Our primary aim was to examine the relationship between endogenous oxytocin and emotion processing in patients with schizophrenia as a follow-up to our earlier demonstration that endogenous oxytocin related to improved symptom severity (Rubin et al. 2010
). Higher average levels of oxytocin were associated with perceiving faces as happier in female patients and controls, but not in men. Despite similar levels of oxytocin, females in this study appeared to benefit more from endogenous oxytocin compared to men. This finding compliments our previous study, where the extent of reduced symptom severity in relation to oxytocin was broader in females (positive, general, and prosocial symptoms) than males (prosocial symptoms only). High oxytocin, especially in females, may help to produce a positive bias in emotion processing and consequently may contribute to improving trust and social interactions in schizophrenia.
Animal studies suggest that oxytocin differentially affects females and males, which may reflect interactions between oxytocin and gonadal hormones (Champagne et al., 2001
; McCarthy et al.,1996
; Razzoli et al., 2003
). Intranasal oxytocin increases amygdala activity during emotional face processing in females (Domes et al., 2010
), but reduces amygdala activity in men (Domes et al., 2007
). Data suggest sex differences in the effects of oxytocin on processing of happy faces; intranasal oxytocin facilitated recognition of happy faces in both men and women in one study (Marsh et al., 2010
), whereas another study in men only found no effect of oxytocin (Di Simplicio et al., 2008
). The present menstrual cycle manipulation allowed us to directly examine whether the relationship between oxytocin and emotion perception depended on the level of circulating estrogen and progesterone. The magnitude of correlation between oxytocin and emotion perception did not differ between menstrual phases, demonstrating that this correlation does not depend on levels of estrogen, progesterone or both. The effects of oxytocin on emotion perception do not appear to depend on interactions with gonadal steroids in either schizophrenia patients or healthy individuals.
A second notable finding in the present study was that emotion recognition fluctuated across the menstrual cycle, and was highest during the follicular phase. Although the inclusion of women with schizoaffective disorders may have contributed to the finding of cycle-related changes in emotion recognition, this finding expands upon previous studies in healthy women in which the effect was similar (Derntl et al., 2008a
). This also complements our previous finding that clinical symptoms fluctuate across the menstrual cycle in women with schizophrenia, and are worst during the follicular phase. However, cycle-related changes in estrogen and progesterone did not explain changes in emotion recognition. Thus, we did not replicate previous reports that estrogen (Guapo et al., 2009
) or progesterone (Derntl et al., 2008a
) relates to the accuracy of emotion recognition across the menstrual cycle in healthy women. The absence of effects of sex steroid hormones (particularly estrogen) on emotion recognition observed in this study may be related to the characteristics of the experimental paradigm employed, which evaluated only two emotions (happiness and sadness) and on a seven-point scale.
Cycle-related changes in emotion processing may reflect ovarian hormone effects on the amygdala; amygdala activation during performance of an emotional recognition task is more extensive in the follicular compared to the luteal phase (Derntl et al., 2008a
). It is interesting that clinical symptoms were worst during the follicular phase when emotion recognition was best. Previous studies suggest that the amygdala is hyperactive during emotion processing in schizophrenia, which relates to worse symptoms (Rauch et al., 2010
). Future studies might explore whether heightened activation of the amygdala during the follicular phase relates to both improved emotion recognition and higher levels of symptom severity. A broad psychological hypothesis is that higher levels of emotional reactivity may contribute to symptom severity and perhaps risk for relapse. Regardless of mechanism of action, oxytocin may have beneficial effects on emotional function, and potentially adverse effects on psychiatric symptomatology. This latter possibility might be considered in future clinical trials of oxytocin.