This is an open access article distributed under the terms of the Creative Commons Attribution License (Kama muta, often referred to colloquially as the emotion of feeling moved (Zickfeld, Schubert, Seibt, Blomster, et al., 2019), is a positive emotion that emerges from a sudden intensification of communal relationships. It is frequently induced by affirmation and re-affirmation of an individual’s positive core values (on feeling moved; Cova & Deonna, 2014), with accompanying displays of moralistic qualities like love and courage. Cognitive appraisals include communal sharing, that is, the witnessing of increases (intensifications) in interpersonal closeness and connection in other people. Kama muta is also associated with strong physiological sensations, such as tears, goosebumps, and subjective feelings of warmth in the chest (Zickfeld et al., 2020). Often, kama muta is experienced in response to difficulties (Zickfeld et al., 2020), such as natural disasters and even the COVID-19 pandemic. With the increase in these globalized threats, it is important to understand the function of kama muta in human society: Based on past literature, we believe that it helps to build up an individual’s resources for prosocial action, which would be societally adaptive in trying times. However, these functions of kama muta, especially in the context of collective threat, have not been studied much in past literature (Zickfeld, Schubert, Seibt, & Fiske, 2019). Our research therefore examines the notion that socially impactful collective threat can enhance experiences of kama muta, which may in turn alleviate individuals' attitudes towards personal troubles, thereby facilitating a broader prosocial function.
Kama Muta Reduces the Perceive Weight of Individual’s Personal Problems
Like many other positive emotions, kama muta may function to help the individual overcome negative situations (Fredrickson et al., 2003), by building resilience through value-reorientation. This reduces the perceived weight of one’s personal problems, enabling individuals to adopt new motivations or concerns for issues they might not have cared about before, increasing self-reflection and changes in what the individual perceives as meaningful (Landmann et al., 2019). This process of reorientation may then promote individuals’ insight into their problems (Lyke, 2009) and improve well-being (Sagiv & Schwartz, 2000; Uchida et al., 2014). Accordingly, we posit that kama muta may exert a downward shift in the perspective of the seriousness of one’s personal problems and anxieties; kama muta may positively reorientate individuals’ existing negative attitudes and appraisals towards their personal problems (henceforth labeled as 'personal problem' appraisals [PPA]; Hypothesis 1).
This can also be understood through cognitive reappraisal processes in emotion regulation (Gross & John, 2003), specifically in positive reorientation (Zhou et al., 2020), where individuals experiencing distress may cognitively re-interpret their negative situation in a positive light. Procedurally, this similarly involves meaning-making processes, such as the reappraising of negative events as (positively) meaningful in the ‘greater scheme of things’ (Stellar et al., 2017), which are linked to downstream effects of experiences of self-transcendental emotions like awe (the ‘small self’ effect, Piff et al., 2015) and kama muta (Pizarro et al., 2021). Contextually, this suggests that kama muta may help individuals find meaning through adversity, thereby reducing the negative attitudes and appraisals towards adverse events (personal problem appraisals).
Kama Muta and Collective Threat: The Context for Value Change
Relatedly, we attempt to determine the conditions by which kama muta can be enhanced, thereby facilitating reductions in personal problem appraisals. Exploratory studies on participants' descriptions of moving experiences have shown that natural disasters form a prototypical backdrop for moving experiences, especially in media (Menninghaus et al., 2015). Despite apparent harm, exposure to collective threat, such as natural disasters, may increase meaning-making processes (Nakayama & Uchida, 2020; Uchida et al., 2014), causing beneficial value shifts in the individual and building up resilience (Bonanno et al., 2010). Individuals exposed to a nation-wide natural disaster, even without experiencing direct harm, reported higher well-being and reevaluation of attitudes towards life, alongside increased commitment to meaning-related self-transcendent values and social connections (Uchida et al., 2014). Kama muta may also facilitate some of these changes, given that natural disasters consistently appear as a backdrop for feeling moved (Menninghaus et al., 2015; Note & Van Daele, 2016). Therefore, in this paper, we define collective threat to be a context in which large-scale threats have direct and indirect adverse consequences on a community of people (e.g., war, natural disasters, pandemic).
We think that kama muta most strongly emerges against a backdrop of difficulties. Kama muta requires some context-related knowledge (Tokaji, 2003), whereby the preciousness of one’s relationships and values is brought to attention when it is threatened, in contexts such as separation or loss (Cullhed, 2020). For example, natural disasters, such as earthquakes, are typically associated with collective threat and loss of life, but juxtaposed within these negative events are often small stories of human connection: a mother reunited with a missing child, or rescue workers showing courage and bravery to save survivors. In these contexts, the sharp contrast between the triumph of positive core values (such as love and courage) and the prevailing adverse circumstances makes every positive action more salient and dearer to the individual (Cova & Deonna, 2014; Strick & van Soolingen, 2018). Witnessing intensifications of heartwarming social bonds (Schubert et al., 2018) thereby leads to strong experiences of kama muta through activating appraisals on the sudden intensification of communal sharing (as an appraisal dimension of kama muta, Zickfeld, Schubert, Seibt, Blomster, et al., 2019). Seibt and colleagues (2023), for example, found that participants who read first-person accounts of a collective threat incident (typhoon) experienced more kama muta than participants who read neutral, fact-based reports of the same incident. Kama muta thus motivates individuals to show more empathic concern, and may be societally adaptive for promoting cooperation: theories from cultural evolution have argued that societies with stronger historical presence of collective threat (like natural disasters or pathogen prevalence) foster greater cooperation and social capital (Henrich & Muthukrishna, 2021; Varnum, 2014). Accordingly, collective threat may form the backdrop for kama muta (Hypothesis 2), allowing for the emergence of kama muta experiences that may, in turn, facilitate value change and meaning making processes.
Overview of Current Studies
Across three studies, we utilize the KAMMUS-Two scale (Zickfeld, Schubert, Seibt, Blomster, et al., 2019), which measures physiological, affective responses and action tendencies of kama muta (Petersen & Martin, 2021).
We examined the effect of feeling kama muta on participants’ personal problem appraisals (stronger experiences of kama muta predicts decreased personal problem appraisals; Hypothesis 1) and the effect of collective threat on kama muta (threat elicits stronger experiences of kama muta; Hypothesis 2). We first establish a connection between collective threat, kama muta and personal problem appraisals, using experimental manipulation (Study 1) and an ongoing natural disaster (Study 2) as the collective threat contexts.
Study 3 then focused on examining the effect of different forms of collective threat on kama muta and replicating the effect of kama muta on personal problem appraisals. We also measured the perceived societal impact of collective threat in each condition as a gauge of the potential consequence of such threat on society. To this end, we manipulated the degree of collective threat via the saliency of societal impact (Study 3). Across all three studies, measures of awe, happiness and gratitude were also included for all studies to examine if feeling moved exerted a unique effect apart from other similarly functioning positive emotions (Koh et al., 2019; Maeura et al., 2020; Stellar et al., 2017).
All studies were approved by the institutional review board of Kyoto University. Portions of the studies were pre-registered (Study 1: Koh, 2019a; Study 3: Koh, 2019b). Unless otherwise stated, all analyses were conducted in jamovi (v1.6.23), and significance testing used alpha = .05. Data, analysis scripts, descriptive statistics, and power analyses are available for all studies in the OSF repository (see Liew & Koh, 2023).
Study 1
Method
Participants
322 Japanese participants were recruited through an online crowdsourcing platform (Lancers [lancers.jp]) and reimbursed 200 yen for participating. Sample sizes were limited by available funding at the time of the study. 20 participants were dropped from the analysis, for failing the attention checks and/or not following the experimental procedure, resulting in a final N = 302 (51% males, mean age = 40.5, SD = 9.34). Participants were assigned to one of three conditions: Threat+Kama muta (N = 104), Neutral+Kama muta (N = 103), and Threat-only (N = 95).
Procedure and Materials
Participants were presented with detailed study information before providing informed consent. They were then prompted to describe in detail up to three different problems that they were currently struggling with. The problems could be anything they were concerned about and had been troubled with. Participants then rated how troubled, worried, difficult, severe and confident in overcoming (reverse-coded) their problems they were, using a slider scale from 1 (not ___ at all) to 100 (extremely ___), for each respective item. All slider scales were initially anchored from the left at 0. The average rating formed the measure for their personal problem appraisals at Time 1 (ProbT1, α = .83). They also rated how common they thought their problems were and their current mood. As these were exploratory measures, further analyses are included in the supplementary materials and will not be discussed within this paper (See Supplementary Materials, Section 2: Problem Commonality and Mood Analyses, Liew & Koh, 2023).
Next, participants proceeded to the threat and feeling moved manipulation tasks according to their assigned condition (“Threat+Kama muta”, “Neutral+Kama muta”, “Threat-only”). Participants assigned to the focal condition, Threat+Kama muta, first watched a video depicting collective threat, that comprised clips of thunderstorms, rain and rolling dark clouds (links are in Supplementary Materials, Section 4, Liew & Koh, 2023). Next, they watched a second video depicting social connection: A Japanese commercial, depicting the opening of a new train line connecting towns and prefectures physically and emotionally/psychologically (e.g., celebrating the new train line together), where people of all ages came together to cheer on the passing train. This was used as the kama muta target stimuli as it allowed participants to observe human connection and bonding without depicting any unfortunate circumstances.
Participants assigned to the Neutral+Kama muta condition, as a control condition for collective threat, first watched a neutral video consisting of a feather resting on a patch of grass, swaying slightly in the breeze. The neutral video was chosen to include nature without its threatening aspects. Participants then watched the commercial described above as the target for kama muta.
Participants assigned to the Threat-only condition, as a control for the kama muta stimuli, watched only the collective threat video before moving on to the next part of the study.
All participants then rated their experiences of kama muta, using the KAMMUS-Two scale, with 24-items (α = .94) from 0 (not at all) to 6 (very much), which were averaged to form the kama muta measure (for alternative calculations see Supplementary Materials: Section 3, Liew & Koh, 2023). They also rated the extent they felt awe1, grateful and happy using the same Likert scale.
Lastly, they were asked to recall the problems they listed earlier and briefly list them again. They rated their personal problems using the same items at Time 1, forming the measure ProbT2. They also completed other unrelated questions2 and a demographics questionnaire.
KAMMUS-Two
The KAMMUS-Two scale is our main measure of kama muta used in this project. It measures the five aspects of kama muta experiences by asking participants about their experiences across five different factors (Zickfeld, Schubert, Seibt, Blomster, et al., 2019): 1) Bodily sensations (12 items; e.g., "To what extent... (the participant experienced) tears"), 2) (appraisals on) intensification of communal sharing relationships (4 items; e.g., "I felt/observed an incredible bond"), 3) social motivation (4 items; e.g., "I wanted to do something extra-nice for someone"), positive valence (1 item; "I had positive feelings"), and vernacular labels (3 items; e.g., "I was moved"). All items were rated on a scale from 0 (not at all) to 6 (a lot).
Results
Manipulation Checks
The video stimuli were tested in a separate study to prevent demand characteristics. 55 Japanese participants watched and rated the collective threat and neutral video using a 7 point-Likert scale, from 1 (not at all) to 7 (extremely). A repeated measure analysis revealed that the collective threat video (M = 4.50, SD = 2.06) was significantly higher in threat than the neutral video (M = 1.29, SD = 0.88), F(1,102) = 215, p < .001, = 0.68.
Outcomes on Personal Problem Appraisals (Hypothesis 1)
ProbT2 was regressed on kama muta and ProbT1, R2 = 0.74, F(2,299) = 430, p < .001. Kama muta significantly and negatively predicted ProbT2 even after controlling for ProbT1 (see Table 2). The stronger the individuals' experiences of kama muta, the lower the personal problem appraisals. Participants’ ratings of awe, gratitude and happiness were then entered as additional controls, R2 = 0.74, F(5,296) = 172, p < .001. Only ProbT1 emerged as a significant predictor (see Table 2). A one-way ANOVA also showed no effect of condition on ProbT2, F(2,299) = .037, p = .964, ω2 = -.006.
ProbT2 was then regressed on Cond and ProbT1, R2 = 0.73, F(3,292) = 271, p < .001, with the Threat-only stimuli as the reference level. No significant effect was observed for Threat+Kama muta nor Neutral+Kama muta (see Table 2).
Collective Threat and Kama Muta (Hypothesis 2)
A Welch’s ANOVA on the experience of kama muta as a function of context was significant, F(2,198) = 32.4, p < .001, ω2 = .172. Post-hoc Games-Howell tests revealed higher experiences of kama muta in the Threat+Kama muta condition than the Threat-only condition, t(187) = 6.93, p < .001, Cohen’s d = 0.97 (95% CI [0.66, 1.28]). Kama muta was also higher in the Neutral+Kama muta condition than the Threat-only condition, t(189) = 6.40, p < .001, Cohen’s d = 0.91 (95% CI [0.59, 1.20]). The Threat+Kama muta and Neutral+Kama muta conditions did not differ, p = .80 (see Figure 1).
Figure 1
The Effect of Condition on Kama Muta
Note. Rated kama muta (KAMMUS-Two scale) after watching the video stimuli (“Threat+Kama muta”, “Neutral+Kama muta”, “Threat-only”). Central tendencies are represented by the medians (line) and means (square). Boxes indicate 25th and 75th percentiles, and whiskers indicate upper and lower boundaries at 1.5 times the interquartile range.
Function of Kama Muta
As an exploratory step, kama muta was tested as a mediator between condition and change in personal problem appraisals (as a difference score of ProbT2-T1). Condition was dummy coded. Neutral+Kama muta contrast coded the Neutral+Kama muta condition as 1 and the others as 0. Threat-only contrast coded the Threat-only condition as 1 and the others as 0. The Threat-only contrast significantly predicted kama muta (b = -0.97, SE = .14, 95% CI [-1.22, -0.68], B = -.40, Z = -7.04, p < .001), which in turn predicted change in personal problem appraisals (b = -1.95, SE = 0.55, 95% CI [-3.04, -0.88], B = -.22, Z = -3.52, p < .001). The indirect effect of Threat-only contrast on change in personal problem appraisals, mediated by kama muta, was significant (b = 1.89, SE = .63, 95% CI [0.81, 3.25], B = 0.09, Z = -3.00, p = .003). The effect of Neutral+Kama muta contrast was not significant: Neutral+Kama muta contrast did not significantly predict kama muta (b = -1.14, Z = -0.94, p = .35), which did not predict change in personal problem appraisals (b = -1.49, Z = -0.93, p = .35), and the indirect effect was not significant (b = .19, Z = 0.60, p = .55). This suggests that the Kama muta stimuli conditions (Threat+kama muta, Neutral+Kama muta), over the Threat-only condition, significantly predicted kama muta ratings, which then significantly predicted reduced PPAs. Confidence intervals were based on a bootstrapped model with a resampling size of 5000 (see Figure 2).
Figure 2
Kama Muta Mediates Threat and PPA
Note. Significant mediation effect of kama muta on the relationship between Threat-only contrast and change in personal problem appraisals. All coefficients reported are unstandardized. Note that total effects (c) and direct effects (c’) were not significant.
***p < .001.
Discussion
Higher ratings of kama muta were associated with lower personal problem appraisals, and this effect remained after controlling for other positive emotions of happiness, gratitude and awe. However, this was dependent not on the threat or kama muta video stimuli, but on the perceived extent to which participants subjectively felt kama muta. Our mediation analysis provides some evidence of the effect of kama muta on PPA: A significant indirect effect of kama muta stimuli conditions was observed on reductions in PPA as mediated by perceived kama muta. To some extent, we think that these results may support Hypothesis 1.
In contrast to Hypothesis 2, no significant effect of threat was observed: no significant difference was observed between threat and neutral videos preceding kama muta induction. One possibility could be the remoteness of the threat to the individual, such that the ‘perceived’ danger of the threat was not salient. Accordingly, we conducted Study 2, which used similar materials and methods as Study 1, but with an actual threat context—participants were recruited in the midst of an ongoing nationwide natural disaster.
Study 2
To test the effects of threat on kama muta induction in an ecologically valid context (Hypothesis 2), we conducted a study using an ongoing natural disaster as the threat context, as opposed to video stimuli as presented in Study 1. Additionally, we examined the reorientation effect of kama muta on negative attitudes towards personal problems (Hypothesis 1), to replicate the findings found in Study 1.
Method
Participants
135 Japanese participants were recruited through the same online platform during an ongoing natural disaster, involving landslides, heavy rainfall and flooding (Chappell, 2019). Due to the time-sensitive nature of the study, our sample size was limited to participants who responded to the recruitment collateral during the natural disaster. They were reimbursed 200 Japanese yen for their participation. 16 participants were dropped from the analysis for the same reasons mentioned in Study 1, resulting in a final sample of N = 119 (51.3% males, mean age = 39.2, SD = 10.8).
Procedure
Unlike Study 1, which was a between-subjects experiment, all participants underwent the same emotion induction procedure as the Neutral+Kama muta condition in Study 2, watching both the neutral and kama muta video stimuli described above. Additionally, after the Time 2 measures, they also rated how much they had been thinking about the ongoing natural disaster and associated damage recently, from 1 (not at all) to 5 (constantly) adapted from previous research (Uchida et al., 2014). The frequency of participants’ thoughts about the ongoing natural disaster represented the salience of the threat context and served as Study 2’s measure of threat in an ecological manner.
Results
Outcomes on Personal Problem Appraisals (Hypothesis 1)
ProbT2 was regressed on kama muta, ProbT1, R2 = 0.79, F(2,116) = 214.5, p < .001. Replicating Study 1, kama muta significantly predicted ProbT2 after controlling for ProbT1. When awe, grateful and happiness were entered as additional controls, R2 = 0.79, F(5,113) = 85.2, p < .001, kama muta remained a significant predictor, but none of the other 3 emotion predictors emerged significant (awe: b = -0.15, p = .84; happy: b = 1.16, p = .23; grateful: b = -0.56, p = .50).
Collective Threat and Kama Muta (Hypothesis 2)
Participants’ disaster thoughts significantly predicted kama muta, b = .391, t(117) = 3.34, p = .001, 95% CI [.159, .624]. Participants who thought more about the ongoing natural disaster around that week felt more kama muta while watching the commercial.
Function of Kama Muta
Kama muta was tested as a mediator between disaster thoughts and change in personal problem appraisals. The regression coefficients between disaster thoughts and kama muta (b = .39, SE = .11, 95% CI [.16, .60], B = .29, Z = 3.50, p < .001) and between kama muta and change in personal problem appraisals (b = -2.63, SE = 0.93, 95% CI [-4.51, -.86], B = -.28, Z = -2.82, p = .005) were significant. The indirect effect of disaster thoughts on change in personal problem appraisals, mediated by kama muta, was significant (b = -1.03, SE = .50, 95% CI [-2.30, -.28], B = -.08, Z = -2.04, p = .041). Confidence intervals were based on a bootstrapped model with a resampling size of 5000 (see Figure 3).
Figure 3
Kama Muta Mediates Disaster Thoughts and PPA
Note. Significant mediation effect of kama muta on the relationship between disaster thoughts and change in personal problem appraisals. All coefficients reported are unstandardized. Note that total effects (c) and direct effects (c’) were not significant.
***p < .001.
Discussion
As with Study 1, subjecting ratings of kama muta were associated with participants’ personal problem appraisals as hypothesized (Hypothesis 1) even after controlling for happiness, gratitude and awe. Additionally, Study 2 showed that kama muta mediated reductions in the burden of personal problems, when collective threat was more salient (Hypothesis 2). Unlike Study 1, in Study 2, the salience of real-life collective threat was associated with increased experiences of kama muta, suggesting that some aspects of collective threat may nevertheless be involved in heightened experiences of kama muta.
Study 3
One explanation could be that the inconsistent effect of threat in Study 1 may be due to differences in saliency of the harmful consequences of collective threat on society. Hence, we asked participants in the threat conditions to additionally read an article showing the direct consequence of the threat on humans (Human Life Threat) or as a general natural phenomenon (Nature Threat). We measured participants’ perceptions of the societal impact of the article as an indication of collective societal threat. We expected social impact to be more salient in the Human Life Threat condition than the Nature Threat condition, and kama muta to be correspondingly enhanced in the former. While already identified in Studies 1 and 2, we again tested if the effects of kama muta on personal problem appraisals (Hypothesis 1) could be consistently replicated.
Method
Participants
347 Japanese participants were recruited and reimbursed similar to Study 1. Sample sizes were limited by available funding at the time of the study. 43 participants were dropped from the analysis, for the same reasons mentioned in Study 1. The remaining 304 participants (49.7% males, mean age = 40.9, SD = 9.67) were assigned to one of the three conditions; Human Life Threat (N = 105), Nature Threat (N = 99) and Neutral (N = 100).
Procedure and Materials
All participants completed the same problem listing task and personal problem appraisal measures as in Study 1. Participants then proceeded to the collective threat manipulation task according to their assigned condition. Participants in both the Human Life and Nature Threat conditions first watched a thunderstorm video (from Study 1). Next, they were presented with an ostensible news article, with instructions to read it and answer related questions later on in the survey. In the Human Life Threat condition, participants read about several human victims who were injured or killed by lightning during a recent thunderstorm, to direct their focus to specific consequences of the collective threat. Participants in the Nature Threat condition read an article about the properties and potential effects of a thunderstorm like flooding and landslides, in order to direct their focus to the general consequences of the collective threat.
Participants in the Neutral condition watched the same neutral video from Study 1. They then read an article about the properties of raindrops, which have natural elements close to a thunderstorm but without its threatening aspects. This was to rule out possible societal impact due to scientific applications of natural phenomenon. All articles were of a similar length.
All participants proceeded to briefly describe the article they had read, and also indicated how much impact they thought the article’s content had on society using a slider scale from 0 (not at all) to 100 (extremely impactful) to measure perceived social (societal) impact. All participants then watched the moving commercial and completed the rest of the procedure described in Studies 1 and 2.
Results
Manipulation Check: Societal Impact
A one-way ANOVA revealed a significant effect of condition, F(2,301) = 42.6, p < .001, ω2 = .215. Contrary to our prediction, post-hoc Tukey tests showed that participants in the Nature Threat condition rated their article significantly higher in societal impact than participants in the Human Life Threat condition, t(301) = -4.76, p < .001, Cohen’s d = -0.67 95% CI [-0.95, -0.38]), and was also rated significantly higher in societal impact than the Neutral condition, t(301) = -9.24, p < .001, Cohen’s d = -1.30 (95% CI [-1.63, -0.97]) (see Table 2; Figure 4: left).
Figure 4
The Effects of Condition on Societal Impact and Kama Muta
Note. Left: Participants’ perceived societal impact (of collective threat) after watching the video stimuli and article reading. Right: Rated kama muta (KAMMUS-Two scale) after watching the video stimuli and article reading. Boxes indicate 25th and 75th percentiles, and whiskers indicate upper and lower boundaries at 1.5 times the interquartile range.
Outcome on Personal Problem Appraisals (Hypothesis 1)
ProbT1 and kama muta were first regressed onto ProbT2 in Step 1, R2 = 0.83, F(2,301) = 722, p < .001, followed by awe, happy and grateful in Step 2, R2 = 0.83, F(5,298) = 292, p < .001. Replicating Studies 1 and 2, kama muta significantly predicted ProbT2 when controlling for ProbT1, and also for other positive emotions and ProbT1. Awe, happiness and gratitude did not emerge as significant predictors, with the closest being Awe, b = 0.82, t(298) = 3.77, 95% CI [-.01, 1.65], p = .053.
Effect of Collective Threat on Kama Muta (Hypothesis 2)
A one-way ANOVA revealed a significant effect of condition, F(2,301) = 3.61, p = .028, ω2 = .017. Post-hoc Tukey tests showed that participants felt more kama muta in the Nature Threat condition (M = 2.80, SD = 1.03) than Neutral condition (M = 2.37, SD = 1.25), t(301) = -2.65, p = .023, Cohen’s d = -0.37 (95% CI [-0.65, -0.09]). All other comparisons were not significant (Human Life Threat – Neutral: t(301) = 0.98, p = .14; Human Life Threat – Nature Threat: t(301) = -1.71, p = .20; see Figure 4: right).
Function of Kama Muta
Kama muta was tested as a mediator between condition and change in personal problem appraisals. Condition was dummy coded. Human Life Threat contrast coded the Human Life Threat condition as 1 and the others as 0. Nature Threat contrast coded the Nature Threat condition as 1 and the others as 0. The Nature Threat contrast significantly predicted kama muta (b = .42, SE = .16, 95% CI [.10, .73], B = .18, Z = 2.61, p = .009), which in turn predicted change in personal problem appraisals (b = -1.86, SE = 0.44, 95% CI [-2.75, -1.003], B = -.24, Z = -4.19, p < .001). The indirect effect of Nature Threat contrast on change in personal problem appraisals, mediated by kama muta, was significant (b = -.79, SE = .35, 95% CI [-1.66, -.24], B = -.04, Z = -2.25, p = .025). However, the effect of Human Life Threat contrast on kama muta was not significant (b = 0.15, Z = -0.94, p = .35), and no significant indirect effect was observed for the effect of Human Life Threat contrast on change in personal problem appraisals (b = -0.28, Z = -0.92, p = .36) (see Figure 5).
Figure 5
Kama Muta Mediates Nature Threat and PPA
Note. Significant mediation effect of kama muta on the relationship between Nature Threat contrast and change in personal problem appraisals. All coefficients reported are unstandardized. Note that total effects (c) and direct effects (c’) were not significant.
***p < .001.
Given that perceived societal impact differed between the Nature and Human Life Threat conditions, we additionally explored if perceived societal impact influenced kama muta and subsequently the change in personal problem appraisals. Societal impact significantly predicted kama muta (b = .005, SE = .003, 95% CI [.00008, .01], B = .12, Z = 2.04, p = .041). However, the indirect effect of societal impact on change in personal problem appraisals, mediated by kama muta, was not significant (b = -.009, SE = .005, 95% CI [-.02, -.001], B = -.029, Z = -1.85, p = .064).
Discussion
On one hand, Study 3 again finds that participants who experienced stronger feelings of kama muta were also more likely to positively reorientate their prevailing negative attitudes towards their own problems (Hypothesis 1). Although the elicitor of collective threat was different from what we initially expected, collective threat with higher perceived societal impact was associated with enhanced experiences of kama muta, which was in turn associated with lower personal problem appraisals, somewhat supporting Hypothesis 2. This aligns with the proposed function of kama muta in lessening individuals’ mental burden in threatening situations. On the other hand, the effect of threat on kama muta was not significant, suggesting that any impact on kama muta ratings or personal problem appraisals did not directly result from our experimental threat conditions. Contrary to our hypothesis, participants who read the Nature Threat article perceived higher societal impact, and then felt more kama muta than those who read the neutral article. Here, what constituted societal impact differed from our initial predictions: the specificity and rarity of the victims’ plight in the Human Life Threat article may have narrowed the threat’s perceived impact on humanity, while the Nature Threat article was more effective in prompting the effect on society (Moreton et al., 2019). In any case, our experimental manipulation of societal impact, which measured its saliency to the participant, did not have a robust effect on kama muta.
Single Paper Metanalysis
Table 1
Group Descriptives for Kama Muta (KAMMUS-Two) Across All Three Studies
| Study | Condition | N | Kama muta | Societal impact | ||
|---|---|---|---|---|---|---|
| M | SD | M | SD | |||
| 1 | Threat + Kama muta | 104 | 2.77 | 1.14 | ||
| Neutral + Kama muta | 103 | 2.67 | 1.08 | |||
| Threat-only | 95 | 1.80 | 0.82 | |||
| 2 | n/a | 119 | 2.59 | 1.05 | ||
| 3 | Nature Threat | 99 | 2.80 | 1.03 | 59.9 | 24.2 |
| Human Life Threat | 105 | 2.53 | 1.10 | 44.0 | 23.6 | |
| Neutral | 100 | 2.37 | 1.25 | 28.6 | 23.8 | |
Table 2
PPA at T2 Regressed on Kama Muta and Condition (Studies 1 and 3) With T1 as a Covariate
| Study | Predictor | Model Covariates | Model R2 | b | t | df | 95% CI | p | |
|---|---|---|---|---|---|---|---|---|---|
| Upper | Lower | ||||||||
| 1 | Kama muta | T1 | 0.74 | -1.78 | -3.56 | 299 | -2.77 | -0.80 | < .001 |
| Kama muta | T1, awe, gratitude, happiness | 0.74 | -0.81 | -0.76 | 296 | -2.90 | 1.29 | .450 | |
| Condition: Threat+Kama muta | T1 | 0.73 | -0.39 | -0.28 | 298 | -3.14 | 2.35 | .778 | |
| Condition: Neutral+Kama muta | T1 | 0.73 | -1.29 | -0.93 | 298 | -4.05 | 1.46 | .356 | |
| 2 | Kama muta | T1 | 0.79 | -2.53 | -2.96 | 116 | -4.22 | -0.84 | .004 |
| Kama muta | T1, awe, gratitude, happiness | 0.79 | -3.09 | -2.05 | 113 | -6.09 | -0.10 | .043 | |
| 3 | Kama muta | T1 | 0.83 | -1.81 | -4.15 | 301 | -2.67 | -0.95 | < .001 |
| Kama muta | T1, awe, gratitude, happiness | 0.83 | -1.71 | -2.13 | 298 | -3.29 | -0.13 | .034 | |
Figure 6
Kama Muta and PPA Across All Studies
Note. Significant negative relationship between kama muta and (residualized) change in personal problem appraisals from all three studies, split by threat. Data from Study 2 is marked as “not applicable” due to the absence of experimental threat induction. Confidence bands show standard errors.
We conducted a participant-level single paper metanalysis to examine if kama muta would predict a reduction of participants’ personal problem appraisals (Hypothesis 1) and if collective threat predicted participants’ experiences of kama muta (Hypothesis 2). We initially followed the recommended protocol for Single Paper Metanalyses using hierarchical (multilevel) regressions as recommended by McShane and Böckenholt (2017), but as we only had 3 studies (and 2, for the effect of threat conditions on kama muta), we used a fixed effects only model, where study number was included as a control variable. For threat, study conditions that presented threat-inducing stimuli (Study 1: Threat+Kama Muta, Threat-only conditions; Study 3: Nature Threat and Human Life Threat conditions) were scored as ‘threat-induction’ conditions, and the remaining conditions were scored as ‘no-threat’ conditions. In examining the effect of self-reported kama muta on personal problem appraisals, we observed a significant and negative association with PPA at T2, controlling for PPA at T1 and study number (Hypothesis 1; model R2 = 0.79, B = -0.106, SE = 0.31, 95% CI [-0.14, -0.07], t = -6.12, p < .001). Regardless of the follow-up kama muta induction, no significant effect was observed for threat (stimuli) on kama muta (Hypothesis 2; model R2 = 0.004, B = -0.04, SE = 0.10, t = -0.43, p = .67) (see Figure 6).
General Discussion
Our results demonstrate that experiences of kama muta were consistently associated with reductions in participants’ negative attitudes and appraisals towards their personal problems. It joins a larger body of positive emotion, such as awe and gratitude (Gander et al., 2013; Koh et al., 2019), that function towards reducing negative affect or attitudes. Given that kama muta also strengthens the sense of connectedness, this result converges with past research, which has shown that a sense of connectedness can alleviate psychological distress (Aanes et al., 2010; Lee et al., 2001), reduce loneliness (Satici et al., 2016), increase positive affect (Adventure-Heart & Proeve, 2017), and improve well-being (Brown et al., 2012). However, one important caveat is that these effects were not observed as an effect of the condition, but as a correlation between participants’ ratings of kama muta, threat, and PPA. This limits any causal claims that can be made from this paper: the causal nature of Hypotheses 1 and 2 were not supported, and the directional effects observed between collective threat, kama muta, and personal problem appraisals only reflect an association between these variables. Nevertheless, there is some indirect evidence that participants induced to feel kama muta (in Study 1) subjectively experienced more intense kama muta, which was then associated with stronger reductions in PPA.
We think that kama muta may inform the individual about what he/she perceives as meaningful (meaning making), which can guide the individual to self-reflect and change their priorities in life (Bartsch et al., 2014; Cova & Deonna, 2014; Landmann et al., 2019). Hence, kama muta may alleviate the negative perceptions of one's personal problems, as the individual senses his/her connection with society and reframes their state of mind. We think that unpacking the specific psychological mechanisms behind the relationship between kama muta and personal problem appraisals would be an area ripe for future research, and that the literature on the cognitive reappraisal processes facilitated by self-transcendent emotions (such as kama muta) may be a good start: kama muta may encourage individuals to cognitively re-appraise their negative situation in a positive light, possibly through finding meaning from adversity and rendering one’s personal problems as less important overall.
However, we also posited that meaning making processes would be enhanced by negative contexts. While we see some evidence for this assertion, in that perceived social impact of collective threat predicted kama muta experiences in Study 3, and that participants concerned about the ongoing natural disaster were more likely to have stronger experiences of kama muta in Study 2, these were based on largely subjective ratings. Direct experimental manipulations of threat had no significant influence on participants’ kama muta ratings through Studies 1 and 3, and the single paper metanalysis showed no significant effects. Our interpretation is that collective threat may exert a multifaceted, and complex effect on the individual, and may not directly affect kama muta ratings. Rather, it is the perceived societal impact of collective threat situations that predicts elicitation of kama muta. Such an explanation would be congruent with past theories on the role of adversity in human society and in forging community and building social capital (see Henrich & Muthukrishna, 2021). However, our experimental evidence remains weak, and as we cannot rule out the presence of confounding factors, the role of threat in heightening kama muta experiences yet remains inconclusive.
Limitations and Future Directions
While we experimentally varied kama muta induction in Study 1, these were linked to threat contexts, meaning that we did not directly examine if adjusting kama muta videos would result in reduced personal problem appraisals. For Studies 2 and 3, only threat was manipulated, and all participants watched the same kama muta eliciting stimulus. This limited scope of the videos used in the induction tasks is especially problematic for the current paper. For Studies 1 and 3, the content from the threat-inducing stimuli used were conceptually unrelated to the content of the kama muta-inducing stimuli. As a result, our manipulation examined generalized and decontextualized videos of intensification of communal sharing and prosocial behaviors that had little to do with the collective threat instances portrayed in our stimuli. Assuming that collective threat does indeed contribute to heightened kama muta experiences, the lack of contextual similarity across threat and kama muta-inducing scenarios in our experiments may have marred our experimental ability to comprehensively investigate Hypothesis 2.
Moreover, kama muta ratings tended to be low across our studies: kama muta is sensitive to the emotional intensity of the video (Kimura et al., 2019). On one hand, this is similar to previous research on feeling moved (Strick & van Soolingen, 2018), where low ratings are acceptable when focusing on differences between conditions. On the other hand, this may also suggest an issue with the experimental stimuli in inducing kama muta, in that a decontextualized one-minute video in an online experiment setting may not be enough to reliably induce high levels of kama muta.
Next, most of the mediation analyses used in this paper follow “measurement of mediation” designs, which cannot provide proof of causal processes (see Pirlott & MacKinnon, 2016). Furthermore, our mediation analyses also used difference scores between two timepoints as outcome variables in examining mediation models of threat and kama muta on PPA, and there are known issues with this methodology (see Edwards, 2001).
Finally, as kama muta was the focal emotion, awe, happiness, and gratitude were measured using single-item measures and may not be as fully captured. The current comparisons with other individual positive emotions are only preliminary, given that they were intended as controls in assessing the effect of kama muta and that there were no control conditions without the kama muta induction for Studies 2 and 3. Future research can utilize more direct manipulations of various positive emotions and include additional controls when focusing on differentiating the effects between these emotions.
In conclusion, our main finding is that kama muta may be functional in relieving the individual’s negative attitudes and appraisals from their personal problems, and to some extent, may be enhanced by collective threat with perceived societal impact. However, the exact relationships between threat and kama muta remains unclear. Nevertheless, our research presents a series of preliminary studies into the role of threat in facilitating individuals' capacity to help others through the emotion of kama muta, and we hope that our results and data can aid like-minded researchers with elucidating these effects. Future studies may examine the specific mechanisms by which kama muta decreases personal problem appraisals, and how (or if) collective threat enhances experiences of kama muta.