in emotion regulation. Indeed, studies using an emotion regulation paradigm have implied frontal regions, in particular the dorsal lateral prefrontal cortex (PFC), and the dorsal medial PFC, as well as regions that deal with responding to emotion such as the amygdala and insula (Ochsner, 2008). Further, the ACC and PFC regions are implied in the application and selection of a reappraisal strategy (Ochsner & Gross, 2005). Further, the dorsal and ventral pathways and their associated structures appear to play a role in self-referent processing (van der Meer et al., 2010), which is undoubtedly also a crucial part of social functioning.

Goldin et al. (2008) used functional Magnetic Resonance Imaging (fMRI) to investigate reappraisal and suppression of negative emotional stimuli in a sample of healthy women. Functional scans of the women’s brains were acquired while they watched short films that were either neutral or negative in content. It was found that reappraisal in down-regulating emotion was employed successfully. During the early stages of reappraisal enhanced responses in the bilateral prefrontal cortex were found, which was associated with decreased left amygdala and left insula responses. The authors concluded that this can be taken as an index of how cognitive control asserted by the PFC interacts with subsequent emotional processes that affect neural, behavioral responses, and the experience of the emotion (Goldin, McRae, Ramel, & Gross, 2008).

As mentioned earlier, schizophrenia is associated with disturbances in both emotional processing and regulation. Structural and functional abnormalities are evident in patients with schizophrenia, and it is thought that these differences may underlie the symptoms specific to schizophrenia (Phillips et al., 2003b). The authors proposed a ventral/dorsal pathway model (Phillips et al., 2003a) in which the ventral pathway was identified to play a role in recognizing the emotional significance of a stimulus, which is in line with the finding of functional abnormalities in schizophrenia that include reduced activation in the amygdala, anterior insula and nucleus accumbens to unambiguous emotional stimuli (Phillips et al., 2003b) On the other hand, it appears that ambiguous stimuli trigger these areas more easily, thereby giving way to delusions (Phillips et al., 2003b). These abnormalities are thought to lead to symptoms associated with alexithymia, such as the impaired identification of emotion, as well as its diminished physical expression. In Phillips’ model (2003a) the dorsal pathway is hypothesized to be engaged in down-regulation of emotion, and indeed patients with schizophrenia have been shown to exhibit attenuated activity within the dorsal prefrontal cortex during tasks that require executive function (Phillips et al., 2003b). For instance, in a task combining emotional stimuli with executive function such as a go/no-go task with positive, negative and neutral words showed increased prefrontal activation in healthy controls, yet schizophrenia patients failed to recruit frontal areas in response to negative emotional words (Vercammen et al., 2012).

Phillips’ model is partially supported by findings from van der Meer et al. (2011), who compared neural activation in healthy controls with schizophrenia patients and their relatives during reappraisal. Most importantly, patients showed decreased activation in the bilateral ventrolateral and dorsolateral PFC, dorsomedial PFC, left interior insula and the inferior parietal lobe (IPL), whereas the relatives of patients showed increased activation in these areas. Thus, the results support Phillips’ model by confirming the hypothesis of decreased activation in frontal areas of patients during an effortful task.

Studying At-Risk Groups

The literature on emotion regulation and schizophrenia does not only encompass patient groups, but also at-risk groups because studying at-risk groups can help to identify vulnerability markers or possible predictors for who will develop a psychosis. At-risk groups are made up of several different subgroups, and for clarification purposes they are briefly mentioned here. Schizotypy or psychosis- proneness (PP) refers to a group that does not experience clinically significant symptoms but scores higher on the character trait of schizotypy. More severe subgroups are early prodromal and late prodromal, both experience symptoms severe enough to seek help, with treatment indicated for the latter. Alternative names for the late prodromal group are ultra-high risk (UHR) or at-risk mental state (ARMS) (Weinberger & Harrison, 2011).

Abbildung in dieser Leseprobe nicht enthalten

Figure 1. Continuum from Healthy Controls through different stages of at-risk groups to Schizophrenia patients.

A literature review of studies examining the outcomes of three different at-risk groups, specifically PP, familial high risk (FHR) and treatment-seeking prodromal patients, found that these groups showed impaired emotion perception, experience increased levels of negative emotion, and anhedonia, although to a lesser extent than in schizophrenia (Phillips & Seidman, 2008).

A study that used fearful facial expressions as emotional stimuli to investigate emotion processing included participants with a schizophrenic treatment-seeking sibling. The findings from this study suggest that at-risk populations show decreased activation in limbic and cortical structures, including the OFC, relative to healthy controls, and that these deviations reflect an increased level of vulnerability to schizophrenia (Venkatasubramanian, Puthumana, Jayakumar, & Gangadhar, 2010). Similarly, Modinos, Ormel and Aleman (2010) studied emotion regulation in undergraduate students who were high on PP. Reappraisal of negative images using fMRI was examined, and both groups showed successful down-regulation of negative emotion. For reappraisal both groups exhibited increased activation in left dorsolateral and dorsomedial PFC, ACC, and bilateral ventrolateral PFC. However, in high PP- group the activation in the left dorsomedial PFC, the ACC as well as the right ventrolateral PFC was found to be greater than in HC (Modinos et al., 2010). One may conclude that the increased prefrontal activation worked in a compensatory fashion, thereby resulting in successful down-regulation of negative affect in the at- risk group.

To our knowledge, there is no study that examines brain activation during emotion regulation in an ultra-high risk for psychosis population. The aim of the present study was to investigate whether UHR-patients show differential brain activations during emotional processing of visual stimuli as well as during reappraisal of negative images in comparison to healthy control (HC) subjects. We hypothesized that negative images will activate the ventral pathway more than neutral stimuli (Phillips et al., 2003). UHR subjects are expected to show attenuated activation of these areas compared to HC. In the reappraise condition compared to the attend condition, we expected to see activation of the dorsal pathway (Phillips et al., 2003). The UHR group is expected to show less activation of these regions than HC, which we think will result in less successful down-regulation compared to HC. Further, questionnaire data will be used to explore brain function during emotion regulation in association with alexithymia, positive and negative affect, and self-rated emotion regulation. In particular, we will test the idea that the emotional component of the alexythymia scale is hypothesized to be associated with the emotional processing task, whereas the cognitive component ought to be related to the reappraisal task.

Materials and Methods

Participants

Twelve young adults with ultra-high risk for developing a psychosis were recruited from GGZ-Friesland, and eleven healthy controls, recruited via local advertisements, participated in this study. The Comprehensive Assessment of At-Risk Mental State interview (CAARMS; Yung, Phillips, & McGorry, 2001) interview was used to define UHR for psychosis in the at-risk group. Exclusion criteria for the healthy control subjects were whether they ever had been in contact with a psychologist, psychiatrist or neurologist, in order to ensure them being free of psychological and neurological disorders. Before scanning, all participants were screened for MRI suitability, and only those who met fMRI safety regulations were scanned. Participants were not scanned if they had red tattoos, permanent piercings, metal in their body from surgery, or if they were or could be pregnant, or suffered from claustrophobia. Drug- and medication use was recorded. Demographic and clinical characteristics are displayed in Table 1. The study was approved by the national ethical committee (Metig UMCG), and carried out in accordance with the current version of the Declaration of Helsinki. All participants provided written informed consent after they were informed about the procedure and before the experiment took place. Participants received 25 euros and refund of travel expenses as financial compensation.

Questionnaires

Diagnostic and clinical characteristics

At GGZ-Friesland all new at-risk patients were screened with the Prodromal Questionnaire-16 (PQ-16, Loewy, Bearden, Johnson, Raine, & Cannon, 2005). A score above 6 indicated a heightened risk, and those patients were invited for the CAARMS interview (Yung et al., 2001). By using both the PQ-16 and the CAARMS interview a two-stage screening process was used to define at-risk for psychosis patients (Rietdijk et al., 2010). An at-risk mental state is defined as having attenuated psychotic symptoms that may consist of unusual thought content, non-bizarre ideas, perceptual anomalies, and disorganized speech.

Further, to assess the current amount of positive and negative symptoms experienced at moment of scanning in the at-risk group the Positive and Negative Syndrome Scale (PANSS; appendix 3, Kay, Flszbein, & Opfer, 1987) was administered.

Emotion Regulation Questionnaire

The ERQ (appendix 1, Gross & John, 2003) is a 10 item self-report questionnaire that assesses the extent to which one uses suppression and reappraisal as emotion regulation strategies in everyday life. The questionnaire is divided into two parts; four questions assessing the use of suppression (e.g. I keep my emotions to myself) and six question assessing reappraisal (e.g. When I want to feel more positive emotion, I change the way I think about the situation). The participant is asked to rate on a 7-point scale how much they agree with the statement (strongly agree-strongly disagree). Reliability and validity of the questionnaire was supported by a multi- sample study conducted by Gross and John (Gross & John, 2003).

Bermond-Vorst Alexithymia Questionnaire

This 40-item self-report questionnaire was administered to all participants in order to assess alexythymia. The BVAQ (appendix 2, Vorst & Bermond, 2001) is made up of five subscales: verbalizing, analyzing, identifying, emotionalizing and fantasizing, each containing eight items. Examples of test items are “When something totally unexpected happens, I remain calm and unmoved” (Emotionalizing), “When I am distressed, I know whether I am afraid or sad or angry” (Identifying). Answers are rated on a scale from 1-5 (certainly does not apply to me – certainly applies to me). Higher scores reflect more alexithymia characteristics. Cronbach’s alpha was found to be 0.79 across the subscales (Vorst & Bermond, 2001).

Positive Affect Negative Affect Scale

The PANAS (appendix 4, Watson, Clark, & Tellegen, 1988) is a 20-item questionnaire measuring the extent to which one feels positive emotion and negative emotion at that particular time. Specifically, it is made up of ten descriptors each, and the participant is asked to rate on a 5-point scale the extent to which he feels that mood state at that time (not at all-very much). For example positive descriptors are “excited” and “enthusiastic”, whereas negative descriptors are “hostile” and “ashamed”.

Emotion Regulation Task

The Emotion Regulation Task (ERT), originally developed by Ochsner, Bunge, Gross and Gabrieli (2002), was adapted for our use. Stimuli were presented using the E-Prime software (Psychology Software Tools, Pittsburgh, PA). The images used were drawn from the International Affective Picture System (IAPS; Lang, Bradley, & Cuthbert, 2005). The task itself consisted of the following three conditions: attend neutral, attend negative and reappraise, with 22 trials each. Every trial was constructed as follows (see Figure 2).

Abbildung in dieser Leseprobe nicht enthalten

Figure 2. Design of an Emotion Regulation Task trial as used in this study.

[...]