https://orcid.org/0000-0001-9611-9759
Figures
Abstract
Susceptibility to emotional contagion is defined as the disposition of how susceptible someone is to catch others’ emotions and it has long been studied in research on mental health, well-being, and social interaction. Given that existing self-report measures of susceptibility to emotional contagion have focused almost exclusively on negative emotions, we developed a self-report measure to assess the susceptibility to emotional contagion of both positive and negative emotions (2 scales). In two studies, we examined their factor structure, validity, and reliability using exploratory factor analysis (Study 1, N = 257), confirmatory factor analysis (Study 2, N = 247) and correlations. Our results confirmed the two-factor structure and demonstrated good internal consistencies. Regarding external validity, our scales showed diverging correlational patterns: While susceptibility to negative emotional contagion was linked to mental health problems and negative emotions, susceptibility to positive emotional contagion was linked to interpersonal functioning and prosocial tendencies. In conclusion, our scales appear to be internally/externally valid and a promising tool for future research.
Citation: Marx AKG, Frenzel AC, Fiedler D, Reck C (2024) Susceptibility to positive versus negative emotional contagion: First evidence on their distinction using a balanced self-report measure. PLoS ONE 19(5): e0302890. https://doi.org/10.1371/journal.pone.0302890
Editor: Majed Sulaiman Alamri, University of Hafr Al-Batin, SAUDI ARABIA
Received: June 1, 2023; Accepted: April 15, 2024; Published: May 14, 2024
Copyright: © 2024 Marx et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All data and code necessary to reproduce the findings presented in this work are available in an OSF repository (see https://osf.io/jfxk4/) together with additional supplemental material and Study 2 has been preregistered prior to conducting any analyses (see https://aspredicted.org/nk8zx.pdf).
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
In a medical context, contagion is defined as transmission of diseases from one individual to another [1]. Accordingly, the term emotional contagion, in a psychological context, is used to describe the transmission of emotions between two or more individuals [2–9]. In addition to this interpersonal process, a trait-like disposition of how susceptible a person is to catch others’ emotions has been proposed [5,6]: An individual’s susceptibility to emotional contagion (SEC). In recent years, scientific interest in SEC has been growing; and previous studies have repeatedly linked an individual’s SEC to increased negative emotions and mental health problems using self-report data [10,11]. However, the majority of items in existing self-report scales on SEC refer almost exclusively to the contagion of negative emotions. Thus, previous findings on SEC seem to be biased towards the contagion of negative emotions, and the SEC of positive emotions has been heavily neglected as of yet. To tackle this issue, the present contribution had two important goals. First, we sought to create and validate a balanced self-report measure to assess an individual’s SEC of both positive and negative emotions separately. Second, we aimed to investigate the relations of positive vs. negative SEC to different variables of emotional experiences, interpersonal functioning, and mental health.
Emotional contagion as an interpersonal process
The idea of emotional contagion has been around since the 18th century [12,13], and it was introduced to scientific fields related to psychology at least 100 years ago [14–17]. Since then, various terms have been used to describe this process, including contagion, transmission, transfer, crossover, resonance, or mirroring [3,7,18–23], and different underlying mechanisms have been proposed [4]. While previous studies have often been unclear and inconsistent in their conceptualizations and definitions [24], emotional contagion can be defined—in agreement with several theorists—as a basic interpersonal process that comprises the mostly automatic and unintentional transmission of emotions from an individual to one or more other individuals resulting in a more or less congruent emotional experience [2,5,7,8].
In that sense, it can be delineated clearly from the much broader concept of empathy, which itself has recently been discussed in recent reviews as a rather vague and manyfold umbrella term comprising a range of more precisely defined lower-level concepts (see [24–26], for critical discussions of empathy definitions). In line with these reviews, empathy definitions typically encompass understanding others’ emotions and taking their perspective (i.e., cognitive empathy), sharing someone else’s emotions and feeling prosocial concern, compassion, or sympathy for others (i.e., affective empathy), and sometimes even the behavioral tendencies and reactions resulting thereof as well as accurately recognizing others’ emotional states [24–26]. However, this rather diverse collection of processes and capacities has been criticized and discussed as a threat to clear and consistent conceptualizations that are needed for scientific advancement and, instead, it has been suggested to bypass the term empathy in favor of more precise definitions of lower-level constructs, such as emotional contagion (e.g., [24,25].
Thus, in contrast to empathy, emotional contagion represents a purely affective response to other individuals’ emotional experiences (instead of cognitive role-/perspective-taking), it is based on unintentional, automatic, and mostly unconscious processes (instead of deliberately imagining being in another person’s situation), and it does not necessarily involve any behavioral reactions towards others (instead of comforting or helping another person). Most critically, though, emotional contagion does not require to distinguish between one’s own emotions and the other person’s emotions (instead of an at least minimum level of self-other distinction or awareness of another individual’s situation that is postulated as a defining feature of empathy) [24–30]. In other words, emotional contagion means that another individual’s emotions become my own emotions, while empathy means understanding and responding to another individual’s emotional experiences while distinguishing between my response and the other person’s emotions or situation.
In previous research, emotional contagion has been investigated within the context of social interaction in different applied fields, such as educational research [20,21,31,32] and clinical psychology [33–41]. In these studies, empirical evidence has been found for the contagious transmission of both positive [20,21,42] and negative emotions [18,32].
Susceptibility to emotional contagion as a personality trait
Above and beyond being a general interpersonal phenomenon, emotional contagion has also been considered from a personality difference perspective. Based on the observation that some individuals seem to catch others’ emotions quite easily and other individuals seem to be relatively immune against such contagion, it has been speculated that people differ in how prone or susceptible they are to catching others’ emotions. Hence, SEC has been proposed as a trait-like disposition. As such, an individual’s SEC is thought to reflect how susceptible an individual is to catching others’ emotions. More specifically, it is defined as an individual’s tendency or proneness to automatically and unintentionally catch other individuals’ emotions through the process of emotional contagion [5–8,43]. In that sense, SEC can be delineated from the broader concept of empathy, which commonly encompasses understanding others’ emotions and taking their perspective (i.e., cognitive empathy), sharing someone else’s emotions while maintaining self-other distinction (i.e., affective empathy), and even feeling compassion and sympathy for someone else (for critical reviews and discussions of empathy definitions, see [24–26]).
Methodologically, previous research relied on self-report measures to assess an individual’s SEC. Following an extensive and systematic literature review (see supporting information S1 Appendix), we identified seven published scales that explicitly addressed SEC (e.g., the Emotional Contagion Scale [44]and the Emotional Empathic Tendency Scale [7]) and 21 scales focusing on different forms of empathy that contained items resembling emotional contagion. Of the seven scales addressing SEC explicitly, four targeted adults ([7,44–46]), one targeted adults and adolescents ([47]), one specifically targeted parents [48], and one targeted children [49], respectively. All of the seven instruments have been constructed based on the assumption that the personal tendency to catch other individuals’ emotions through the process of emotional contagion is a global and unidimensional construct irrespective of the valence of the transmitted emotion.
Within all scales containing items on an individual’s SEC, we identified 132 items that focused directly on being susceptible to emotional contagion in their wording. Conspicuously, most of these items focus on negative emotions (54%) as compared to items focusing on positive emotions (26%) or items with no explicit emotional valence (21%). Even more so, within the seven scales addressing SEC explicitly, we identified 50 items focusing directly on SEC, of which 60% addressed negative emotions, 24% addressed positive emotions, and 16% did not make the emotional valence explicit. Taken together, existing scales—and especially the ones explicitly addressing SEC which represent the basis of most previous literature on SEC—seem to be biased towards an individual’s tendency to catch negative emotions and there seems to be a lack of more balanced measures to assess both the SEC of positive and negative emotions.
On a more substantial level, previous studies on individuals’ SEC have largely focused on different socially interactive contexts, for example, among health care practitioners [10,50,51], social workers [11,52], or salespersons [53]. In these studies, self-reported SEC was linked to an increased experience of negative emotions and to various indicators of mental health problems (e.g., emotional exhaustion, depressive symptoms, or professional impairment). However, given that typically in prior research, composite global scores for SEC were used, it is unclear whether it is the susceptibility to negative or positive emotions, or both, which underlies the associations with these indicators of health problems and other aversive experiences. Due to the overemphasis on negative emotions in the items of the existing scales, we postulate that previous research on individuals’ SEC is biased towards negative emotions. Research explicitly teasing apart the SEC of positive versus negative emotions is called for. To this end, a self-report measure to separately assess the susceptibility to catching positive and negative emotions in a balanced way is needed. While this separation may seem to be a solely methodological desideratum, it is also highly conceptually reasonable, as we set out below.
Theoretical framework and empirical support for positive versus negative SEC
Originally, SEC has been conceptualized as a unidimensional and global construct. In the present work, we argue for a more complex perspective on an individual’s SEC. Theoretically, our reasoning is rooted in Watson and Tellegen’s [54] consensual model of positive and negative affect. In contrast to emotion theorists who argued for a specific set of discrete emotions [55–58], Watson and Tellegen proposed a more parsimonious model based on the reanalysis of several datasets with positive and negative affect as two general and relatively independent dimensions of emotional experiences that can be subdivided into specific discrete emotions reflecting the valence of the dimensions, such as happiness or anxiety [54,59–61].
Since Watson and Tellegen’s proposal in 1985 [54], a range of empirical studies have provided evidence for this assumption. For example, in a study investigating college students’ affect over several class sessions and how it is altered by the feedback on their performance, positive and negative affect were found to be unrelated in each measurement. Moreover, while success feedback increased positive affect, it did not impact students’ negative affect, and vice versa [62]. In addition, several studies investigated certain context factors influencing the relationship between positive and negative affect. For instance, in a study examining the influence of culture and gender on the relationship between positive and negative affect, different correlational patterns were found for the subgroups of American men (weak negative correlation between positive and negative emotions) and women (strong negative correlation) versus Chinese men (weak positive correlation) and women (strong positive correlation), respectively [63]. Additionally, previous studies examining the influence of experienced stress on this association found that greater levels of stress lead to a stronger negative correlation between positive and negative affect [64,65]. In studies investigating the relation between positive and negative affect for both state and trait measures, the two dimensions were found to be unrelated at the trait level but showed a significant small negative correlation when considering a shorter time interval [66,67]. Taken together, the existing evidence suggests that people who frequently and intensively experience positive emotions do not systematically experience elevated or decreased levels of negative emotions. As such, the proposal of positive and negative affect as two separable and -more or less- independent dimensions provides a plausible and promising theoretical framework for our reasoning.
In line with this framework, we propose two distinct dimensions of SEC: The SEC of positive emotions (Positive SEC) and the SEC of negative emotions (Negative SEC). These dimensions reflect the valence of the emotions transmitted through the process of emotional contagion. In line with Watson and Tellegen’s model, we do not necessarily expect an antagonistic relation between Positive and Negative SEC (i.e., a higher Positive SEC does not necessarily lead to a lower Negative SEC). While we do not rule out any covariation between the two dimensions, we do not propose that there is a strong higher-order personal tendency to catch emotions of either valence from other people (resulting in a strong positive correlation between Positive and Negative SEC and explaining most of the variance). Instead, we expect Positive SEC and Negative SEC to be largely independent from one another and linked differentially to variables in several important domains of human functioning, namely emotional experiences, social interaction, and individuals’ mental and physical health.
Empirical support for a distinction of positive versus negative SEC comes from more recent studies in empathy research. For example, Light and colleagues [68] investigated an explicitly positive form of empathy using a newly developed self-report measure. They found that Empathic Happiness, an emotion-specific form of positive empathy (see also [69]), was negatively related to the experience of anhedonia and depressive symptoms [68]. Further, Lundkvist [70] examined the factor structure of a Swedish adaptation of Doherty’s [44] proposed initially unidimensional Emotional Contagion Scale (ECS) and reported that the items were grouped into two factors—SEC of positive and negative emotions. Similarly, Murphy and colleagues [22] investigated the validity of the Empathy Index [71], a self-report measure comprising the subscales Behavioral Contagion and Empathy, of which Jordan and colleagues [71] had explicitly claimed to represent a general form of being susceptible to the transmission of emotions. In their study, Murphy and colleagues [22] conducted factor analyses that revealed three different distinguishable factors, which they labeled Positive/Neutral Contagion (3 items, but only 2 refer to positive emotions; example item: “If I see a video of a baby smiling, I find myself smiling“), Distress Contagion (4 items; example item: “If I see someone fidgeting, I’ll start feeling anxious too”), and Physical Mimicry, representing a purely motoric reaction to other individuals’ behaviors (3 items; example item: “I catch myself crossing my arms or legs just like the person I’m talking to”). Subsequent correlational analyses revealed that Positive/Neutral Contagion was positively associated with sympathetic caring for others, extraversion, agreeableness, and openness, but negatively associated with attachment avoidance, emotional detachment, and various interpersonal aspects of personality disorders (e.g., meanness). Distress Contagion, on the other hand, was not associated with sympathetic caring, extraversion, agreeableness, and openness, but positively associated with anxiety, depressive symptoms, emotional distress, negative affect, attachment anxiety, and features of personality disorders (e.g., disinhibition). In a clinical context, Trautmann and colleagues [72] examined if an individual‘s tendency to vicariously experience others’ positive versus negative emotions influences an individual’s stress response when being exposed to a neutral film versus a film depicting a traumatic event. In this study, SEC was assessed using Lundkvist’s [70] two-factor conceptualization of Doherty’s ECS [44]. Trautmann and colleagues [72] found that being susceptible to negative emotions was positively related to the perceived stressfulness of the stimulus as well as increased anxiety and heart rate, whereas being susceptible to positive emotions was not related to these variables. However, this study focused solely on emotional responses to an extremely aversive and negatively valenced stimulus and it did not include any positively valenced stimuli or outcome variables. Nevertheless, their reported correlational pattern supports the distinction of positive versus negative SEC.
In conclusion, scattered empirical findings support our proposition that SEC should not be treated as a global and unidimensional construct. An individual’s Positive SEC vs. Negative SEC appear to differ in their associations to relevant socio-emotional outcome variables (i.e., emotional experiences, interpersonal functioning, and well-being). Negative SEC has previously been associated with an increased experience of negative emotions and mental health problems [10,11,22,68,72]. Positive SEC has not shown similar or opposite relations in previous research, but has been associated with greater interpersonal functioning instead [22]. Taken together, there is both theoretical and empirical support for a distinct and independent conceptualization of Positive and Negative SEC.
Study 1
Study 1 was part of a larger research project ([73]; see [74] for information on the overarching project) and designed to provide first evidence on the psychometric quality of our novel scales, their statistical separability, and their proposed differential relations with a relevant outcome variable, namely emotional exhaustion being one core component of burnout [75]. As elaborated above, we suggest that Negative SEC, but not Positive SEC, should be positively linked with this measure of psychological ill-being. In this study, we chose to sample from the population of secondary school teachers, given that teaching is not only highly emotionally demanding and characterized by severe burnout risks, but also an extremely social and interpersonal profession [76,77].
Materials and methods
Sample and procedure
The sample of Study 1 consisted of N = 257 teachers (70.4% female; no answer, 1.2%) from more than 40 schools in different German states (Bavaria, 81.3%; Baden-Wurttemberg, 11.3%; other, 6.6%; no answer, 0.8%). Their age ranged from 26 to 64 years (M = 41.8; SD = 10.9; no answer, 2.7%) and 96.9% had obtained the German "Abitur", a qualification level granted to students after usually 12 to 13 years after passing their final exams at the end of secondary education in Germany giving them access to academic higher education (no answer, 1.2%). They had an average teaching experience of 13.2 years (SD = 10.3; no answer, 6.6%) and they worked at different types of secondary schools (13.6% lowest, 19.1% middle, 52.9% highest academic track), in primary schools (2.3%), in vocational schools (3.5%), or in more than one of these school types at the same time (6.2%; no answer, 1.6%). They taught a wide range of subjects (e.g., Math, German, English) and their average teaching hours per week ranged from 6 to 28 hours (M = 20.7; SD = 5.2; no answer, 5.1%). All participants were recruited voluntarily in 2017 through convenience sampling and direct contact with the schools or the teachers (i.e., schools and teachers were contacted via email, telephone, and written informational letters). Overall, more than 480 paper-pencil questionnaires were sent out, with a response rate of 53.2% resulting in our final sample size and exceeding response rates of previous paper-pencil studies in this specific population [78,79]. No incentives were given for participation and no identifiers that could link individual participants to their results were obtained. Thus, all analyses were conducted on anonymous data. All participants gave their informed consent and the study was conducted in accordance with the ethical standards of the APA and the Declaration of Helsinki. The research project received a formal waiver of ethical approval by the ethics review board of the first author’s institution.
Measures.
Positive and Negative SEC—Scale Development and Item Selection: Within our two newly developed scales, we chose to formulate items with regard to specific discrete emotional experiences, such as being happy or stressed out. We aimed to create a balanced measure with items assessing the SEC of positive emotional experiences and of negative emotional experiences being equally represented. We were inspired by existing measures of SEC or affective empathy, including the ECS [44] and the Emotional Empathic Tendency Scale [7], and specific items within these measures served as the basis for newly created items. The selected items were all related to Positive SEC or Negative SEC in their content and wording and were either taken from published German translations or carefully transferred to the German language by means of translation-back-translation by two bilingual experts who reached a consent in a committee approach in case of alternative translations (see S1 Table). To eliminate existing deficits in published self-report measures of SEC, such as ambiguous items (e.g., “I am happy when I am with a cheerful group and sad when the others are gloom”; [46,80]) or competence-oriented items (e.g., “I am able to remain calm even though those around me worry”; [7]), some of the items have been reworded and new items have been created according to our two-fold conceptualization of SEC. In total, five items each were created to assess individuals’ Positive SEC and Negative SEC. All items were presented in mixed order within the questionnaire and answered on a five-point Likert Scale (ranging from 1 = “strongly disagree” to 5 = “strongly agree”).
Emotional Exhaustion: We used the German version of the emotional exhaustion subscale of the Maslach Burnout Inventory ([81,82]; 9 items; response scale 0 to 6; example item: “I feel emotionally drained from my work”; Cronbach’s alpha = .86; M = 1.83; SD = 0.96; Min/Max = 0.11/4.56), which has been used frequently as a standalone measure of burnout in previous research [75].
Statistical analyses.
First, we conducted exploratory factor analysis (EFA) to explore the factor structure of our newly developed scale, and potentially exclude items based on empirical and theoretical considerations; to result in a short and economic self-report measure. Second, we examined the psychometric properties of the scales and assessed their internal consistency. Given that McDonald’s omega has the advantage of considering the strength of association between items and constructs as well as the item-specific errors [83], we used both Cronbach’s alpha and McDonald’s omega coefficients. Third, we calculated bivariate Pearson correlations to explore the relations between Positive and Negative SEC and with emotional exhaustion separately. Data processing and analysis were done in R and all data and code to reproduce the results of Study 1 are available in an OSF-repository together with additional supporting information (see https://osf.io/jfxk4/).
Results and discussion
Preliminary analyses.
All items aiming to assess an individual’s SEC showed deviations from univariate normality as tested using the Shapiro-Wilk test, but no significant multivariate outliers were detected using the Mardia test [84]. While the items regarding Negative SEC appeared to be relatively normally distributed based on their skewness levels, those regarding Positive SEC appeared to be negatively skewed (see Table 1 for the items’ psychometric properties). As tested using Little’s MCAR test, missing data in all SEC items was completely at random and did not exceed 2% of all cases. These cases were listwise deleted resulting in N = 252 cases for EFA.
Exploratory factor analyses.
EFA was used to test our theoretical assumption of two separate dimensions underlying the idea of SEC. To investigate the adequacy of the collected data for factor analyses, we inspected the inter-item correlations and used the Kaiser-Meyer-Olkin (KMO) criterion and Bartlett’s test of sphericity. Inter-item correlations ranged between .06 and .65, the KMO measure of sampling adequacy was .84, and Bartlett’s test of sphericity was significant [χ2(45) = 892.41, p < .001], collectively indicating good factorability of the data for further analyses ([85]). To determine the number of factors to retain, we first inspected the factors’ eigenvalues revealing two factors with eigenvalues greater than one. Second, when looking at the Scree Plot, we identified a break after two factors which were confirmed by parallel analysis. As this was in line with these results and our theoretical reasoning, we decided to retain two factors. As we would not exclude the existence of at least small-sized correlations between the two factors, we used an oblique factor rotation of the factors (“oblimin” method). Because the data violated the normality assumption, we used an ordinary least squares (OLS) method to find the minimum residual solution (“minres” method).
After conducting EFA, five items had significant factor loadings on Factor 1 and five items on Factor 2, respectively (ranging between .49 and .85). To create a parsimonious and economic yet still balanced measure of SEC, we excluded the two items with the weakest factor loadings and/or cross-loadings > .1 (see S2 Table for the factor loadings of all 10 items). Based on item content, Factor 1 was labeled “Positive SEC” and Factor 2 was labeled “Negative SEC” (see Table 2 for the factor loadings of the final 8 items in Study 1 and Study 2). The two factors showed an interfactor correlation of r = .37 and cumulatively accounted for 48% of the total variance (Factor 1 = 26%, Factor 2 = 22%). In sum, EFA revealed two separable yet positively related factors, and the items representing these two factors corresponded to our proposed theoretical conceptualization of Positive SEC and Negative SEC.
Descriptive statistics.
Across all variables (incl. emotional exhaustion), missing values were found in 5.8% of the cases which were listwise deleted for subsequent descriptive and correlational analyses, which is thus based on N = 242 cases. Participants reported relatively high levels of item endorsement for Positive SEC (M = 4.16; SD = .59; Min/Max = 2/5) as compared to Negative SEC (M = 2.76; SD = .75; Min/Max = 1/4.5). However, standard deviations were sufficiently large to preclude ceiling or floor effects. Internal consistencies of the two scales were examined by calculating Cronbach’s alpha and McDonald’s omega coefficients showing acceptable to good reliabilities for Positive SEC (α = .83; CI 95% [.80, .87] and ω = .83; CI 95% [.80, .87]) as well as Negative SEC (α = .76; CI 95% [.71, .81] and ω = .77; CI 95% [.72, .81]), according to conventions [85].
Correlational analyses.
Given the good internal consistencies of both SEC scales, we calculated mean scores for Positive SEC and Negative SEC. Bivariate Pearson Correlation revealed a significant positive relation between Positive SEC and Negative SEC (r = .27; p < .001; CI 95% [.15, .38]). Further, Negative SEC was significantly positively related to self-reported emotional exhaustion (r = .27; p < .001; CI 95% [.15, .38]), whereas Positive SEC and emotional exhaustion were uncorrelated (r = -.02; p = .765; CI 95% [-.14, .11]).
Taken together, Study 1 provided initial evidence for our theoretical assumption regarding two separable dimensions of SEC (Positive SEC vs. Negative SEC). Our findings suggest that our new balanced SEC measure represents a reliable instrument in terms of the scales’ internal consistencies. Additionally, Positive and Negative SEC proved to differ in their relation to a theoretically meaningful external criterion. Negative SEC, but not Positive SEC, was linked with emotional exhaustion as a prototypical measure of psychological ill-being.
Study 2
In Study 2, we had two primary goals: First, we aimed to corroborate and expand the evidence on our novel scales’ validity and reliability. We expected to replicate the proposed two-factor structure in an independent and more generalized sample using confirmatory factor analysis (CFA), clearly identifying Positive SEC and Negative SEC as two distinct, yet related, factors, with the two-factor model’s fit being superior to the single-factor model (internal validity). Second, regarding the scales’ external validity, we aimed to investigate the associations of self-reported Positive SEC vs. Negative SEC with measures of emotional experiences, interpersonal functioning, health/well-being (criterion-oriented validity), and measures of cognitive and affective empathy components (divergent and convergent validity). Study 2 was preregistered prior to conducting any analyses (see https://aspredicted.org/nk8zx.pdf) and all data and code to reproduce the results of Study 2 are available in an OSF-repository (see https://osf.io/jfxk4/) together with additional supplemental material.
Materials and methods
Sample and procedure
The sample of Study 2 consisted of N = 247 participants (48.6% female) that were recruited in 2018 via Clickworker (www.clickworker.com), a German online recruiting service similar to Amazon MTurk (www.mturk.com). Their age ranged from 18 to 75 years (M = 39.9 years; SD = 12.8 years), 55.9% had obtained the German "Abitur", and 38.5% indicated to be married. All data was collected anonymously using SoSciSurvey, a German open-access provider for online-surveys (www.soscisurvey.de). Due to data privacy reasons, no information on the participants’ place of residence or professional occupation was assessed. The participants received a small monetary incentive (3 €) for completing the survey. On average, they required M = 21.2 minutes to complete the questionnaire (SD = 7.82, Min = 6.2, Max = 59.7). While all participants who started to fill out the online questionnaire also completed it (completion rate = 100%), three participants were excluded prior to any further data processing, as preregistered, because they filled out the questionnaire unrealistically fast (i.e., in under five minutes). The acquired sample size (N = 250) was determined based on statistical power estimations with statistical power for correlational analyses of 1 – β > .99 for medium and large effects (alpha level = 0.01). The study was conducted in accordance with the ethical standards of the APA and the Declaration of Helsinki and the research project was approved by the ethics review board of the first author’s institution.
Measures.
In addition to our two novel SEC scales, all participants completed the following self-report measures of (1) cognitive/affective empathy, (2) positive/negative emotionality and distress, (3) well-being and mental/physical health, and (4) social/interpersonal functioning (see Table 3 for sample items and detailed characteristics for each measure including example items and response scales).
Measures of Empathy: To assess different components of empathy, we used scales from the Interpersonal Reactivity Index (IRI; [86,95]) and the Adolescent Measure of Empathy and Sympathy (AMES; [87]), which has been validated in adult samples as well [96]. The IRI Perspective-Taking scale (Cronbach’s alpha α = .76; McDonald’s omega total ω = .81) measures an individual’s tendency to adopt another’s perspective to find out what another person might be thinking, thus, representing the cognitive empathy component. The IRI Fantasy scale (α = .75; ω = .81) measures an individual’s tendency to imaginatively transpose oneself into the feelings and actions of fantasy characters. The AMES Affective Empathy scale (α = .81; ω = .84) aims to measure the tendency to vicariously experience another person’s emotions which is quite similar to the concept of emotional contagion. The AMES Cognitive Empathy scale (α = .82; ω = .86) aims to measure an individual’s capacity to cognitively understand another person’s emotions.
Measures of Positive/Negative Emotionality and Distress: To assess different aspects of individuals’ emotionality and distress, we used scales of the IRI and the Big-Five-Inventory 2 (BFI2; [93,97]) as well as the Positive and Negative Affect Schedule (PANAS; [59,98]). The IRI Personal Distress scale (α = .76; ω = .80) measures an individual’s self-oriented tendency to experience distress and anxiety in highly emotional interpersonal situations, such as emergencies. Within the BFI2, the Big Five personality traits can be further subdivided in more differentiated facets. From the Negative Emotionality trait (formerly Neuroticism), we used the BFI2 Emotional Volatility (α = .75; ω = .80), BFI2 Depressiveness (α = .83; ω = .87), and BFI2 Anxiety (α = .69; ω = .77) facets and the BFI2 Activity/Energy Level (α = .73; ω = .79) facet from the Extraversion trait, respectively. The PANAS scales (Positive Affect and Negative Affect) assess an individual’s general experience of positive (α = .88; ω = .91) and negative affect (α = .89; ω = .92) and consist of adjectives describing discrete emotional experiences that are either positive or negative in their valence (instruction: “How do you feel in general?”).
Measures of Well-being and Mental/Physical Health: We used the Edinburgh Depression Scale (EDS; [88,99,100]) to assess current depressive symptoms (α = .87; ω = .90), the General Anxiety Disorder 7 questionnaire (GAD-7; [89,101]) to assess current symptoms of general anxiety (α = .88; ω = .91), and the Perceived Stress Scale (PSS-10; [90,102]) to assess individuals’ current levels of stress (α = .87; ω = .90). To assess individuals’ current physical symptoms (α = .80; ω = .86), we used a short version of the Cohen-Hoberman-Inventory of Physical Symptoms (CHIPS; [91,103]) and to assess individuals’ satisfaction with their lives and current living conditions (α = .93; ω = .94), we used the Satisfaction with Life Scale (SWLS; [92]).
Measures of Social/Interpersonal Functioning: To assess different aspects of individuals’ social and interpersonal functioning, we used scales of the IRI, the AMES, and specific facets of BFI2 personality traits. The AMES Sympathy subscale (α = .68; ω = .77) aims to measure an individual’s tendency to feel concerned or sorrow for another person. The IRI Empathic Concern subscale (α = .69; ω = .71) aims to measure an individual’s tendency to feel concern and sympathy towards others. From the personality trait extraversion, we used the BFI2 Sociability (α = .88; ω = .82) and BFI2 Assertiveness (α = .78; ω = .80) facets and the BFI2 Compassion (α = .64; ω = .71), BFI2 Trust (α = .61; ω = .73), and BFI2 Respectfulness (α = .73; ω = .80) facets from the personality trait Agreeableness, respectively.
Social Desirability: Given that self-reported health-related and clinical psychological variables have been reported to be associated with social desirability [104], we decided to assess the participants’ social desirability response tendencies. To this end, we used a short German measure (KSE-G; [94]) of the tendency to either exaggerate one’s positive qualities (α = .62; ω = .64) or to conceal negative qualities (α = .70; ω = .72)
Statistical analyses.
Regarding the scales’ internal validity, we conducted CFA using the package “lavaan” ([105], version 0.6.4) in R ([106], version 3.6.0). To evaluate model fit, we inspected a range of fit indices, including the comparative fit index (CFI), the Tucker Lewis Index (TLI; also called the non-normed fit index), the root-mean-square error of approximation (RMSEA), and the standardized root-mean-square residual (SRMR). In line with Hu and Bentler [107], model fit was recognized as acceptable with a CFI and TLI of .95 or higher, an RMSEA of .06 or lower, and an SRMR of .08 or lower [107,108]. Because the data violated the normality assumption, we used robust estimators of model fit (MLR) with robust (Huber-White) standard errors and a scaled test statistic that is (asymptotically) equal to the Yuan-Bentler test statistic [109]. We conducted model comparisons using the “anova” function (χ2 Difference Test with “satorra.bentler.2001” method) in the R package “lavaan”.
Given that several variables were substantially related to self-reported social desirability, we calculated partial correlations of the two scales (Positive and Negative SEC) with these measures while controlling for social desirability (see S3 Table for bivariate correlations of all measures with social desirability and S4 Table for bivariate zero-order correlations of Positive SEC and Negative SEC with all measures without controlling for social desirability). While the distribution of almost all variables were judged as non-normal based on Shapiro-Wilk tests, we still consider the distribution of our variables sufficiently normal based on visual inspection of the collected data (all data visualizations are included in the analysis scripts in the supplement) and the respective skewness levels [85], especially in light of the generally assumed robustness of Pearson correlations against violations of the normality assumption (e.g., [110]; see S5 Table for supplemental non-parametric Spearman correlations of Positive and Negative SEC with all measures.). To account for multiple testing and the risk of alpha cumulation, we adjusted alpha significance levels using the Bonferroni correction for 23 correlational tests resulting in a new alpha significance level of p < .002 and calculated confidence intervals for all correlation coefficients [111]. As in Study 1, all data processing and statistical analyses were done in R, and reproducible scripts for all reported results were generated.
Results
Descriptive statistics and reliability.
Table 4 shows descriptive statistics for all Positive and Negative SEC items. Again, relatively strong item endorsement was found for Positive SEC (means > 3 on the five-point scale for all items) as compared to Negative SEC (means < 3 on the five-point scale for all items), yet with standard deviations again being sufficiently large to preclude ceiling or floor effects in both scales. Item total correlations (part-whole-corrected) ranged between r = .59 and .73 for Positive SEC and .57 and .75 for Negative SEC, indicating good item discrimination capacities and item homogeneity in both subscales [86]. All items of the SEC scales deviated from univariate normality as tested using the Shapiro-Wilk test. Still, no significant multivariate outliers were detected using the Mardia test [85]. Based on their skewness levels, the items regarding Negative SEC seem to be relatively normally distributed, whereas the items regarding Positive SEC appear to be negatively skewed. No missing values were found in the responses to the items of the SEC scale.
Table 5 shows descriptive statistics of all scales reported in this study, including Positive and Negative SEC. Internal consistencies of our SEC scales were examined by calculating Cronbach’s alpha and McDonald’s omega indices showing good reliability for Positive SEC (α = .83; CI 95% [.79, .86] and ω = .83; CI 95% [.80, .86]) as well as Negative SEC (α = .82; CI 95% [.79, .86] and ω = .83; CI 95% [.79, .86]), according to conventions ([85]). The two factors Positive SEC and Negative SEC showed a small positive zero-order correlation between each other (r = .17; p < .001; CI 95% [.04, .29]) and a small to medium-sized positive correlation (r = .27; p < .001; CI 95% [.15, .39]), when controlling for social desirability, respectively; indicating that individuals reporting higher levels of Positive SEC also tended to report higher levels of Negative SEC.
Internal validity.
The model fit for the two-factor model (Positive SEC and Negative SEC) was very good (CFI = .982; TLI = .973; SRMR = .045; RMSEA = .054 [.006, .088]) while the model fit of the single-factor model was unacceptable (CFI = .520; TLI = .327; SRMR = .189; RMSEA = .267 [.242, .293]). The superiority of the two-factor model over the single-factor model was further supported by a lower Akaike Information Criterion (AIC) value for the two-factor model (AIC = 4376.1) as compared to the single-factor model (AIC = 4712.9) as well as a highly significant model comparison (Δχ2(1) = 1238.8, p < .001). As expected, all items showed high factor loadings with their corresponding factor, with positive standardized coefficients ranging from λ = .63 to λ = .88 (see Table 2).
External validity.
To explore the external validity of both SEC scales, we obtained correlations of Positive SEC and Negative SEC with all criterion variables included in the study, controlling for social desirability (see Table 6).
Correlations with Measures of Empathy: While Positive SEC only showed small to medium-sized correlations with the assessed measures of empathy, Negative SEC showed a strong positive correlation with the scale AMES Affective Empathy (r = .69) and a small positive correlation with the IRI Fantasy scale (r = .21), but no correlations were found with the scales IRI Perspective Taking and AMES Cognitive Empathy.
Correlations with Measures of Positive/Negative Emotionality and Distress: Positive SEC was positively correlated with the scale BFI2 Activity/Energy Level (r = .36), but uncorrelated with the other measures. Negative SEC showed a strong positive correlation with the scales IRI Personal Distress (r = .58) and BFI2 Anxiety (r = .52) and small to medium-sized correlations with the scales BFI2 Emotional Volatility (r = .44), BFI2 Depressiveness (r = .39), and PANAS Negative Affect (r = .27); but no significant relations with the scales PANAS Positive Affect (r = -.19) and BFI2 Activity/Energy Level (r = -.12).
Correlations with Measures of Well-Being and Mental/Physical Health: Negative SEC showed significant medium-sized positive correlations with the scales EDS Depressiveness (r = .29), GAD7 Anxiety (r = .43), PSS Stress (r = .36), and CHIPS Physical Symptoms (r = .36) as well as a small negative correlation with the scale SWLS Life Satisfaction (r = -.26). In contrast, Positive SEC was uncorrelated with all of these measures.
Correlations with Measures of Social/Interpersonal Functioning: Positive SEC showed medium-sized positive correlations with the scales AMES Sympathy (r = .35), IRI Empathic Concern (r = .42), BFI2 Sociability (r = .34), and BFI2 Compassion (r = .32). Negative SEC, in turn, was only weakly or uncorrelated with these measures, with only a few exceptions of small- to medium-sized correlations (i.e., BFI2 Assertiveness, IRI Empathic Concern, and AMES Sympathy).
Discussion
First, regarding correlations with self-report measures of empathy (in terms of convergent/divergent validity), we did not find any substantial correlations between Negative SEC and most empathy scales. The only exception was a strong positive correlation with the AMES Affective Empathy scale. However, this positive association supports the convergent validity of our Negative SEC scale, given that the AMES Affective Empathy scale shows substantial overlap with our definition of SEC as well as with the wording of our Negative SEC items. In comparison, Positive SEC showed only small to medium correlations with both cognitive and affective empathy components (not all of them significant). Overall, these findings suggest reasonable divergent validity in terms of both SEC scales’ relations to self-reported empathy.
Second, regarding correlations with self-report measures of positive/negative emotionality and distress, our findings suggest asymmetric correlational associations with Positive SEC vs. Negative SEC. This asymmetric pattern implies that Negative SEC is related to an increased experience of emotional instability, distress, and negative emotions, including depressiveness and anxiety, while Positive SEC does not seem to be related to these measures. There were only few exceptions to this pattern: An individual’s level of activity/energy as measured by the BFI2 scale was positively related to Positive SEC and, rather unexpectedly, both SEC scales were uncorrelated with positive affect as measured using the PANAS Positive Affect scale.
Third, this asymmetrical pattern was also evident regarding self-report measures of well-being and mental/physical health. Negative SEC was positively related to current experiences of depressiveness, anxiety, stress, and physical health and negatively related to their subjective well-being. Positive SEC, on the other hand, wasn´t correlated with any these variables.
Fourth, regarding measures of social/interpersonal functioning, our findings showed the opposite asymmetrical pattern. While Negative SEC did not seem to be substantially related to most of these variables, Positive SEC was related to greater interpersonal functioning and prosocial behavioral tendencies as measured by the scales AMES Sympathy, IRI Empathic Concern, BFI2 Sociability, and BFI2 Compassion. Again, there were a few exceptions from this asymmetric pattern: Assertiveness was negatively related to Negative SEC and uncorrelated with Positive SEC and Respectfulness was uncorrelated with both SEC scales. However, being assertive as measured by the BFI2 Assertiveness items reflects assuming control and influencing others and being respectful as measured by the BFI2 Respectfulness items reflects being polite and obliging. In that respect, these two scales differ from the other scales and possibly represent less prosocial facets of interpersonal interaction.
General discussion
In the present work, we first aimed to investigate our novel scales’ internal validity and reliability in terms of their psychometric properties, internal consistencies, and factor structure. Second, regarding the scales’ external validity, we aimed to explore diverging patterns in the relations of Positive and Negative SEC with relevant criterion variables. To this end, we conducted two studies to examine our scales’ internal validity using EFA (Study 1) and CFA (Study 2) and their external validity using correlations with measures of empathy, emotional experiences, mental health problems, and indicators of social functioning.
Internal validity and reliability
Regarding the measure’s internal validity, EFA in Study 1 revealed a two-factor structure with Positive SEC and Negative SEC as clearly distinct yet correlated factors. The factor structure was confirmed in Study 2 using CFA with the two-factor model showing a good model fit; clearly superior to a single-factor model as demonstrated by a chi-square difference test. In both studies, the two scales of our SEC Scale demonstrated good reliability in terms of Cronbach’s alphas and McDonald’s omegas, ranging between .76 and .83. Hence, the two newly developed SEC measures appear to be an internally valid and reliable self-report measures.
External validity
Regarding the scales’ external validity, we explored their relations to (1) self-report measures of empathy, (2) of positive/negative emotionality and distress, (3) of well-being and mental/physical health, and (4) of social/interpersonal functioning. Overall, while both SEC scales seemed to show reasonable convergent and divergent validity based on their relations with the assessed empathy measures, we found diverging correlational patterns of Positive SEC vs. Negative SEC with the other criterion variables. On the one hand, Negative SEC was related to increased trait-like negative emotional experiences (BFI2 Depressiveness), emotionality (BFI2 Emotional Volatility), distress (IRI Personal Distress), and greater current mental health problems (EDS Depressiveness, GAD7 Anxiety, PSS Stress, SWLS Life Satisfaction) and physical symptoms (CHIPS Physical Symptoms). In contrast, Positive SEC appeared to be unrelated to these measures. On the other hand, Positive SEC was related to increased indicators of interpersonal functioning and prosocial tendencies, and other-oriented behaviors and cognitions (AMES Sympathy, IRI Empathic Concern, BFI2 Sociability, BFI2 Compassion). In turn, Negative SEC appeared to be unrelated to these variables.
Taken together, while previous research had provided initial scattered evidence of a certain degree of asymmetry in the pattern of relations of Positive SEC and Negative SEC with external variables–which is why we proposed to differentiate more clearly between the two dimensions in the first place–the pattern of relationships we found in our studies corroborated this divergence using two separate SEC scales. We conclude that future research will profit from differentiating these two dimensions of SEC, and explicitly considering the valence of emotions being transmitted form one person to another in the context of research on emotional contagion as a process and SEC as an individual’s disposition.
Limitations
While this work bears important new insights in the context of research on emotional contagion and SEC, it also bears some limitations which need to be taken into consideration when interpreting its findings. First, in both studies, we used a cross-sectional research design to assess the relations of Positive and Negative SEC to selected criterion variables. Therefore, it is important to be aware that our analyses cannot determine any temporal or causal relationship between the assessed variables. Second, we solely used self-report measures to assess SEC and all other variables. Thus, the reported results could be biased due to common method variance [112]. Third, while research on the process of emotional contagion in human social interaction has seemingly been increasing over the past years, SEC as an individual’s disposition has been investigated less frequently in past research. As a result, some of the cited publications in the present work date back more than 10 years. Thus, more current scientific inquiries into SEC and its associations with relevant outcomes are needed, such as the present studies. Last, while both samples were sufficiently large in terms of statistical power for detecting medium and large effects as well as stability of correlations [113], the generalizability of the findings could still be limited due to our chosen sampling approaches and the fact that recruitment and data collection happened before the COVID-19 pandemic. In Study 1, we used a convenience sample of voluntarily recruited teachers. Therefore, it is possible that only highly motivated and healthy individuals took part in the study. In Study 2, we used an online sample recruited through a crowd-based online platform, also voluntarily but with a monetary incentive. While such crowd-based online samples have been found to be reasonably representative of the general population and to provide high-quality data [114], participants in these samples have been found to report increased levels of depressiveness with up to three times higher levels as compared to prevalence estimations in the general population [115]. Moreover, all data had been collected prior to the onset of the COVID-19 pandemic, which had been shown to have impacted individuals’ well-being and mental health in the general population worldwide ([116]) and especially in highly vulnerable subgroups (e.g., [117,118]). Finally, both samples were relatively homogeneous regarding cultural background and educational levels, and the language of all measures in our studies was German. Taken together, more diverse samples, recruited especially during or after the pandemic, are needed to replicate and corroborate our findings, and an English version of our novel self-report measure should be developed and validated in English-speaking populations.
Directions for future research
First, to sustainably overcome the confusion regarding definitions and operationalizations of emotional contagion and SEC in future research and to foster a more consistent use of terms and concepts, we suggest—in line with previous theorists (e.g., [24)—that researchers explicitly define and transparently operationalize their view on emotional contagion and/or SEC in future studies.
Second, in addition to replicating our findings, more studies are needed to investigate the temporal stability and predictive validity of individuals’ Positive SEC and Negative SEC. More specifically, their relations to other important variables, such as emotion regulation capacities, attachment security/insecurity, mentalization capacities, relationship quality, or prosocial behavior, should be investigated further.
Third, to go beyond self-report data in future studies, we suggest including other measures of emotional experiences, personality traits, and socially interactive behavior to cross-validate our new self-report measure, such as experience sampling [119,120], smartphone-based mobile sensing [121,122], or systematic behavioral coding. In that direction, the Facial Action Coding System (FACS; [123]) has proven to be a particularly suitable tool to quantify human facial expressions which represent a highly relevant channel to nonverbally communicate emotions to other individuals in socially interactive situations. Subsequently, the interpersonal transmission of facially expressed emotions could be investigated using nonlinear time series analyses, such as Cross-Recurrence Quantification [124]. Such fine-grained and dynamic methods are promising approaches to further validate our balanced Positive and Negative SEC scales as a measure of an individual’s disposition to catch either positive and/or negative emotions when interacting with others.
Supporting information
S1 Appendix. Systematic review of published self-report measures of susceptibility to emotional contagion.
This review strives to (1) give a systematic overview of existing self-report instruments to measure individuals’ susceptibility to emotional contagion (SEC), (2) to describe the theoretical framework of existing scales, and (3) to review the specific items that are used to assess individuals’ SEC in different scales.
https://doi.org/10.1371/journal.pone.0302890.s001
(DOCX)
S1 Table. German items and related items in existing scales.
https://doi.org/10.1371/journal.pone.0302890.s002
(DOCX)
S2 Table. Factor matrix of all 10 items initially tested in Study 1.
https://doi.org/10.1371/journal.pone.0302890.s003
(DOCX)
S3 Table. Bivariate correlations of all measures with social desirability.
https://doi.org/10.1371/journal.pone.0302890.s004
(DOCX)
S4 Table. Zero-order correlations of Positive and Negative SEC with all measures.
https://doi.org/10.1371/journal.pone.0302890.s005
(DOCX)
S5 Table. Non-parametric Spearman correlations of Positive and Negative SEC with all measures.
https://doi.org/10.1371/journal.pone.0302890.s006
(DOCX)
References
- 1. Barreto ML, Teixeira MG, Carmo EH. Infectious diseases epidemiology. Journal of Epidemiology and Community Health. 2006; 60(3): pp. 192–195. pmid:16476746
- 2. Bischof-Köhler D. Über den Zusammenhang von Empathie und der Fähigkeit, sich im Spiegel zu erkennen. Schweizersche Zeitschrift für Psychologie. 1988; 47(2–3): pp. 147–159.
- 3. Dezecache G, Jacob P, Grèzes J. Emotional contagion: Its scope and limits. Trends in Cognitive Sciences. 2015; 19(6): pp. 297–299. pmid:25891260
- 4. Elfenbein HA. The many faces of emotional contagion: An affective process theory of affective linkage. Organizational Psychology Review. 2014; 4(4): pp. 326–362.
- 5. Hatfield E, Cacioppo JT, Rapson RL. Emotional contagion. Current Directions in Psychological Science. 1993; 2(3): pp. 96–100.
- 6. Hatfield E, Bensman L, Thornton PD, Rapson RL. New perspectives on emotional contagion: A review of classic and recent research on facial mimicry and contagion. Interpersona: An International Journal on Personal Relationships. 2014; 8(2): pp. 159–179.
- 7. Mehrabian A, Epstein N. A measure of emotional empathy. Journal of Personality. 1972; 40(4): pp. 525–543. pmid:4642390
- 8. Miller KI, Stiff JB, Ellis BH. Communication and empathy as precursors to burnout among human service workers. Communication Monographs. 1988; 55(3): pp. 250–265.
- 9. Schoenewolf G. Emotional contagion: Behavioral induction in individuals and groups. Modern Psychoanalysis. 1990; 15(1): pp. 49–61.
- 10. Omdahl BL, O’Donnell C. Emotional contagion, empathic concern and communicative responsiveness as variables affecting nurses’ stress and occupational commitment. Journal of Advanced Nursing. 1999; 29(6): pp. 1351–1359. pmid:10354229
- 11. Siebert DC, Siebert CF, Taylor-McLaughlin A. Susceptibility to emotional contagion: Its measurement and importance to social work. Journal of Social Service Research. 2007; 33(3): pp. 47–56.
- 12.
Hume D. A treatise of human nature, reprinted from the original edition in three volumes and edited by Selby-Bigge L.A., M.A. (Oxford, UK: Clarendon Press, 1896) [Facsimile PDF version]. Available from: https://oll.libertyfund.org/titles/342
- 13.
Smith A. The theory of moral sentiments and essays on philosophical subjects, reprinted from the original edition and edited by Black Joseph and Hutton James (London: UK: Alex, Murray & Son, 1869) [Facsimilie PDF version]. Available from: https://oll.libertyfund.org/titles/2721
- 14. Lipps T. Psychologische Untersuchungen—Band 1. W. Engelmann; 1905. Available from: https://archive.org/details/psychologischeu01lippgoog/mode/2up
- 15. Scheler M. Zur Phänomenologie und Theorie der Sympathiegefühle und von Liebe und Hass: Mie einem Anhang über den Grund zur Annahme der Existenz des fremden ich. M. Niemeyer; 1913.
- 16. Scheler M. Wesen und Formen der Sympathie: Der "Phänomenologie der Sympathiegefühle" (2. vermehrte und durchgesehene Auflage). F. Cohen; 1923.
- 17. Zahavi D. Debating empathy: Historical awareness and conceptual precision. Emotion Review. 2022; 14(3).
- 18. Bolger N, DeLongis A, Kessler RC, Wethington E. The contagion of stress across multiple roles. Journal of Marriage and the Family. 1989; 51(1): pp. 175–183.
- 19.
Eisenberg N, Eggum ND. Empathic responding: Sympathy and personal distress. In Decety J, Ickes W, editors. The social neurosciences of empathy. MIT Press; 2009. pp. 71–83. https://doi.org/10.7551/mitpress/9780262012973.003.0007
- 20. Frenzel AC, Goetz T, Lüdtke O, Pekrun R, Sutton RE. Emotional transmission in the classroom: Exploring the relationship between teacher and student enjoyment. Journal of Educational Psychology. 2009; 101(3): pp. 705–716.
- 21. Frenzel AC, Becker-Kurz B, Pekrun R, Goetz T, Lüdtke O. Emotion transmission in the classroom revisited: A reciprocal effects model of teacher and student enjoyment. Journal of Educational Psychology. 2018; 110(5): pp. 628–639.
- 22. Murphy BA, Costello TH, Lilienfeld SO. Is empathic contagion helpful or harmful? Overlooked heterogeneity in the Empathy Index. Psychological Assessment. 2018; 30(12): pp. 1703–1708. pmid:30035555
- 23. Parkinson B. Interpersonal emotion transfer: Contagion and social appraisal. Social and Personality Psychology Compass. 2011; 5(7): pp. 428–439.
- 24. Hall JA, Schwartz R. Empathy present and future. The Journal of Social Psychology. 2019; 159(3): pp. 225–243. pmid:29781776
- 25.
Batson CD. These things called empathy: Eight related but distinct phenomena. In: Decety J, Ickes W, editors. The social neuroscience of empathy. MIT Press; 2009. https://doi.org/10.7551/mitpress/9780262012973.003.0002
- 26. Cuff BMP, Brown SJ, Taylor L, Howat DJ. Empathy: A review of the concept. Emotion Review. 2016; 8(2): pp. 144–153.
- 27. Decety J, Meyer M. From emotion resonance to empathic understanding: A social developmental neuroscience account. 2008. Development and Psychopathology. 20(4): pp. 1053–1080. pmid:18838031
- 28. Rogers CR. The necessary and sufficient conditions of therapeutic personality change. 1957. Journal of Consulting Psychology. 21(2): pp. 95–103. pmid:13416422
- 29. Steinbeis N. The role of self–other distinction in understanding others’ mental and emotional states: Neurocognitive mechanisms in children and adults. 2016. Philosophical Transactions of the Royal Society B: Biological Sciences. 371(1686). pmid:26644593
- 30. Zahavi D, Rochat P. Empathy ≠ sharing: Perspectives from phenomenology and developmental psychology. 2015. Consciousness and Cognition. 36: pp. 543–553. pmid:26070850
- 31. King RB, Datu JA. Happy classes make happy students: Classmates’ well-being predicts individual student well-being. Journal of School Psychology. 2017; 65: pp. 116–128. pmid:29145940
- 32. Oberle E, Schonert-Reichel KA. Stress contagion in the classroom? The link between classroom teacher burnout and morning cortisol in elementary school students. Social Science & Medicine. 2016; 159: pp. 30–37. pmid:27156042
- 33. Goodman CR, Shippy RA. Is it contagious? Affect similarity among spouses. Aging & Mental Health. 2002; 6(3): pp. 266–274. pmid:12217095
- 34. Kurtz LE, Algoe SB. Putting laughter in context: Shared laughter as behavioral indicator of relationship well-being. Personal Relationships. 2015; 22(4): pp. 573–590. pmid:26957946
- 35. Niedtfeld I. Experimental investigation of cognitive and affective empathy in Borderline Personality Disorder: Effects of ambiguity in multimodal social information processing. Psychiatry Research. 2017; 253: pp. 58–63. pmid:28351003
- 36. Reck C, Hunt A, Fuchs T, Weiss R, Noon A, Moehler E, Downing G, Tronick EZ, Mundt C. Interactive regulation of affect in postpartum depressed mothers and their infants: An overview. Psychopathology. 2004; 37(6): pp. 272–280. pmid:15539778
- 37. Rosner R, Beutler LE, Daldrup RJ. Vicarious emotional experience and emotional expression in group psychotherapy. Journal of Clinical Psychology. 2000; 56(1): pp. 1–10. pmid:10661364
- 38. Sels L, Ceulemans E, Bulteel K, Kuppens P. Emotional interdependence and well-being in close relationships. Frontiers in Psychology. 2016; 7. pmid:27014114
- 39. Tronick EZ, Reck C. Infants of depressed mothers. Harvard Review of Psychiatry. 2009; 17(2): pp. 147–156. pmid:19373622
- 40. Waters SF, West TV, Mendes WB. Stress contagion: Physiological covariation between mothers and infants. Psychological Science. 2014; 25(4): pp. 934–942. pmid:24482403
- 41. Waters SF, West TV, Karnilowicz HR, Mendes WB. Affect contagion between mothers and infants: Examining valence and touch. Journal of Experimental Psychology: General. 2017; 146(7): pp. 1043–1051. pmid:28493755
- 42. Moll T, Jordet G, Pepping G-J. Emotional contagion in soccer penalty shootouts: Celebration individual success is associated with ultimate team success. Journal of Sports Sciences. 2010; 28(9): pp. 983–992. pmid:20544488
- 43. Wheeler L. Toward a theory of behavioral contagion. Psychological Review. 1966; 73(2): pp. 179–192.
- 44. Doherty RW. The Emotional Contagion Scale: A measure of individual differences. Journal of Nonverbal Behavior. 1997; 21(2): pp. 131–154.
- 45. Carré A, Stefaniak N, D’Ambrosio F, Bensalah L, Besche-Richard C. The Basic Empathy Scale in adults (BES-A): Factor structure of a revised form. 2013. Psychological Assessment. 25(3): pp. 679–691. pmid:23815121
- 46. Reniers RLEP, Corcoran R, Drake R, Shryane NM, Völlm BA. The QCAE: A Questionnaire of Cognitive and Affective Empathy. 2011. Journal of Personality Assessment, 93(1): pp. 84–95. pmid:21184334
- 47.
Caruso DR, Mayer JD. A measure of emotional empathy for adolescents and adults [Unpublished Manuscript]. 1998. Retrieved from https://mypages.unh.edu/sites/default/files/jdmayer/files/empathy_article_2000.pdf
- 48. Rieffe C, Ketelaar L, Wiefferink CH. Assessing empathy in young children: Construction and validation of an Empathy Questionnaire (EmQue). 2010. Personality and Individual Differences. 49(5): pp. 362–367.
- 49. Richaud MC, Lemos VN, Mesurado B, Oros L. Construct validity and reliability of a new spanish empathy questionnaire for children and early adolescents. 2017. Frontiers in Psychology. 8. pmid:28659848
- 50. Bakker AB, Schaufeli WB, Sixma, HJ, Bosveld W. Burnout contagion among general practitioners. Journal of Social and Clinical Psychology. 2001; 20(1): pp. 82–98.
- 51. Le Blanc PM, Bakker AB, Peeters MCW, van Heesch NCA, Schaufeli WB. Emotional job demands and burnout oncology care providers. Anxiety, Stress & Coping: An International Journal. 2001; 14(3): pp. 243–263.
- 52. Miller KI, Birkholt M, Scott C, Stage C. Empathy and burnout in human service work: An extension of a communication model. Communication Research. 1995; 22(2): pp. 123–147.
- 53. Verbeke W. Individual differences in emotional contagion of salespersons: Its effect on performance and burnout. Psychology & Marketing. 1997; 14(6): pp. 617–636.
- 54. Watson D, Tellegen A. Toward a consensual structure of mood. Psychological Bulletin. 1985; 98(2): pp. 219–235. pmid:3901060
- 55.
Ekman P. Basic emotions. In Dalgleish T, Power MJ, editors. Handbook of Cognition and Emotion. John Wiley & Sons Ltd.; 1999. pp. 45–60. https://doi.org/10.1002/0470013494.ch3
- 56. Ekman P, Cordaro D. What is meant by calling emotions basic. Emotion Review. 2011; 3(4): pp. 364–370.
- 57.
Izard CE. Differential emotions theory. In Izard CE, editor. Human emotions. Springer US; 1977.
- 58. Tomkins SS, McCarter R. What and where are the primary affects? Some evidence for a theory. Perceptual and Motor Skills. 1964; 18(1): pp. 119–158. pmid:14116322
- 59. Watson D, Clark LA, Tellegen A. Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology. 1988; 54(6): pp. 1063–1070. pmid:3397865
- 60. Watson D, Wiese D, Vaidya J, Tellegen A. The two general activation systems of affect: Structural findings, evolutionary considerations, and psychobiological evidence. Journal of Personality and Social Psychology. 1999; 76(5): pp. 820–838.
- 61. Watson D, Clark LA. Affects separable and inseparable: On the hierarchical arrangement of the negative affects. Journal of Personality and Social Psychology. 1992; 62(3): pp. 489–505.
- 62. Goldstein MD, Strube MJ. Independence revisited: the relation between positive and negative affect in a naturalistic setting. Personality and Social Psychology Bulletin. 1994: 20(1): pp. 57–64.
- 63. Bagozzi RP, Wong N, Yi Y. The role of culture and gender in the relationship between positive and negative affect. Cognition and Emotion. 1999; 13(6): pp. 641–672.
- 64. Blaxton JM, Nelson NA, Bergeman CS. The positive and negative affect relation in the context of stress and age. Emotion. 2021; 21(8): pp. 1712–1720. pmid:34941321
- 65. Potter PT, Zautra AJ, Reich JW. Stressful events and information processing dispositions moderate the relationship between positive and negative affect: Implications for pain patients. Annals of Behavioral Medicine. 2000; 22: pp. 191–198. pmid:11126463
- 66. Diener E, Emmons RA. The independence of positive and negative affect. Journal of Personality and Social Psychology. 1984; 47(5): pp. 1105–1117. pmid:6520704
- 67. Schmukle SC, Egloff B, Burns LR. The relationship between positive and negative affect in the Positive and Negative Affect Schedule. Journal of Research in Personality. 2002; 36(5): pp. 463–475.
- 68. Light SN, Moran ZD, Zahn-Waxler C, Davidson RJ. The measurement of positive valence forms of empathy and their relation to anhedonia and other depressive symptomatology. Frontiers in Psychology. 2019; 10. pmid:31031680
- 69. Morelli SA, Lieberman MD, Zaki J. The emerging study of positive empathy. Social and Personality Psychology Compass. 2015; 9(2): pp. 57–68.
- 70. Lundqvist L-O. A swedish adaption of the Emotional Contagion Scale: Factor structure and psychometric properties. Scandinavian Journal of Psychology. 2006; 47(4): pp. 263–272. pmid:16869859
- 71. Jordan MR, Amir D, Bloom P. Are empathy and concern psychologically distinct? Emotion. 2016; 16(8): pp. 1107–1116. pmid:27668802
- 72. Trautmann S, Reineboth M, Trikojat K, Richter J, Hagenaars MA, Kanske P, Schäfer J. Susceptibility to others’ emotions moderates immediate self-reported and biological stress responses to witnessing trauma. Behaviour Research and Therapy. 2018; 110: pp. 55–63. pmid:30243101
- 73. Frenzel AC, Fiedler D, Marx AKG, Reck C, Pekrun R. Who enjoys teaching, and when? Between-and within-person evidence on teachers’ appraisal-emotion links. Frontiers in Psychology. 2020; 11:1092. pmid:32636775
- 74. Schwartze MM, Frenzel AC, Goetz T, Marx AKG, Reck C, Pekrun R. Excessive boredom among adolescents: A comparison between low and high achievers. 2020; PLoS ONE. 15(11): e0241671. pmid:33152022
- 75.
Schaufeli WB, Buunk A. Burnout: An overview of 25 years of research and theorizing. In Schabracq MJ, Winnubst JAM, Cooper CL, editors. The Handbook of Work and Health Psychology. John Wiley & Sons Ltd.; 2003.
- 76.
Anderman LH, Klassen RM. Being a teacher: Efficacy, emotions, and interpersonal relationships in the classroom. In: Corno L, Anderman LH, editors. Handbook of Educational Psychology. Routledge; 2015. pp. 416–428. https://doi.org/10.4324/9781315688244-42
- 77. García-Carmona M, Marín MD, Aguayo R. Burnout Syndrome in secondary school teachers: A systematic review and meta-analysis. Social Psychology of Education: An International Journal. 2019; 22(1): pp. 189–208.
- 78. Taxer JL, Frenzel AC. Facets of teachers’ emotional lives: a quantitative investigation of teachers’ genuine, faked, and hidden emotions. Teaching and Teacher Education. 2015; 42: pp. 78–88. 03.003
- 79. Mertler CA. Patterns of response and nonresponse from teachers to traditional and web surveys. Practical Assessment, Research, and Evaluation. 2003; 8: pp. 353–371.
- 80. Batchelder L, Brosnan M, Ashwin C. The development and validation of the Empathy Components Questionnaire (ECQ). PLOS ONE. 2017; 12(1): e0169185. pmid:28076406
- 81.
Maslach C, Jackson SE, Leiter MP. Maslach Burnout Inventory: Third edition. In Zalaquett CP, Wood RJ, editors. Evaluating stress: A book of resources. Scarecrow Education; 1997.
- 82.
Enzmann D, Kleiber D. Helfer-Leiden: Streß und Burnout in psychosozialen Berufen. Asanger; 1989.
- 83. Sijtsma K. On the use, the misuse, and the very limited usefulness of Cronbach’s alpha. Psychometrika. 2009; 74(1): pp. 107–120. pmid:20037639
- 84. Korkmaz S, Goksuluk D, Zararsiz G. MVN: An R package for assessing multivariate normality. The R Journal. 2014; 6: pp. 151–162.
- 85.
Field A, Miles J, Field Z. Discovering statistics using R. SAGE; 2012.
- 86. Davis MH. Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality and Social Psychology. 1983; 44(1): pp. 113–126.
- 87. Vossen HGM, Piotrowski JT, Valkenburg PM. Development of the Adolescent Measure of Empathy and Sympathy (AMES). Personality and Individual Differences. 2015; 74: pp. 66–71.
- 88. Cox JL, Holden JM, Sagovsky R. Detection of postnatal depression. Development of the 10-item Edinburgh Postnatal Depression Scale. The British Journal of Psychiatry: The Journal of Mental Science. 1987; 150: pp. 782–786. pmid:3651732
- 89. Spitzer RL, Kroenke K, Williams JBW, Löwe B. A brief measure for assessing Generalized Anxiety Disorder: The GAD-7. Archives of Internal Medicine. 2006; 166(10): pp. 1092–1097. pmid:16717171
- 90. Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. Journal of Health and Social Behavior. 1983; 24(4): pp. 385–396. pmid:6668417
- 91. Cohen S, Hoberman HM. Positive events and social supports buffers of life change stress. Journal of Applied Social Psychology. 1983; 13(2): pp. 99–125.
- 92. Glaesmer H, Grande G, Braehler E, Roth M. The german version of the Satisfaction With Life Scale (SWLS). European Journal of Psychological Assessment. 2011; 27(2): pp. 127–132.
- 93. Soto CJ, John OP. The next Big Five Inventory (BFI-2): Developing and assessing a hierarchical model with 15 facets to enhance bandwidth, fidelity, and predictive power. Journal of Personality and Social Psychology. 2017; 113(1): pp. 117–143. pmid:27055049
- 94. Kemper CJ, Beierlein C, Bensch D, Kovaleva A, Rammstedt B. Eine Kurzskala zur Erfassung des Gamma-Faktors sozial erwünschten Antwortverhaltens: Die Kurzskala Soziale Erwünschtheit-Gamma (KSE-G). GESIS Working Papers; 2012.
- 95. Paulus C. Der Saarbrücker Persönlichkeitsfragebogen SPF (IRI) zur Messung von Empathie: Psychometrische Evaluation der deutschen Version des Interpersonal Reactivity Index [Internet]. 2009. Available from: https://psycharchives.org/en/item/2b090feb-4f9e-4ea2-9524-13151172f615
- 96. Zengin H, Çaka SY, Çinar ND. Adaptation of the Adolescent Measure of Empathy and Sympathy (AMES) to Turkish: A validity and reliability study. Alpha Psychiatry. 2017; 19(2): pp. 184–191.
- 97. Danner D, Rammstedt B, Bluemke M, Lechner C, Berres S, Knopf T, Soto CJ, John OP. Das Big Five Inventar 2. Diagnostica. 2019; 65(3): pp. 121–132.
- 98. Krohne HW, Egloff B, Kohlmann C-W, Tausch A. Untersuchungen mit einer deutschen Version der "Positive and Negative Affect Schedule" (PANAS). Diagnostica. 1996; 42(2): pp. 139–156.
- 99. Matijasevich A, Munhoz TN, Tavares BF, Barbosa APPN, da Silva DM, Abitante MS, Dall’Agnol TA, Santos IS. Validation of the Edinburgh Postnatal Depression Scale (EPDS) for screening of Major Depressive Episode among adults from the general population. BMC Psychiatry. 2014; 14. pmid:25293375
- 100. Matthey S, Barnett B, Kavanagh DJ, Howie P. Validation of the Edinburgh Postnatal Depression Scale for men, and comparison of its item endorsement with their partners. Journal of Affective Disorders. 2001; 64(2–3): pp. 175–184. pmid:11313084
- 101. Löwe B, Müller S, Brähler E, Kroenke K, Albani C, Decker O. Validierung und Normierung eines kurzen Selbstratinginstruments zur Generalisierten Angst (GAD-7) in einer repräsentativen Stichprobe der deutschen Allgemeinbevölkerung. PPmP—Psychotherapie · Psychosomatik · Medizinische Psychologie. 2007; 57(2): A050.
- 102. Reis D, Lehr D, Heber E, Ebert DD. The German version of the Perceived Stress Scale (PSS-10): Evaluation of dimensionality, validity, and measurement invariance with exploratory and confirmatory bifactor modeling. Assessment. 2019; 26(7): pp. 1246–1259. pmid:28627220
- 103. Hall NC, Chipperfield JG, Perry RP, Ruthig JC, Goetz T. Primary and secondary control in academic development: Gender-specific implications for stress and health in college students. Anxiety, Stress, & Coping. 2006; 19(2): pp. 189–210.
- 104. Perinelli E, Gremigni P. Use of social desirability scales in clinical psychology: A systematic review. Journal of Clinical Psychology. 2016; 72(6): pp. 534–551. pmid:26970350
- 105. Rosseel Y. lavaan: An R package for structural equation modeling. Journal of Statistical Software. 2012; 48(2): pp. 1–36.
- 106.
R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing; 2019.
- 107. Hu L, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling. 1999; 6(1): pp. 1–55.
- 108.
Kline RB. Principles and practice of structural equation modeling, 2nd edition. Guilford Press; 2005.
- 109.
Rosseel Y. The lavaan tutorial [Unpublished Manuscript]. 2020. Available from: http://lavaan.ugent.be/tutorial/tutorial.pdf
- 110. Havlicek LL, Peterson NL (1976). Robustness of the Pearson correlation against violations of assumptions. Perceptual and Motor Skills. 1976; 43(3_suppl): pp. 1319–1334.
- 111. Cumming G, Fidler F, Kalinowski P, Lai J. The statistical recommendations of the American Psychological Association publication manual: Effect sizes, confidence intervals, and meta-analysis. Australian Journal of Psychology. 2012; 64(3): pp. 138–146.
- 112. Avolio BJ, Yammarino, FJ, Bass BM. Identifying common methods variance with data collected from a single source: An unresolved sticky issue. Journal of Management. 1991; 17(3): 571–587.
- 113. Schönbrodt FD, Perugini M. At what sample size do correlations stabilize? Journal of Research in Personality. 2013; 47(5): pp. 609–612.
- 114. McCredie MN, Morey LC. Who are the turkers? A characterization of MTurk workers using the Personality Assessment Inventory. Assessment. 2019; 26(5): pp. 759–766. pmid:29490470
- 115. Ophir Y, Sisso I, Asterhan CSC, Tikochinski R, Reichart R. The turker blues: Hidden factors behind increased depression rates among Amazon’s Mechanical Turkers. Clinical Psychological Science. 2020; 8(1): pp. 65–83.
- 116. Sun Y, Wu Y, Fan S, Dal Santo T, Li L, Jiang X et al. Comparison of mental health symptoms before and during the covid-19 pandemic: evidence from a systematic review and meta-analysis of 134 cohorts. BMJ. 2023; 380:e074224. pmid:36889797
- 117. Kaubisch LT, Reck C, von Tettenborn A, Woll CFJ. The COVID-19 pandemic as a traumatic event and the associated psychological impact on families—a systematic review. Journal of affective disorders. 2022; 319: pp. 27–39. pmid:36089074
- 118. Kiltz L, Trippenzee M, Fleer J, Fokkens-Bruinsma M, Jansen EPWA. Student well-being in times of COVID-19 in the Netherlands: basic psychological need satisfaction and frustration within the academic learning environment. European Journal of Psychology of Education. 2023.
- 119. Depow GJ, Francis Z, Inzlicht M. The experience of empathy in everyday life. Psychological Science. 2021; 32(8): pp. 1198–1213. pmid:34241543
- 120. De Vuyst H-J, Dejonckheere E, Van der Gucht K, Kuppens P. Does repeatedly reporting positive or negative emotions in daily life have an impact on the level of emotional experiences and depressive symptoms over time? PLOS ONE. 2019; 14(6). pmid:31247033
- 121. Harari GM, Lane ND, Wang R, Crosier BS, Campbell AT, Gosling SD. Using smartphones to collect behavioral data in psychological science: Opportunities, practical considerations, and challenges. Perspectives on Psychological Science: A Journal of the Association for Psychological Science. 2016; 11(6): pp. 838–854. pmid:27899727
- 122. Harari GM, Müller SR, Stachl C, Wang R, Wang W, Bühner M, Rentfrow PJ, Campbell AT, Gosling SD. Sensing sociability: Individual differences in young adults’ conversation, calling, texting, and app use behaviors in daily life. Journal of Personality and Social Psychology. 2020; 119(1): pp. 204–228. pmid:31107054
- 123. Ekman P, Friesen WV, Hager JC. Facial Action Coding System—The Manual (on CD ROM). Research Nexus division of Network Information Research Corporation; 2002.
- 124. Coco MI, Dale R. Cross-recurrence quantification analysis of categorical and continuous time series: An R package. Frontiers in Psychology. 2014; 5. pmid:25018736