The relationship between social anxiety, subjective stress and salivary cortisol

Table of Contents

Abstract

Zusammenfassung

Tables

Figures

Introduction
The Present Study

Method
Recruiting and Participants
Experimental Procedure
Measure of Social Anxiety (Independent Variable)
Measure of Subjective Stress (Dependent Variable)
Measure of Cortisol Increase (Proposed Mediator/ Moderator Variable)
Statistical Analyses

Results
Manipulation Check of Stress Induction
Correlational Analyses
Correlation between social anxiety and subjective stress
Correlation between social anxiety and cortisol increase
Correlation between cortisol increase and subjective stress
Intercorrelations
Correlations between baseline measures and other variables
Mediation of Subjective Stress
Moderation of Subjective Stress

Discussion

References

Tables

Table 1. Pearson correlations between the variables measured in the study

Table 2. Regression models for the mediation analysis on subjective stress

Table 3. Regression model for the moderation analysis on subjective stress

Figures

Figure 1. Graphic representation of the proposed mediation model (A) and the proposed moderation model (B)

Figure 2. Sequence of events and timeline of the experimental procedure

Figure 3. Mean levels of subjective stress (A) and salivary cortisol (B) before, during (shaded area), and after the stress phase

Abstract

Socially anxious individuals perceive social interactions as threatening to their safety and consequently exhibit specific physiological and psychological anxiety reactions, referred to as stress responses. Recent studies in patients with social anxiety have shown that pharmacologically induced cortisol increase reduced the psychological stress reactivity to a socio-evaluative stressor. These results suggest a potential mechanism by which cortisol and the psychological stress response interact. Thus, we hypothesized that the relationship between social anxiety and subjective stress would be mediated, or moderated by endogenous cortisol increase during stress. Specially, we hypothesize that cortisol increase acts as a buffer for the psychological stress response. A sample of 120 healthy participants underwent the Trier Social Stress Test for groups (TSST-G; von Dawans, Kirschbaum, & Heinrichs, 2011). Although, we found a direct relationship of social anxiety with both the perceived subjective stress, and the cortisol increase, neither our mediation, nor moderation hypothesis were confirmed.

Zusammenfassung

Soziale Interaktionen werden von sozial ängstlichen Personen als bedrohlich empfunden. Daher reagieren sie in diesen Situationen auch mit spezifischen physiologischen und psychologischen Angst- und Stressreaktionen. Kürzlich durchgeführte Studien mit sozial phobischen Patienten konnten zeigen, dass ein pharmakologisch induzierter Cortisolanstieg die psychische Stressreaktion während eines sozio-evaluativen Stresstests reduziert. Diese Ergebnisse legen einen möglichen Mechanismus nahe, in dem Cortisol und die psychologische Stressreaktion miteinander interagieren. Daher vermuteten wir, dass der Zusammenhang zwischen sozialer Angst und subjektivem Stressempfinden durch den Anstieg von körpereigenen Cortisol während Stress mediiert oder moderiert wird.

In einer Stichprobe von 120 gesunden Männern, führten wir den Trier Social Stress Test für Gruppen durch (TSST-G; von Dawans, Kirschbaum, & Heinrichs, 2011). Obwohl sich ein Zusammenhang zwischen sozialer Angst und mit subjektivem Stressempfinden, wie auch mit dem Anstieg von Cortisol zeigte, konnte weder unsere Mediations- noch Moderationshypothese bestätigt werden.

Introduction

For most people the daily life consists of a concatenation of social interactions: we start the day with breakfast together with our partners, kids, or flat mates. At work, we not only have to deal with our tasks, but also with our colleagues. Furthermore, we like to spend our free time with friends. However, not all social interactions are a source of recreation, but can be challenging and demanding. For example, we may be confronted with the presence of unfamiliar people and subject to scrutiny. Some people perceive these latter kinds of social situations as potentially threatening to their safety and consequently exhibit specific psychological and physiological anxiety reactions, usually referred to as stress responses (Vanin & Helsley, 2008; Bourne, 2010). The necessary condition for the stress response to be elicited is that people do not have the confidence that they possess adequate capabilities for managing or controlling the situation successfully (O'Leary, 1992).

It is known that anxiety states cause physiological changes that enable individuals to respond to threatening situations (Fredrikson, Sundin, & Frankenhauser, 1985; Martel, Hayward, Lyons, Sanborn, Varady, & Schatzberg, 1999; Condren, O`Neill, Ryan, Barett, & Thakore, 2002). While numerous psychological and physiological processes are responsive to stress, one system, the hypothalamic-pituitary-adrenal axis (HPA axis), is centrally involved in regulating physiological responses to stress (Lovallo & Thomas, 2000; Tsigos & Chrousos, 2002)^[1]. The stress-induced activation of this endocrine system results in a series of hormonal adaptations, in particular an increased release of the hormone cortisol (Soravia, 2005; Roelofs, Bakvis, Hermans, Van Pelt, & Van Hon, 2007). This increase of cortisol is known to prepare the organism to cope with the potential dangerous situations elicited by the stressor (Lovallo & Thomas, 2000).

The physiological stress response is accompanied by a behavioral and psychological response, typically described as fear. These anxiety symptoms include shaking, sweating, tension and feelings of physical discomfort and subjective stress. In the special case of the stressor being a social situation, symptoms include blushing, avoiding eye contact, hysterical laughs and feelings of situational uncontrollability or the desire to leave the situation (von Dawans, 2008). In sum, the anxiety induced by the stress response is an adaptive process to the uncontrollable environment with the aim of keeping the body functioning (Kasper, den Boer, & Sitsen, 2003).

As mentioned above, some individuals exhibit a stress response in social situations. If anxiety arises only in response to the presence or anticipation of social-evaluative interaction or performance situations, it is called social anxiousness (Bourne, 2010). In contrast to a low level of social anxiousness, high socially anxious individuals (also called high socially phobic) are more likely to fear being negatively judged or humiliated by others. If social anxiousness reaches a critical level, anxiety symptoms become excessive, maladaptive, inappropriate, and are associated with marked decrements in a person’s quality of life (Vanin & Helsley, 2008). According to the American Psychiatric Association (DSM-IV-TR, 2000) social anxiousness reaches the level of social anxiety disorder, when quality of life is impaired by avoidance behaviors and the considerably inability to handle social encounters without feeling persistently stressed. In sum, degrees of social anxiety severity vary on a continuum from low to high social anxiousness and social anxiety disorder, and differ on the degree of intensity and impact in everyday life (Shirotsuki et al., 2009).

The paradigm typically used to elicit social anxiety is the Trier Social Stress Test (TSST; Kirschbaum, Pirke, & Hellhammer, 1993). In this test, participants have to give a public speech and perform mental arithmetic’s in front of evaluators. These characteristics are known to reliably induce psychosocial evaluative situations, which are the specific phobic stimuli for social anxiety (Dickerson & Kemeny, 2004).

Recently a number of studies have shown a relationship between the severity of social anxiety, cortisol reactivity and the psychological stress response induced by a socio-evaluative stressor. However, the picture is not yet clear. Some studies have found that socially anxious individuals exhibit lower cortisol reactivity but higher psychological stress responses than individuals with low social anxiety. For example, Shirotsuki et al. (2009) administered the TSST to a group of 22 healthy college students, whereat socially anxious individuals exhibit lower cortisol reactivity, but higher psychological stress response than non-socially anxious individuals. On the other hand, other studies show higher cortisol reactivity and higher psychological stress responses for socially anxious individuals. For example, Condren and colleagues (Condren et al., 2002), who administered a procedure similar to the TSST, examined 15 patients with social phobia and 15 healthy controls and found that patients showed a higher increase of cortisol levels and psychological stress response. Furlan and colleagues (Furlan , DeMartinis, Schweizer, Rickels, & Lucki, 2001) found a mixed pattern. In their study with 20 social phobic patients and 17 healthy controls, they found that some patients showed increased cortisol reactivity and other patients impaired, or even absent cortisol reactivity in response to a speech task compared to the controls. However, psychological stress responses were higher in all social patients. A study by von Dawans (2008), who compared 65 social phobic patients with 79 healthy controls, found no difference in cortisol reactivity to the TSST but a difference in psychological stress response (i.e. stronger subjective stress response in social phobic patients). However, von Dawans (2008) also manipulated social support and administered oxytocin to half of her sample, so it is unclear if the physiological responses to stress interacted with these variables.

To shed light on these inconsistent results, and to examine the underlying mechanism behind it, Soravia and colleagues pursued a different approach and induced pharmacological cortisol release. In their first study (Soravia et al., 2006) they administered glucocorticoides (which are metabolized into cortisol) to 20 social phobic patients and a placebo to another group of 20 social phobic patients. They found that pharmacologically induced cortisol release reduced the psychological stress reactivity to the TSST in comparison with the placebo. In their second study (Soravia, De Quervain, & Heinrichs, 2009) they administered glucocorticoides to 24 healthy participants and placebos to 26 participants. This is also the first of the studies mentioned that examined male and female participants, with the other studies only examining males. They found that a pharmacologically induced cortisol release did not reduce the psychological stress response compared with the placebo treatment.

Taken together, these results suggest a potential mechanism by which cortisol reactivity and the psychological stress response interact. Specifically, the studies of Soravia et al. (2006, 2009) indicate that for high-level socially anxious individuals, cortisol reactivity during stress may weaken the psychological stress response. Furthermore, it seems that only standardized socio-evaluative stressors (the TSST; Shirotsuki et al., 2009), and not unstandardized procedures (Condren et al., 2002; Furlan et al., 2001), reliably induce endocrine stress responses.

The Present Study

The present study differs from the previous studies, which have examined the psychological and endocrine responses to a standardized socio-evaluative stressor in socially anxious participants in one important aspect. To our knowledge, this is the first study which investigates whether the endogenous stress-induced release of cortisol (i.e., no administration of glucocorticoides) influences perceived subjective stress in a general population (i.e., non-clinical sample).

[...]

^[1] This work only focuses on the HPA axis as a key system for cortisol release during stress. Thus, other central components of the stress system are not described, as they are not relevant to the present work.