Endogenous oxytocin levels are associated with the perception of emotion in dynamic body expressions in schizophrenia

https://doi.org/10.1016/j.schres.2015.01.022Get rights and content

Abstract

Lower endogenous oxytocin levels have been associated with impaired social cognition in schizophrenia, particularly facial affect identification. Little is known about the relationship between oxytocin and other forms of emotion perception. In the current study, 41 individuals with schizophrenia (SZ) and 22 demographically matched healthy controls (CN) completed a forced-choice affective body expression classification task. Stimuli included dynamic videos of male and female actors portraying 4 discrete emotions: happiness, sadness, anger, and neutral. Plasma oxytocin levels were determined via radioimmunoassay. Results indicated that SZ had significantly higher plasma oxytocin concentrations than CN. SZ were also less accurate at identifying expressions of happiness and sadness; however, there were no group differences for anger or neutral stimuli. A group × sex interaction was also present, such that female CN were more accurate than male CN, whereas male SZ were more accurate than female SZ. Higher endogenous oxytocin levels were associated with better total recognition in both SZ and CN; this association was specific to females in SZ. Findings indicate that sex plays an important role in identifying emotional expressions in body gestures in SZ, and that individual differences in endogenous oxytocin predict emotion perception accuracy.

Introduction

There is consistent evidence for emotion perception abnormalities in individuals with schizophrenia (SZ) (Kohler et al., 2010). These impairments have been observed across a number of stimulus types (e.g., facial, vocal, audio-visual) and task formats (e.g., identification, differentiation, intensity judgment), and are more pronounced for negative (e.g., fear, anger, sadness) than positive (e.g., happiness, surprise) emotions (Edwards et al., 2002, Kohler et al., 2010). Several variables moderate the magnitude of emotion perception deficits in SZ, including severity of negative and positive symptoms, age, illness duration, antipsychotic medications, phase of illness, and inpatient vs. outpatient status (Edwards et al., 2002, Kohler et al., 2010). Recently, lower endogenous oxytocin levels have also been associated with poorer facial affect perception, particularly among females with SZ (Rubin et al., 2011). Intranasal administration of oxytocin has also been shown to improve several aspects of social cognition in SZ, including facial affect perception (Goldman et al., 2011, Pedersen et al., 2011, Averbeck et al., 2012, Davis et al., 2013, Davis et al., 2014, Fischer-Shofty et al., 2013, Gibson et al., 2014, Woolley et al., 2014). In SZ, aberrant oxytocin functioning may therefore be an important biological correlate of emotion perception abnormalities.

However, facial affect identification is only one form of emotion perception. To determine whether associations between oxytocin and emotion perception are specific to facial affect processing, it will be important to administer tasks using emotional stimuli other than faces. Recently, the field of affective science has developed and validated a series of stimulus sets portraying expressions of emotion in body gestures. Healthy individuals can rapidly and accurately identify discrete emotional states from faceless stimuli depicting body expressions of emotion; such stimuli may be ideal for examining whether the association between oxytocin and emotion perception is unique to faces or a generalizable effect. To our knowledge, only one published study has examined emotion perception in body gestures in participants with SZ. Van den Stock et al. (2011) presented participants with static target photographs displaying a faceless male or female actor portraying either sadness, anger, or fear that had to be matched to one of two simultaneously presented static probe images, one of which displayed the same emotion as the target and the other a foil from another affective category. Results indicated that participants with SZ were less accurate and slower than controls at identifying expressions of anger, sadness, and fear. However, Van den Stock et al. (2011) failed to include stimuli representing discrete positive emotions. Therefore, it is unclear whether the valence-related patterns of performance that are observed during facial affect tasks (i.e., impaired perception of negative and intact perception of positive) also occur during the identification of emotion in body gestures. Additionally, it is also unknown whether the perception of emotion in body gestures is associated with endogenous oxytocin levels.

The current study addressed these gaps in the literature by administering a forced-choice affective body expression classification task to a sample of outpatients diagnosed with SZ and demographically matched healthy controls. The following primary hypotheses were made: 1) consistent with Van den Stock et al. (2011) who reported that participants with SZ performed more poorly than controls across all stimulus conditions, a significant main effect of Group was expected; 2) similar to studies examining emotion perception for facial stimuli (see Kohler et al., 2010), participants with SZ were expected to display poorer performance than healthy controls for negatively valenced, but not positively valenced body gestures; and 3) emotion perception impairments were expected to be associated with lower endogenous oxytocin levels in participants with SZ and healthy controls. Several secondary hypotheses were also made in relation to sex differences. It is well documented that females have higher oxytocin levels than males (Carter et al., 2007). Significant associations between endogenous oxytocin and facial affect perception have also been reported in females with SZ, but not males (Rubin et al., 2011). Given these sex-related effects, we hypothesized that females with SZ would evidence significant associations between oxytocin and the identification of emotion in body gestures, and that these associations would be nonsignificant in males.

Section snippets

Participants

Participants included 40 individuals meeting DSM-IV-TR criteria for schizophrenia (SZ) and 22 healthy controls (CN). Individuals with SZ were recruited through the Outpatient Research Program at the Maryland Psychiatric Research Center, and evaluated during a period of clinical stability as evidenced by no changes in medication type or dosage for a period greater than or equal to four weeks. Consensus diagnosis was established via a best-estimate diagnostic approach based on the Structured

Group differences in plasma oxytocin levels

One-way ANOVA revealed that individuals with schizophrenia had significantly higher endogenous oxytocin levels than healthy controls, F(1,60) = 2.84, p < 0.02.

Emotion perception of body gestures

Emotion perception performance for SZ and CN is presented in Fig. 2, panel A. A 2 Group (SZ, CN) × 4 Emotion (Happiness, Sadness, Anger, Neutral) mixed-models ANOVA revealed a significant effect of Group, F (1,61) = 15.6, p < 0.001, and a significant within-subjects effect of emotion, F (3,183) = 13.6, p < 0.001; however, the Group × Emotion interaction

Discussion

The current study presented some of the first evidence that emotion perception deficits in SZ extend beyond facial and vocal channels and also occur when people with SZ are required to identify emotion in body gestures. Specifically, individuals with SZ displayed poorer performance than CN for body gestures depicting happiness and sadness, but not anger or neutral. These findings extend the results of Van Den Stock et al. (2011), which found that SZ were more impaired than CN at identifying

Role of funding source

Research supported in part by US National Institutes of Mental Health Grant P50-MH082999 (WT Carpenter) and a Department of Veterans Affairs Mental Illness Research Education Clinical Centers VISN 5 pilot grant (GP Strauss).

Contributors

Gregory Strauss, William Keller, Robert Buchanan, James Koenig, and James Gold designed the study. Statistical analyses and writing of the first draft of the manuscript were performed by Gregory Strauss. James Koenig and his lab conducted oxytocin radioimmunoassays. All authors contributed to and approved the final manuscript.

Conflict of interest

Authors have no conflicts of interest relevant to the current study.

Acknowledgments

The authors would like to thank the participants who completed the study, as well as staff at the Maryland Psychiatric Research Center who contributed to data collection. We are especially thankful to Dana Brady for processing oxytocin levels and members of Dr. Strauss' team who conducted subject recruitment and testing: Lauren Catalano, Adam Culbreth, Bern Lee, Jamie Adams, Travis White, Tehreem Galani, and Carol Vidal.

References (27)

Cited by (36)

  • Endogenous oxytocin levels are associated with impaired social cognition and neurocognition in schizophrenia

    2019, Journal of Psychiatric Research
    Citation Excerpt :

    Despite these inconsistencies in mean group differences, lower endogenous OT levels have fairly consistently been associated with greater symptom severity (positive and negative) (Rubin et al., 2010, 2011, 2015, 2017; Sasayama et al., 2012; Strauss et al., 2015a; Walss-Bass et al., 2013; however, see Rubin et al., 2013, 2014) and poorer performance on tasks measuring higher-order and lower-level social cognition (e.g., social cue perception, facial affect perception, identification of emotional body gestures, hedonic judgments) (Goldman et al., 2008; Rubin et al., 2011; Strauss et al., 2015a; 2015b, 2015c). Associations between peripheral OT and social cognition may be particularly strong among females with SZ (Rubin et al., 2011; Strauss et al., 2015c). It is unclear whether lower endogenous OT is also associated with poor general cognition in SZ.

View all citing articles on Scopus
View full text