Edited by: Tom Johnstone, University of Reading, UK
Reviewed by: Alessio Avenanti, Alma Mater Studiorum - University of Bologna, Italy; Ignazio Puzzo, University of Reading, UK
*Correspondence: Gabriele Buruck, Department of Psychology, Institute of Work, Organizational and Social Psychology, Technische Universität Dresden, Zellescher Weg 17, Dresden 01069, Germany e-mail:
This article was submitted to Emotion Science, a section of the journal Frontiers in Psychology.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Psychosocial stress affects resources for adequate coping with environmental demands. A crucial question in this context is the extent to which acute psychosocial stressors impact empathy and emotion regulation. In the present study, 120 participants were randomly assigned to a control group vs. a group confronted with the Trier Social Stress Test (TSST), an established paradigm for the induction of acute psychosocial stress. Empathy for pain as a specific subgroup of empathy was assessed via pain intensity ratings during a pain-picture task. Self-reported emotion regulation skills were measured as predictors using an established questionnaire. Stressed individuals scored significantly lower on the appraisal of pain pictures. A regression model was chosen to find variables that further predict the pain ratings. These findings implicate that acute psychosocial stress might impair empathic processes to observed pain in another person and the ability to accept one's emotion additionally predicts the empathic reaction. Furthermore, the ability to tolerate negative emotions modulated the relation between stress and pain judgments, and thus influenced core cognitive-affective functions relevant for coping with environmental challenges. In conclusion, our study emphasizes the necessity of reducing negative emotions in terms of empathic distress when confronted with pain of another person under psychosocial stress, in order to be able to retain pro-social behavior.
Pain comprises manifold sensory, affective, and cognitive experiences that often mirror personal life events and depend on individual differences. Hence, for an observer wishing to understand a person's pain and to empathize with the other's feelings, this complexity constitutes a great challenge. Considering this, the social communications model of pain (Hadjistavropoulos and Craig,
“Empathy in the broadest sense refers to the reactions of one individual to the observed experiences of another” (Davis,
Previous research concludes that empathic reactions to pain in others are generated by two distinct ways. A stimulus-response, perception-based route is triggered in the presence of concrete visual stimuli depicting, for example, other people or body parts in painful situations. Additionally, in situations where such direct perceptual evidence is missing, affective states of others can be inferred by the creation of representations of the other's potential mental state, which constitutes a more abstract, inferential route (Singer and Lamm,
The perception-based route activates core empathy-related brain networks via simulation of the affective state observed through the engagement of action-perception networks (Preston and De Waal,
A partial neural overlap between the experience of pain in the Self and the observation of pain in others has been reported in the somatosensory cortex/posterior insula, which is associated with the sensory discriminative dimension of pain (Avenanti et al.,
However, as stated above, empathy includes perspective taking, Self-other distinction, and knowledge of the causal relation between one's own and the other's affective state (Walter,
To summarize, empathy for pain does not rely on a full overlap between the Self and the other. In contrast, experiencing another person's pain or distress in the same way as one's own pain experience would lead to an “empathic over-arousal” (Eisenberg,
Hence, “the best response to others' distress may not be distress, but efforts to soothe that distress” (Decety and Lamm,
Following this, emotion regulation provides the opportunity to modulate our emotional experience and behavior. It refers to a set of different strategies by which individuals “influence which emotions we have, when we have them and how these emotions are experienced or expressed” (Gross,
Research about the effects of stress on empathy (for pain) primarily focuses on problems associated with chronic stress in health care employees. In this sector, studies show that empathy is blunted by stressors such as high workload, exposure to suffering patients or patient death, and ethical conflicts (Koehl-Hackert et al.,
The influence of stress on empathy can partly be explained by stress-related effects on the prefrontal cortex because of its involvement in processes concerning emotion regulation, working memory, self-regulatory processes, and goal-directed behavior (Miller,
Neurophysiological studies have repeatedly shown that voluntary top–down emotion regulation by different cognitive strategies rests upon executive control regulating (negative) emotions by an activation of brain regions like the PFC and the parietal cortex. These regulation processes modulate the emotional experience processed by bottom-up emotion processing structures, e.g., the amygdala (Ochsner et al.,
Supporting this, Decety and Meyer (
As mentioned above, there is a lack of experimental studies on the potentially adverse effects of acute psychosocial stress on empathic reactions to other persons in pain. Therefore, the aim of the present study was to fill this gap by analyzing the influence of acute psychosocial stress on ratings to the observed pain in another person in an experimental setting by means of a pain paradigm (Jackson et al.,
Specifically, we investigated whether and in what way (1) acute psychosocial stress influences pain ratings to another person's pain, (2) acute psychosocial stress and individual differences in emotion regulation skills predict individual differences in pain ratings to another person's pain, and (3) the influence of stress on pain ratings to another person's pain is modulated by emotion regulation skills.
The selection of participants occurred non-randomizedly (convenience sampling). Furthermore, there were few exclusion criteria for participation in the study such as health issues, language barriers, or prior experience with the Trier Social Stress Test (TSST). Two female participants had to be excluded based on psychological health issues. None had to be excluded due to language barriers or prior experience with the TSST. Thus, 120 subjects participated in the study and were randomly assigned to a Stress-Group vs. a control group (Placebo-Group). However, 14 subjects had to be excluded from the statistical analyses because the assessment of the pain ratings—our key dependent variable—failed due to technical artifacts. Furthermore, two subjects (age > 40 years) were identified as outliers with regard to their age. Out of the remaining 104 participants 52 were assigned to each group (Stress-Group:
The research design was a randomized, standardized, multivariate experimental/control group comparison with pre-/post-/repeat-measurement. The participants were randomly assigned to one of the two experimental conditions [Trier Social Stress Test (TSST) = Stress-Group vs. Placebo-TSST = Placebo-Group]. A manual with a precise description of the lab procedure was developed. Two laboratory assistants were trained in the TSST and in multiple trial runs before commencing the study. The main examination lasted approximately 90 min and was conducted in the behavioral observation laboratory of the Institute for Work, Organizational, and Social Psychology at the TU Dresden. The experimental procedure is illustrated in Figure
The TSST was developed by Kirschbaum et al. (
As a psychophysiological manipulation check for the stress induction, we assessed the cardiovascular reaction. Heart rate data was continuously sampled beat-to-beat during the complete experimental procedure by the Polar © RS800cx (Polar Electro Oy, Kempele, Finland) heart rate monitor. Acceptable validity and reliability of these devices has been demonstrated (Goodie et al.,
The MDBF (Multidimensional mood questionnaire; Steyer et al.,
Ratings of the perceived pain in another person were measured by means of a paradigm proposed by Jackson et al. (
Headphones were worn by the participants in order to minimize interfering acoustic stimuli. As the instrument demands quick reactions from the participants, effects due to social desirability are minimized.
Emotion regulation skills were measured by the SEK-27 (Self-Report Measure for the Assessment of Emotion Regulation Skills). The SEK-27 was developed by Berking and Znoj (
Since females showed a trend toward higher pain ratings (on painful pictures) than males [
To test for differences in pain ratings and reaction times between the different pain intensities of the pictures we conducted repeated measure analyses of covariance (ANCOVA) with pain ratings or reaction times, respectively, as dependent variable, pain Intensities as repeated measures factor (non-pain pictures, pain pictures with intensity 1, pain pictures with intensity 2, pain pictures with intensity 3), Group (Stress vs. Placebo) as independent variable and gender as covariate.
For all subsequent analyses, the ratings to painful pictures were combined across the three different pain intensities resulting in two picture categories as repeated measures factor (painful and non-painful pictures).
In order to investigate whether our stress induction via the TSST was effective (dependent variables: MDBF mood, MDBF alertness, MDBF calmness, and heart rate) and to analyze the influence of stress on pain ratings (dependent variable), we used repeated measures ANCOVAs. The ANCOVAs included Group (Stress vs. Placebo) as independent variable, gender as covariate and Time (preTSST, postTSST, postPain) or Pain (non-painful pictures, painful pictures) as repeated measures factor, respectively.
To analyze associations between the covariates age, gender, trait empathy, and habitual emotion regulation, the independent variable Group (Stress vs. Placebo), as well as between the predictor emotion regulation skills and pain ratings to painful pictures, non-parametrical correlation analyses (Kendall's Tau) were performed. This step was included in order to select potential predictor variables for the following regression analyses. Variables were included when correlation coefficients were significant at least at
We used moderated regression analysis to test the potential moderating role of emotion regulation skills with the
First, we analyzed whether the participants gave pain ratings according to the pain intensities depicted in the presented stimuli. Results showed that the pain ratings increased with the presented pain intensity [Main Effect Intensities
In comparison to the Placebo-Group, the Stress-Group reported a significant reduction in mood from preTSST to postTSST [Interaction Time × Group:
Additionally, the participants assigned to the Stress-Group showed a higher increase in heart rate during stress induction than participants in the Placebo-Group [Interaction Time × Group:
Whereas pain pictures were rated significantly more painful than non-painful pictures [Main Effect Pain:
First, we used a partial correlational analysis (adjusting for gender) to examine the specific role of cardiovascular activation for the prediction of the pain ratings to pain pictures. However, individual heart rate during postTSST (
Subsequently, to select potential predictors of the pain ratings to pain pictures in addition to psychosocial stress, we conducted a further correlational analysis. There were small, but significant by trend correlations of the pain judgments during painful situations with gender, Group (Stress vs. Placebo-Group), and the emotion regulation skill clarity (
Gender ( |
−0.15 |
1 | ||||||||
Age ( |
0.02 | −0.01 | 1 | |||||||
Fantasy | −0.03 | −0.22 |
0.12 | 1 | ||||||
Perspective taking | −0.12 | −0.05 | −0.05 | 0.20 |
1 | |||||
Empathic concern | −0.08 | −0.26 |
−0.14 | 0.32 |
0.11 | 1 | ||||
Personal distress | 0.12 | −0.33 |
0.05 | 0.12 | −0.13 | 0.17 |
1 | |||
Reappraisal | −0.05 | −0.13 | −0.07 | 0.03 | 0.25 |
−0.02 | −0.14 |
1 | ||
Suppression | 0.12 | 0.28 |
0.10 | −0.20 |
−0.04 | −0.30 |
−0.17 | 0.05 | 1 | |
Stress vs. Placebo ( |
−0.15 |
0.03 | −0.06 | −0.10 | 0.10 | 0.02 | −0.00 | 0.01 | −0.10 | 1 |
Awareness | 0.02 | −0.32 |
−0.06 | 0.11 | 0.11 | 0.22 |
0.06 | 0.08 | −0.20 |
0.03 |
Body sensations | −0.03 | −0.16 |
−0.23 |
0.07 | 0.14 | 0.13 | 0.05 | 0.09 | −0.15 |
0.06 |
Clarity | −0.13 |
−0.14 | −0.21 |
0.07 | 0.14 | 0.04 | −0.06 | 0.09 | −0.21 |
0.03 |
Understanding | −0.06 | −0.12 | −0.20 |
0.10 | 0.14 | 0.06 | −0.00 | 0.00 | −0.24 |
−0.07 |
Modification | −0.07 | 0.05 | −0.05 | −0.03 | 0.25 |
−0.02 | −0.18 |
0.25 |
−0.01 | −0.00 |
Acceptance | −0.17 |
−0.01 | −0.17 |
0.08 | 0.18 |
0.06 | −0.19 |
0.11 | −0.16 |
−0.04 |
Tolerance | −0.15 |
0.12 | −0.09 | −0.00 | 0.25 |
−0.05 | −0.26 |
0.14 |
−0.00 | −0.01 |
Self-support | −0.02 | −0.09 | −0.10 | 0.05 | 0.24 |
−0.01 | −0.14 | 0.39 |
−0.04 | 0.01 |
Readiness to confront | 0.05 | −0.13 | −0.09 | −0.05 | 0.11 | 0.00 | −0.16 |
0.30 |
0.01 | 0.00 |
Subsequently, based on the intercorrelations (Table
Firstly, we computed interaction terms between Group and the mean centered value for each emotion regulation skill (SEK27) and included these terms in the correlation analysis. This resulted in small, but significant correlations of the pain ratings on painful pictures with Group × clarity (
Secondly, we conducted a moderated regression analysis with the covariate gender, Group as independent variable, emotion regulation skills as predictors (clarity, understanding, acceptance, tolerance), and with the respective interaction terms between group and emotion regulation skill. Our assumption concerning the potential moderating effect of specific emotion regulation skills on the relation between acute psychosocial stress and pain ratings had to be rejected (
The aim of the present study was to analyze the influence of acute psychosocial stress and emotion regulation skills on the judgments of another person's pain—an indicator of empathic feelings. We found that after inducing psychosocial stress, participants rated pictures of others in painful situations significantly less painful than participants that did not undergo a stress induction. In addition to the effect of stress induction, individual differences in the ability to regulate one's own emotions by acceptance predicted the judgment of pain in others. To be specific, subjects that reported a higher acceptance rated other people's pain as significantly lower. Moreover, our results further suggest that the ability to tolerate negative emotions modulated the association between stress and pain ratings.
In this final section, the findings of the experimental study are summarized and discussed. Firstly, we will discuss the influence of acute psychosocial stress on the appraisal of pain others and how this relates to empathic processing. Subsequently, we integrate and interpret the findings of the regression and moderation analyses. Finally, research limitations are discussed and suggestions for further research are provided.
In the present study, we attempted answering the question whether the experience of psychosocial stress influences the empathic reaction to perceived pain in others. Referring to prior research (Smeets et al.,
Furthermore, our study demonstrates that participants who had been exposed to acute psychosocial stress, and accordingly showed higher ratings concerning their subjective stress experience as well as elevated physiological responses, revealed significantly lower values in their pain ratings. This effect could not be observed in participants assigned to the non-stress control condition.
It has been shown that negative emotions increase pain unpleasantness ratings on own pain (Villemure et al.,
There are several factors that might contribute to the differences between our findings and those of the cited studies. First, almost all of the studies mentioned above did not investigate the observation of pain in others. Second, they reported results on unpleasantness ratings while we found effects on intensity ratings which might tap into different processes. Finally, we did not observe any association between mood and pain ratings. Hence, the effect of psychosocial stress on the intensity ratings of another person's pain may constitute a specific empathy related process, which is independent of the emotional state of the Self. Consistent with our results, a study by Guo et al. (
Research considering the effects of stress on empathy predominantly focuses on problems associated with chronic stress in health care employees (Koehl-Hackert et al.,
The results of our study show that acute stress significantly impacts the appraisal of pain in another person, which can be interpreted as an effect on empathic feelings. Although the reductions in pain ratings in the stressed as compared to the non-stressed group are comparatively small, they are particularly relevant insofar as the stress induction was only short-termed as confirmed by psychological and physiological measurements. Thus, our results strongly confirm prior findings on consequences of chronic stress using a cost-efficient experimental setting to test the relations between acute psychosocial stress and empathy for pain. We also assume that the contribution of the predictors (emotion regulation skills acceptance and tolerance) to individual differences in in the pain ratings remained low, because we used an acute stress paradigm to induce stress reactions in our participants. As compared to chronic stress experiences that last for months or even years, acute psychosocial stress is not only much shorter, but also better manageable for most subjects.
It could be argued that the reducing effect of stress on the appraisal of pain in others may be explained by the detrimental influences of stress on PFC functions, which have been shown to impair social cognition (Smeets et al.,
Our next question concerned the potential role of acute psychosocial stress and specifically associated emotion regulation skills as predictors of the pain ratings, with emotion regulation skills being conceptualized as prolonged state (current state including the previous week). By means of a stepwise multiple regression analysis, the impact of the potential predictors was tested. Overall, 10% of the variance in the ratings to pain in others was explained by (objective) acute stress and the emotion regulation skill acceptance. In detail, acute psychosocial stress reduced the pain ratings, as already discussed above. In addition, the higher participants rated their ability to use acceptance as emotion regulation strategy, the lower they rated pain in others.
Kohl et al. (
Following this, to extend our findings on the direct effects of stress and emotion regulation, we will subsequently discuss our results on indirect effects. Thereby, we will consider interactions between stress and emotion regulation skills with regard to their impact on the appraisal of pain in others, which might provide additional implications for pain empathy.
Decety (
Our results further strengthen the importance of acceptance/tolerance as emotion regulation strategies which refuse to focus on altering one's negative emotions. Strikingly, one of the two psychopathology-related features of reappraisal is termed emotional resistance or not-acceptance of emotional events (Werner and Gross,
As mentioned above, the contribution of the predictors to the pain ratings is comparably low, which is probably due to the application of an acute stress paradigm. We also assume that the time between stress induction and pain measurements was too long to observe more profound effects. Future studies should prefer conducting an empathy paradigm during or shortly after stress induction. Above, potential variables that might be of further predictive value like, for example, current chronic stress, development of empathic behavior, depression, anxiety, or emotional repertoire (see also De Vignemont and Singer,
Another limitation of the empathy for pain paradigm used in our study is that it merely consists of pictures showing hand and feet under painful stimulation. An empathy paradigm in which the subjects are confronted with more emotional cues such as movements or gestures and facial expressions might have resulted in higher emotional responses. Another limitation may concern the reference point within the pain ratings which might differ between the participants: It was left up to the participants whether they related the task to an episode of pain experienced at present, or to anticipated pain occurring in the future.
It has been shown that women show stronger reactions in emotion studies (Domes et al.,
In view of the experimental setting applied in this study, we assume a high internal validity of the analysis resulting in a low likelihood of alternative explanations for the findings demonstrated here. By the randomized assignment of the participants to the experimental conditions individual confounding factors are controlled for. In addition, further potential confounders (e.g., health) were eliminated by taking additional measures prior to randomization.
A further limitation results from the sampling method applied here: We used a convenience sample of students taking part voluntarily rather than a probabilistic sample. In addition, participants were aware of taking part in a scientific study, which may have been associated with several behavioral changes. Finally, our sample was very homogenous concerning age and education (i.e., largely psychology students). This could have limited the variance in stress reaction and empathy for pain.
To our knowledge, the present study is the first that experimentally supports a direct effect of acute psychosocial stress on reactions to pain in others, drawing from an interdisciplinary scholarly perspective on stress, emotion and social (neuro-) science. Moreover, our results emphasize the important role of functional emotion regulation for a healthy reaction to other peoples' pain and, hence, for empathy for pain (i.e., how an “observer” understands the pain of a “sufferer;” Singer et al.,
It has also been discussed that empathic sharing of negative feelings might raise the vulnerability to stress and negative emotions. Thus, by comparing empathy training with compassion training it could be shown that compassion is crucial in counteracting the activation of negative emotions (Klimecki et al.,
Underestimations of the experienced pain intensity in another person by a caregiver or significant other carry the risk of the person in pain feeling misunderstood or, more importantly, the risk of increasing the physiological harm to that person (Hadjistavropoulos and Craig,
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The study was funded by the Berufsgenossenschaft für Gesundheitsdienst und Wohlfahrtspflege, (BGW Dresden, Prof. Richter) and was partly supported by the Deutsche Forschungsgemeinschaft (SFB 940, project A5). We want to thank the TSST-Team, Nina Kavaldjeva, and Marit Zimmermann for assistance in data acquisition. Furthermore, we thank Jens Helmert for his assistance in data analysis.