Pre-experience of social exclusion suppresses cortisol response to psychosocial stress in women but not in men

Abstract

Lack of social support and social exclusion is associated with adverse effects for mental and physical health. Additionally, women appear to be more vulnerable to social triggers of health disturbances. The hypothalamus–pituitary–adrenocortical-axis (HPA axis) might play a key role in this context as it has been shown both to relate to psychosocial conditions and health outcomes and to respond differentially depending on gender. In a previous experiment we found no effects of exclusion alone (operationalized via Cyberball) on cortisol secretion. Here we examine the effects of a social exclusion pre-experience on psychological and cortisol responses to a public speaking stressor. Subjects (33 m, 34 f) were randomly assigned to social exclusion (SE) or one of two control conditions (exclusion attributed to technical default (TD) and social inclusion (SI)). Afterwards salivary cortisol and psychological responses to a public speaking paradigm were assessed. Exclusion pre-treatment does not affect psychological responses to public speaking stress though with respect to cortisol significant. Cyberball by gender and Cyberball by gender by time interactions are found. SE-women show a blunted cortisol stress response to public speaking while cortisol responses of SE-men fall between SI-men and TD-men. Pre-experience of social exclusion leads to a blunted cortisol response to stress in women but not in men. This factor might contribute to the higher vulnerability to social triggers of health disturbances observed in women.

Introduction

Becoming involved in social relationships is a fundamental human motivation. Social support has several protective effects; it ameliorates the psychological impact of stressful situations and is associated with positive health outcomes (House et al., 1988). Lack of support and social exclusion, on the other hand, is often observed for instance in depression (Hawthorne, 2008), anxiety disorders (Torgrud et al., 2004), fibromyalgia (Montoya et al., 2004), chronic fatigue syndrome (Prins et al., 2004), cardiovascular disease (Sorkin et al., 2002), cancers and HIV/AIDS (Reynolds and Kaplan, 1990, Leserman et al., 2000). Gender differences are reported in this context, but findings remain inconclusive. Some studies indicate that the beneficial impact of support on health is more pronounced in men than in women (House et al., 1982), whereas others state the opposite (Denton et al., 2004, Berkman and Syme, 1979). However, women seem to be more vulnerable to social triggers of health disturbances (Denton et al., 2004, Troisi, 2001).

Mechanisms mediating between the degree of social integration and physical and mental health are only partially understood. Particularly with regard to physiological mediators, little is known to date. Remarkably, however, many of the aforementioned diseases are also characterized by dysregulation of the HPA axis (Leserman et al., 2000, Burke et al., 2005, Furlan et al., 2001, Wingenfeld et al., 2008, Van Den Eede et al., 2007, Van Den Eede et al., 2008, Koertge et al., 2002, Ehlert et al., 2001, McEwen, 1998). Thus, the question arises whether the degree of social integration interferes with HPA-axis regulation.

The HPA-axis is activated by psychosocial stress resulting in increased release of cortisol from the adrenal glands, with women tending to show a dampened cortisol response due to basal biological regulatory mechanisms (Kajantie and Phillips, 2006, Hellhammer et al., 2009). A few studies indicate that salivary cortisol responses to stress can be reduced by social support (Heinrichs et al., 2003, Ditzen et al., 2007, Ditzen et al., 2008, Kirschbaum et al., 1995). Support by their male partners enhances the cortisol release of women, while men, on the other hand, show a dampened cortisol response when supported by their female mates (Kirschbaum et al., 1995). Data on cardiovascular stress responses suggests that this effect might be attributed rather to the gender of the supporting person than to that of the person being supported. Social support provided by women but not by men reduces cardiovascular stress responses in both women and men (Glynn et al., 1999). Considerable research is still needed with respect to gender differences in the physiological response to social support. Even less is known about the effects of social exclusion or ostracism. In particular, effects of social exclusion on the HPA response to a subsequent stressor have not yet been analysed. Studies assessing direct effects of social exclusion on salivary cortisol (Zwolinski, 2008, Blackhart et al., 2007, Stroud et al., 2002) are rare as well. The results of these studies are equivocal, which, at least in part, might be due to the exclusion paradigm that includes a direct social interaction hardly to standardize.

Indeed, experimental research on exclusion faces the problem of how to induce a standardized exclusion condition. “Cyberball,” a virtual ball game, represents one possibility of inducing social exclusion in a highly standardized manner (Williams et al., 2000). In this paradigm, participants are made to believe that they are playing with three other participants (which in fact are computer generated). During the game the degree of social inclusion (i.e., how often they receive the ball from the other participants) is manipulated: “included” participants receive the ball regularly throughout the game while “excluded” participants receive no further ball after the first throws. Previous research on this paradigm indicates that “excluded” participants not only perceive themselves as being excluded but also suffer from lower self-esteem compared to the included participants (Williams et al., 2000). Furthermore, fMRI studies have shown that the dorsal anterior cingulate cortex, a region also activated while experiencing physical pain, is activated during exclusion (Eisenberger et al., 2003).

In a previous study (Zöller et al., 2010) we analysed the immediate effects of Cyberball exclusion on cortisol secretion and psychological parameters in women. While exclusion increased depression and anger no effects on cortisol were found. In the present study we used the Cyberball paradigm to assess the effects of a social exclusion pre-experience on HPA stress responsiveness to a standard laboratory stressor in men and in women.