Insights on Modelling Physiological, Appraisal, and Affective Indicators of Stress using Audio Features.

Stress is a major threat to well-being that manifests in a variety of physiological and mental symptoms. Utilising speech samples collected while the subject is undergoing an induced stress episode has recently shown promising results for the automatic characterisation of individual stress responses. In this work, we introduce new findings that shed light onto whether speech signals are suited to model physiological biomarkers, as obtained via cortisol measurements, or self-assessed appraisal and affect measurements. Our results show that different indicators impact acoustic features in a diverse way, but that their complimentary information can nevertheless be effectively harnessed by a multi-tasking architecture to improve prediction performance for all of them.

The association between genetic variability in the NPS/NPSR1 system and chronic stress responses: A gene-environment-(quasi-) experiment.

The neuropeptide S (NPS) and its receptor (NPSR1) have been implicated in stress regulation and stress-related disorders. The present study aimed at investigating the association between overall genetic variability in the NPS/NPSR1 system and psychological and cortisol stress regulation in everyday life. Our study was conceptualized as a gene-environment-(quasi-) experiment, a design that facilitates the detection of true GxE interactions. As environmental variable, we used the preparation for the first state examination for law students. In the prospective and longitudinal LawSTRESS project, students were examined at six sampling points over a 13-months period. While students who prepared for the exam and experienced long-lasting and significant stress, formed the stress group, law students experiencing usual study-related workload were assigned to the control group. As phenotypes we assessed changes over time in the cortisol awakening response (CAR; n = 176), perceived stress levels (n = 401), and anxiety symptoms (n = 397). The CAR was assessed at each sampling point immediately upon awakening and 30 as well as 45 min later. Perceived stress levels in daily life were measured by repeated ambulatory assessments and anxiety symptoms were repeatedly assessed with the anxiety subscale of the Hospital Anxiety and Depression Scale. With gene-set analyses we examined the joint association of 936 NPS/NPSR1 single nucleotide polymorphisms with the phenotypes to overcome well known limitations of candidate gene studies. As previously reported, we found a blunted CAR during the exam as well as significant increases in perceived stress levels and anxiety symptoms until the exam in the stress group, compared to the control group. The gene-set analysis did not confirm associations between genetic variability in the NPS/NPSR1 system and changes in perceived stress levels and anxiety symptoms. Regarding the CAR, we found a significant GxE interaction for the area under the curve with respect to the ground (p = .050) and a trend towards a significant effect for the area under the curve with respect to the increase (p = .054). When the analysis was restricted to the SG, associations for both CAR parameters were significant (ps < .050). This finding suggests that the association between genetic variability in the NPS/NPSR1 system and the CAR becomes visible under the environmental condition 'chronic stress exposure'. We conclude that the present study complements findings from animal models and that it provides novel evidence for a modulatory influence of the NPS/NPSR1 system on cortisol regulation in humans.

Daily life stress and the cortisol awakening response over a 13-months stress period - Findings from the LawSTRESS project.

The LawSTRESS project is a controlled prospective-longitudinal study on psychological, endocrine, central nervous and genetic predictors of responses to long-lasting academic stress in a homogenous cohort. In this first project report, we focused on the association between daily life stress and the cortisol awakening response (CAR). The CAR, a distinct cortisol rise in the first 30-45 min after morning awakening, is a well-established marker of cortisol regulation in psychoneuroendocrinology. Law students from Bavarian universities (total n = 452) have been studied over a 13-months period at six sampling points starting 12 months prior exam. The stress group (SG) consisted of students experiencing a long-lasting and significant stress period, namely the preparation for the first state examination for law students. Law students assigned to the control group (CG) were studied over an equally long period without particular and sustained stress exposure. To investigate stress related alterations in the CAR, we examined a subsample of the LawSTRESS project consisting of 204 students with 97 participants from the SG (69.1% female, mean age = 22.84 ± 1.82) and 107 from the CG (78.5% female, mean age = 20.95 ± 1.93). At each sampling point, saliva samples for cortisol assessment were collected immediately upon awakening and 30 as well as 45 min later. Perceived stress in daily life was measured by repeated ambulatory assessments (about 100 queries over six sampling points). The time course of perceived stress levels in the two groups differed significantly, with the SG showing an increase in perceived stress until the exam and a decrease thereafter. Stress levels in the CG were relatively stable. The CAR was not significantly different between groups at baseline. However, a blunted CAR in the SG compared to the baseline measure and to the CG developed over the measurement timepoints and reached significance during the exam. Remarkably, this effect was neither associated with the increase in perceived stress nor with anxiety and depression symptoms, test anxiety and chronic stress at baseline. The present study successfully assessed multidimensional stress trajectories over 13 months and it documented the significant burden, law students preparing for the first state examination are exposed to. This period was related to a blunted CAR with presumed physiological consequences (e.g., on energy metabolism and immune function). Mean psychological stress levels as well as the CAR returned to baseline levels after the exam, suggesting a fast recovery in the majority of the participants.

Sustained threat and phasic fear in the laboratory and cognitive-emotional processes of anxiety in everyday life - An ambulatory assessment study.

Fear is a phasic state of apprehension to an imminent threat, whereas anxiety is a more sustained state of expecting a potential threat leading to tension and worry. The NPU-threat test is a laboratory startle paradigm allowing a reliable and valid assessment of both, fear- and anxiety-potentiated reactions. It is suggested to differentiate between anxiety disorders, but little is known on associations with everyday life experiences of cognitive-emotional processes regarding anxiety in non-clinical samples. In the present project, the NPU-threat test was applied in three studies with (1) unselected healthy individuals, (2) participants with extreme manifestations of trait anxiety (low vs. high) and (3) individuals preparing for a high-stakes exam. Self-reported states of emotionality and worry were assessed during a four-day ambulatory assessment (AA). Overall, NPU-threat test measures did not significantly differ between studies, while the AA dependent measures were sufficiently sensitive to capture differences between groups. However, there was no significant association between psychophysiological measures of the NPU-threat test and AA state measures across participants. In participants recruited for low vs. high trait anxiety we found an association with AA worry and emotionality, but no interaction with potentiated startle. The present findings do not support the idea of a link between our laboratory biomarker and adaptive regulation of cognitive-emotional states in everyday life in healthy individuals. We speculate that an association between laboratory physiological measures and everyday experience of anxious states may be detectable in clinical samples.

Effect of sugar administration on cortisol responses to acute psychosocial stress.

Sugar administration prior acute psychosocial stress exposure was shown to enhance subsequent salivary cortisol responses. However, this finding is based on studies that have administered high doses of glucose to male subjects after long fasting periods. Therefore, in the present study, we investigated the effect of different sugar-containing drinks on acute cortisol stress responses under experimental conditions that are commonplace in stress research and our sample included females and males. Our primary aim was to derive feasible recommendations for a standardized sugar administration in future studies. Of the 103 healthy young participants (49 females, 54 males), 72 were confronted with the Trier Social Stress Test after being randomly assigned to one of three sugar conditions (200 ml of grape juice, a 75 g glucose or a 75 g maltodextrin drink); 31 subjects served as control sample and were exposed to the TSST without sugar administration. Cortisol stress responses were significantly enhanced in the grape juice as well as the glucose group as compared to the control group. Post hoc analysis revealed that this effect seemed to be more pronounced in males than in females. We did not find a significant effect of maltodextrin. Cortisol responder rates in all three experimental groups were higher than in the control group. Our results suggest that, at least in males, the administration of 200 ml of grape juice is sufficient to facilitate HPA axis reactivity and to minimize confounding effects due to interindividual differences in energy availability while being exposed to a laboratory stress paradigm. The unexpected gender-specific effect is of potential relevance and should be scrutinized in future studies.

HPA axis responses to psychological challenge linking stress and disease: What do we know on sources of intra- and interindividual variability?

Stress is an ubiquitous phenomenon with significant impact on human physiology when it lasts too long, when it is too intense, or when it hits vulnerable individuals. Examining the mechanisms linking stress exposure with health and disease is an important endeavor in psychoneuroendocrine research. Empirical evidence so far revealed large intra- as well as inter-individual variability in hypothalamic-pituitary-adrenal (HPA) axis responses to acute psychosocial stress, showing that the HPA axis is a highly adaptive system. Thus, the characterization of intra- und inter-individual patterns of HPA axis reactivity is of high scientific interest and forms the basis on which mechanistic links between stress response (dys)regulation and health impairments can be examined. To date, basic knowledge has been, and still is, accumulated on demographic, biological (including genetic and epigenetic) factors, lifestyle behavioral variables, consumption of substances and medication, psychological and personality factors, as well as on methodological aspects. Besides this, there is also very recent progress in respect to the development of laboratory stress paradigms that can be applied in virtual reality or inside an MRI-scanner. In sum, the present review updates our current knowledge on moderating and intervening factors as sources of intra- und inter-individual variability in human cortisol stress responses and offers recommendations for future research designs.

Acute psychosocial stress and everyday moral decision-making in young healthy men: The impact of cortisol.

In everyday life, moral decisions must frequently be made under acute stress. Although there is increasing evidence that both stress and cortisol affect moral judgment and behavior as well as decision-making in various domains unrelated to morality, surprisingly few attempts have been made to explore the effects of stress on everyday moral decision-making. Therefore, in the present study, we exposed 50 young healthy men to the Trier Social Stress Test (TSST) or its non-stressful placebo version (PTSST). We investigated the impact of acute stress exposure and stress-related cortisol levels on decision-making, decision certainty, and emotions in 28 everyday moral conflict situations with altruistic versus egoistic response alternatives. Results showed that the TSST-exposed group made more altruistic decisions than the non-stress control group, while groups did not differ in decision certainty and emotion ratings. Moreover, in correlational as well as regression analyses, additionally controlling for confounding variables, we observed significant positive associations between cortisol levels and altruistic decision-making. Further analyses revealed that altruistic decisions came along with significantly higher decision certainty and significantly more positive emotion ratings than egoistic decisions. Notably, our data also raise the idea that the personality trait agreeableness plays an important role in everyday moral decision-making. In sum, our findings provide initial evidence that both acute stress exposure and cortisol levels have prosocial effects on everyday moral decision-making in young healthy men.

Sex-specific association between functional neuropeptide S receptor gene (NPSR1) variants and cortisol and central stress responses.

The brain neuropeptide S (NPS) system has recently generated substantial interest and may be of major relevance for central stress regulation. The NPS receptor (NPSR1) is highly expressed in the limbic system, exogenous NPS exerts pronounced anxiolytic and fear-attenuating effects in rodents and extensive close crosstalk between the NPS system and the hypothalamic-pituitary-adrenal (HPA) axis has been demonstrated. In humans, associations between NPSR1 variants and anxiety and panic disorder, as well as amygdala responsiveness to fear- relevant faces and prefrontal cortex activity in a fear conditioning paradigm have been reported. Moreover, a NPSR1 sequence variant was found to be associated with cortisol stress responses in males. Here, we performed a haplotype-based analysis covering three functional NPSR1 single nucleotide polymorphisms in the promoter (rs2530547), in exon 3 (rs324981) and exon 6 (rs727162) in 277 healthy subjects who were exposed to the Trier Social Stress Test (TSST). A significant sex-specific association with salivary cortisol responses to acute psychosocial stress was detected for the common TTC haplotype 2 (frequency of about 20%). In an additional study using an imaging genetics approach, 65 healthy subjects were exposed to a stress paradigm for scanner environments ("ScanSTRESS"). We found a significant and, again, sex-specific interaction between rs324981 (whose minor T-allele is harbored by haplotype 2) and the neural stress response in a cluster close to the parahippocampal gyrus (whole brain corrected). Moreover, as in the TSST sample, NPSR1 variation was associated with salivary cortisol responses (on a trend level) in a sex-specific way. In summary, our preliminary findings in two independent cohorts exposed to different stress paradigms suggest that the NPS system significantly influences acute stress responses and that sequence variation in NPSR1 may contribute to sex differences in stress regulation.

Testing the ecological validity of the Trier Social Stress Test: Association with real-life exam stress.

The Trier Social Stress Test (TSST) is the most widely used laboratory stress protocol in psychoneuroendocrinology. Despite its popularity, surprisingly few attempts have been made to explore the ecological validity of the TSST. In the present study, 31 young healthy subjects (24 females) were exposed to the TSST about 4 weeks before completing an oral exam on a separate day. Salivary cortisol levels increased significantly in response to both stimuli (TSST: F(2.21, 66.33)=5.73, p=0.004; oral exam: F(1.98, 59.28)=4.38, p=0.017) with similar mean response curves and significant correlations between cortisol increases and areas under the response curves (increase: r=0.67; AUC: r=0.56; both p≤0.01). Correspondingly, changes in positive and negative affect did also show significant correlations between conditions (increase: positive affect: r=0.36; negative affect: r=0.50; both: p≤0.05; AUC: positive affect: r=0.81; negative affect: r=0.70; both p≤0.01) while mean time course dynamics were significantly different (positive affect: F(2.55, 76.60)=10.15, p=0.001; negative affect: F(1.56, 46.82)=23.32, p=0.001), indicating that the oral exam had a more pronounced impact on affect than the TSST. Our findings provide new evidence for the view that cortisol as well as subjective stress responses to the TSST are indeed significantly associated with acute stress responses in real life.