Emotion Regulation Predicts Everyday Emotion Experience and Social Function in Schizophrenia

Methods

Participants

Thirty-four individuals meeting Diagnostic and Statistical Manual of Mental Disorders (4th edition, text rev.; DSM-IV-TR; American Psychiatric Association, 2000) criteria for schizophrenia (n = 27) or schizoaffective disorder (n = 7) participated in this study. Exclusion criteria included (a) DSM-IV-TR diagnosis of substance abuse or dependence in the past 6 months, (b) DSM-IV-TR diagnosis of a current mood disorder, (c) changes in medication 2 weeks prior to consent, (d) IQ less than 70 as measured by the Wechsler Test of Adult Reading (WTAR; Wechsler, 2001), and (e) history of severe head trauma and/or loss of consciousness. All participants were stable outpatients who had not changed medications 2 weeks prior to study participation. Diagnoses were confirmed using the Structured Clinical Interview for DSM-IV Axis I disorders. Psychotic symptoms were assessed using the Brief Psychiatric Rating Scale (BPRS; Overall & Gorham, 1962). Negative symptoms were assessed using the Clinical Assessment Interview for Negative Symptoms (CAINS; Kring, Gur, Blanchard, Horan, & Reise, 2013). The CAINS consists of a motivation and pleasure scale (MAP) and an expression subscale (EXP). Higher CAINS scores relates to greater pathology. All participants provided written informed consent to the protocol approved by the Washington University Institutional Review Board. Participant demographics are presented in Table 1. Thirty-two participants were taking an antipsychotic at the time of study participation, while two participants were medication free for at least 1 month prior to study participation. Participant ethnicity reflects that of the St. Louis community. There were no relationships between participant demographics (age, ethnicity, gender, parental education, medication status) and study variables; thus, they were not included in the presented analyses.

OPEN IN VIEWER

Table 1.

Demographics and Clinical and Self-Report Measures

	M	SD	%	n
Age	39.48	8.40
Sex (% female)			48
Education	12.81	2.86
Parental education	14.23	3.56
Estimated IQ	95.21	16.65
Race
African American			67
Caucasian			33
Clinical ratings
BPRS positive symptoms	8.52	5.18
CAINS MAP	18.23	6.22
CAINS EXP	5.85	3.59
Medication
Typical antipsychotics				6
Atypical antipsychotics				26
Medications				2
Self-report measures
ERQ reappraisal	28.17	8.53
ERQ suppression	15.53	6.15
SBI savoring	5.73	5.62
SLOF—Informant	28.90	2.34
SLOF—Self-report	27.19	6.24
EMA ratings
Positive affect	3.52	0.96
Negative affect	1.65	0.84
Social interaction	3.03	0.84
Social interest	3.39	0.81

Note: BPRS Psychosis = Psychosis Subscale of the Brief Psychiatric Rating Scale; CAINS EXP = Expression Subscale of the Clinical Assessment Interview for Negative Symptoms; CAINS MAP = Motivation and Pleasure Subscale of the Clinical Assessment Interview for Negative Symptoms; ERQ = Emotion Regulation Questionnaire; Estimate IQ = premorbid IQ based on Wechsler Test of Adult Reading (Wechsler, 2001); SBI = Savoring Beliefs Inventory; SLOF = specific level of functioning Interpersonal Relationship Sub-Scale; Social interaction = mean number of people interacted with daily; Social interest = mean interest in daily social interactions.

Results

Cognitive reappraisal

We first examined the relationship between cognitive reappraisal, emotional experience, and social functioning in daily life. As shown in Table 2, we found a significant relationship between self-reported habitual use of reappraisal and positive affect in daily life such that greater use of reappraisal was associated with more positive emotion in daily life. However, inconsistent with our hypotheses, we did not find a significant relationship between reappraisal and negative emotion, social interaction, or social interest in daily life. There was no significant relationship between reappraisal and social functioning as measured by the SLOF self (r = .03, p = .87) or informant report (r = .11, p = .63). Further, cognitive reappraisal was not significantly related to CAINS MAP (r = –.30, p = .12), EXP (r = –.19, p = .32), or BPRS Positive symptoms (r = –.13, p = .51).

OPEN IN VIEWER

Table 2.

Results From Hierarchical Linear Models Using Emotion Regulation Strategies to Predict Daily Emotional Experience and Social Interaction

	Reappraisal			Suppression			Savoring
Dependent variable	b	t	p	b	t	p	b	t	p
Positive emotion	.34	5.32	.001	−.28	−5.36	.005	.17	3.36	.02
Negative emotion	−.03	−1.12	.48	.36	6.59	.001	−.19	−4.50	.01
Social interaction	.09	1.20	.33	−.21	−4.36	.01	.24	5.96	.002
Social interest	.11	1.30	.40	−.30	5.76	.003	.31	6.32	.001

Discussion

The goal of the current study was to examine the relationship between self-reported habitual use of emotion regulation strategies and emotional and social functioning in the daily lives of those with schizophrenia. Consistent with research in healthy controls, we found that habitual use of emotion regulation strategies has important links to both the experience of emotion and social interaction in daily life. We discuss each regulation strategy and related findings below.

Similar to findings in healthy adults, we found that individuals with schizophrenia who reported greater habitual use of cognitive reappraisal were more likely to report greater positive emotion while going about their daily lives. In contrast to previous findings in healthy controls (Goldin, McRae, Ramel, & Gross, 2008; Gross & John, 2003), habitual use of cognitive reappraisal in schizophrenia was not related to reduced negative emotion. This finding is also inconsistent with laboratory-based studies that found negative emotion decreased in schizophrenia following specific instruction to reappraise their emotional response to a sad mood induction (e.g., Perry et al., 2012). It may be that in schizophrenia, reappraisal is effective at reducing negative emotion immediately following reappraisal in a more controlled setting but may not relate to more sustained reductions in negative emotion in more natural settings. It may also be the case that reappraisal did not relate to the down-regulation of sadness and anxiety assessed in the current study but would relate to reductions in other negative emotions, such as anger or guilt. Future studies utilizing EMA should investigate the short- and long-term effects of cognitive reappraisal in schizophrenia and healthy controls to highlight the temporal relationship between reappraisal and negative emotional experience, clarifying if and when negative emotional response is reduced. It is also noteworthy that while reappraisal did relate to daily positive emotion, it did not relate to negative symptoms or social functioning. While this replicates prior findings from Henry and colleagues (2008), who found no relationship between reappraisal and symptom ratings, this lack of relationship is somewhat surprising given that reappraisal is thought to be an adaptive strategy linked to functioning and emotional well-being. These findings suggest that reappraisal may not be effectively linked to important clinical outcome measures that are used across studies. It will be important for future work to examine whether consistent use of reappraisal over time does relate to changes in symptoms or functioning, to help clarify whether reappraisal is an important therapeutic intervention.

The second regulation strategy we focused on in the present study was emotional suppression. Given that schizophrenia has long been associated with diminished emotional expression, which is linked to poor outcomes (e.g., Gur et al., 2006), suppression may be a regulation strategy relevant to understanding poor psychosocial function in schizophrenia. Our current findings suggest that those with schizophrenia who report greater habitual use of suppression also report greater negative emotion and reduced positive emotion when assessed in daily life. Moreover, emotional suppression was related to less social interaction and interest in daily life and clinician-rated reductions in expressivity. These findings fit within the model of emotional suppression as a potential maladaptive strategy in healthy controls (e.g., John & Gross, 2004) and in schizophrenia (Perry et al., 2012). These findings also fit with previous literature linking flat affect with reduced social relationships and social functioning in schizophrenia (Evensen et al., 2012). Indeed, a main component of social skills training for people with schizophrenia involves teaching common forms of communication, such as a smile when greeting others or expressing and sharing feelings with others, which may help to reduce the use of suppression.

The final regulation strategy investigated in the present study was the savoring of positive emotion. Basic emotion research has shown that savoring of positive emotion is linked to a host of positive outcomes including increased positive emotion, reduced negative emotion, and greater feelings of well-being and social functioning (e.g., Bryant, 2003; Livingstone & Srivastava, 2012; Quoidbach et al., 2010). Consistent with these findings, habitual use of savoring was related to greater positive emotion, reduced negative emotion, and greater social interaction and interest in the daily lives of those with schizophrenia. Habitual use of savoring was also related to clinician-rated negative symptom ratings of motivation and pleasure, pointing to the potential importance of savoring of positive emotion in the lives of those with schizophrenia. In a pilot study examining the effectiveness of meditation in schizophrenia, Johnson and colleagues (2011) found that following a brief meditation intervention, people with schizophrenia showed an increase in positive emotion, which was linked to savoring of future positive emotion. Our findings support this link between savoring and experience of positive emotion in daily life, suggesting that savoring may be a potential target for intervention. Given that savoring has received limited attention thus far in schizophrenia, future work should look to examine regular usage of savoring and further explore potential adaptive and maladaptive correlates of this regulation strategy.

One additional aim of the current study was to examine multiple measures of social functioning and their relationship to emotion regulation. We found that daily social interest, and interactions, but not self- or informant reports of social function on the SLOF, related to emotion regulation strategy use. These findings may suggest that emotion regulation relates to the day-to-day social experiences of individuals with schizophrenia but is not an important influence in the overall social functioning as measured by the SLOF. This contrast in findings may be due to differences between the two measures of social functioning. Our EMA measures of social involvement assessed both the number of people participants interacted with each day along with their interest in those social interactions. In contrast, the SLOF asks participants and their informant to reflect on current social functioning across a variety of domains including effective communication, participation in groups, and maintaining friendships. Further, our measure asked participants to report their interactions each day, rather than sum their experience over 30 days as the SLOF does, thus requiring participants to recall experiences over a shorter period of time. It is unclear if one measure of social involvement is more or less important to our understanding of patient functioning; rather, they may represent somewhat distinct components of social functioning. Moreover, we found that daily social interest, but not quantity of social interactions, related to informant-reported social functioning (as measured by the SLOF). This suggests that level of interest may be a more relevant marker of overall functioning than a measure relating solely to the quantity of social interactions. Indeed, while a person may report a number of social interactions, this does not gauge the quality of these interactions. It may be that a handful of highly engaged interactions are more relevant to social functioning than overall number of interactions.

Finally, neither daily reports of social interest nor interaction were related to self-reported social functioning on the SLOF. This finding is consistent with prior work suggesting an important disconnect between informant and self-reported functioning with patients over- or underestimating functioning relative to their informant (e.g., Bowie et al., 2007; Sabbag et al., 2012). Future work examining the link between self- and informant-based social functioning measures and ecological assessments of social interactions will be important to clarify and extend our understanding of how to assess social functioning in the daily lives of people with schizophrenia. Taken together, our findings suggest that daily interest in social activity is closely related to both informant reports of social functioning and habitual emotion regulation but may not be related to self-assessment of global social skills.

The present study had several limitations. First, we chose to assess only four distinct emotions and collapsed these into composite measures of positive and negative affect. It may be the case that certain discrete emotions may be more impacted by various regulatory strategies (e.g., suppressing sadness vs. suppressing anger). It will be important for future studies to not only assess a wider range of emotions but to potentially take a discrete emotion approach to investigate whether distinct emotions (e.g., fear, happiness) are more specifically related to emotion regulation and well-being in schizophrenia. Second, while the structure of EMA data allows for the assessment of emotional variations across the day/week, we did not have specific hypotheses about how emotional responses would vary by time of day or day of the week. Thus, the present analyses did not include specific predictors related to time and do not provide insight into variations in emotional experience or social interaction as a function of emotion regulation. Future studies, assessing emotion regulation via EMA will be better able to hypothesize and examine variance in emotional response following emotion regulation and thus be better able to clarify the immediate linkages between regulation and emotional responses. Third, while we had the power to detect significant relationships, our sample size was modest. Fourth, nighttime surveys assessing participants’ social interaction throughout the day required participants to reflect upon the day rather than report their experience in the moment. While this relies on participants’ ability to retrospectively recall their social interactions across the day, this still represents an important complement to measures that require participants to recall their interactions across a week or more. Future studies should include assessment of social experience throughout the day to gain a more refined view of social response. Finally, control participants were not included in the current study; thus, we are unable to make claims about group differences in use of emotion regulation or in how these strategies relate to emotional and social experience in the everyday life of healthy controls. Based on extensive prior literature, we would expect that healthy controls would show similar relationships as seen in the present study, though we do not think this diminishes the importance of establishing the relevance of habitual use of emotion regulation strategies in relationship to daily life function in schizophrenia.

The current study adds to a growing literature investigating emotion regulation, emotional experience, and social functioning in schizophrenia. Our findings demonstrate that the habitual use of emotion regulation strategies such as cognitive reappraisal and savoring relate to the experience of greater positive emotion in the daily lives of those with schizophrenia. In contrast, we found that the habitual use of suppression is related to greater negative emotion, reduced positive emotion, and decreased social engagement in daily life. These findings are largely consistent with results in healthy control populations, suggesting that emotion regulation has similar benefits and costs. Future research should look to clarify the temporal dynamic of these relationships to better understand the relationship between regulation, emotional experience, and functioning in daily life.