How self-efficacy beliefs in dealing with negative emotions are associated to negative affect and to life satisfaction across gender and age.

The present study examines the extent to which individuals' self-efficacy beliefs about their capacity to manage distinct emotions, such as anger, sadness, fear, shame and guilt, are associated with negative affect and life satisfaction in a Spanish population of diverse ages. The results attest to the validity of the Multidimensional Negative Emotions Self-Regulatory Efficacy Scale (MNESRES) and show that self-efficacy beliefs related to basic and self-conscious/moral emotions are associated differently with negative affect and life satisfaction. These findings corroborate previous findings from American and Italian populations, and they support the view that discrete emotions deserve distinct attention, either regarding their management or their association with individuals' well-being and adjustment.

Introduction

Social cognitive theorists view personality as a cognitive-affective system whose functioning depends upon the concerted action of psychological structures that have gradually been established to enable individuals to exert some control over their experiences, and the course of their lives [1]. Self-efficacy beliefs have been assigned a special role in attesting to human agency’s distinct capacities for self-reflection and self-regulation, and in promoting successful adaptation [2]. A large body of research reflects how the beliefs people hold about their capacity to cope efficaciously with arduous challenges and to face demanding situations, exert a pervasive influence over thought, motivation and action across different domains of functioning [1]. Originally, self-efficacy beliefs were considered as highly contextualized knowledge structures that affect appraisal, which in turn guide actions [2, 3]. This led researchers to focus on specific tasks and to pursue a multifaceted approach to study the various expressions of self-efficacy in diverse situations. However, it later became evident that the influence of self-efficacy beliefs extends beyond specific skills, tasks and situations, as individuals reflect upon the consequences of their actions and use their experience to broaden their understanding, and to control their feelings and thoughts. This raised the question as to how self-efficacy beliefs might affect individuals’ capacity to manage their emotions [4].

Regulation of emotions

Research on emotion management has flourished in recent decades, with a special focus on the regulation of emotions [5-7]. This phenomenon covers the expression and management of both positive and negative emotions [8], and adequate regulation has been associated with health and well-being [9, 10]. Individuals differ in how they think about emotions, and their beliefs about the controllability of emotions significantly affect how they acknowledge the need to regulate them by selecting and implementing adequate strategies, and monitoring the outcomes of their endeavours. Indeed, the control that individuals believe they can exert over the causes of their emotions and over the consequences of their reactions strongly influences the regulatory strategies they employ [11, 12]. It is likely that the more people who believe that emotions can be regulated, the more of them who would be willing to manage these emotions. As self-efficacy beliefs have been proven to exert a pervasive influence over thought, motivation and action across different functional domains, individuals’ beliefs about the control they can exert on their emotions are thought to be crucial to achieve effective regulation [2, 4, 5].

Measuring individuals’ perceived efficacy in the regulation of basic and self-conscious/moral negative emotions

Recent studies on perceived self-efficacy have broadened to consider how self-efficacy beliefs affect people’s capacity to manage their emotions, and their impact on individuals’ psychosocial functioning and well-being. Initially, this led to the development of the Regulatory Emotional Self-Efficacy scale (RESE), which has been used widely in several countries [13-15]. The scale conserves the distinction between positive and negative affect, and it includes five items that assess individuals’ perceived self-efficacy in expressing positive emotions like joy, enthusiasm and pride, as well as nine items that assess individuals’ perceived self-efficacy in managing anger/irritation and despondency/sadness. The RESE scale is becoming increasingly well validated, and there is a growing body of evidence that people’s beliefs about their capacity to express positive emotions and to control negative emotions exert a crucial influence on their successful development and social adaptation [14-16].

Most of the earlier research on affective self-regulatory efficacy addressed basic emotions, and their relevance to health and social functioning [13, 17–19]. Focusing on negative emotions has led to the development of the Multidimensional Negative Emotions Self-Regulatory Efficacy Scale (MNESRES) [19], a new measure that includes five subscales designed to assess self-efficacy beliefs in relation to the management of anger/irritation, sadness, fear, shame and guilt (SE_Anger, SE_Sadness, SE_Fear, SE_Shame and SE_Guilt, respectively). In validating MNESRES it was found that five distinct mastery beliefs were uniquely associated with social adjustment and well-being [19]. However, confirmatory factor analyses showed that the scales could be traced to either five correlated factors or to two second-order factors, which in turn could be traced to a third-order factor. Specifically, while SE_Anger, SE_Sadness and SE_Fear could be traced to a common second-order factor that may be considered to be related to basic negative emotions, SE_Shame and SE_Guilt could be traced to another second-order factor thought to be related to self-conscious moral emotions. Both models fitted reasonably well with the data, but the one that best fitted the data in Italy and the U.S. was a third model, one that posited a third-order factor above (1) a second-order factor that included SE_Anger and SE_Sadness (2), SE_Fear as a first-order factor by itself, and (3) another second-order factor that included SE_Shame and SE_Guilt [19].

Self-efficacy beliefs related to basic and self-conscious/moral emotions

People display basic emotions like anger, sadness and fear from the early stages of life. By contrast, self-conscious/moral emotions like guilt and shame do not appear until self-structures and processes develop, enabling individuals to reflect on their experiences, and to acknowledge the discrepancies between their behaviour and their values [20]. The basic emotions anger/irritation, sadness and fear are thought to have evolved over time as part of humans’ basic equipment to cope with events such as loss, failure and threat to survival [21, 22]. In general, they occur with moderate engagement of cognitive processes, and they are associated with the triggering of rapid and sometimes automatic responses [23, 24]. The regulation of these emotions has received much attention due to the adverse consequences arising from their dysregulation in terms of health and social integration [25]. Anger has been mostly associated with externalizing behavioural problems [26, 27], whereas sadness and fear have been associated with internalizing problems, and sometimes with externalizing problems as well [28].

To date, fewer studies have explicitly addressed the regulation of self-conscious/moral emotions like shame and guilt, although one should note the growing interest in studying these emotions given their significant influence on social behaviour [29, 30]. Shame and guilt engage advanced cognitive processes, resting upon the development of internal self-referent structures and processes, and largely depending upon social and cultural norms. They reflect individuals’ internalization of moral obligations or socially expected conduct, the violation of which carries feelings of discontent, and interferes with self-acceptance and respect [31]. In addition, like basic emotions shame and guilt have been shown to affect mental health and well-being, although to a different degree [32-35].

Whereas each emotion can be viewed as the result of a unique organization of affect, cognition and behaviour under given situations and cultural constraints [36], it is likely that self-efficacy beliefs affect the regulation of both basic and self-conscious/moral emotions, and that they influence health and well-being, albeit to different extents. Indeed, negative basic emotions like anger, fear and sadness have a pervasive impact over individuals’ life, and their regulation has a primary influence on adaptation when compared to self-conscious/moral emotions like guilt and shame. In terms of the regulation of basic emotions vs. self-conscious/moral emotions, our reasoning was based on two considerations: 1) Basic emotions are fundamental for the individual’s survival and as such, they are largely prewired in our brain, manifesting themselves early in development and with similar features across cultures. By contrast, self-conscious/moral emotions appear later in development, and they rely upon the appearance of cognitive structures that underlie the formation of a self-system to which they are instrumental and that maintains behavior within the limits of social and personal standards. Moreover, these emotions are largely conditional, and they are shaped by social experiences and cultural references [37-39]. 2) Basic negative emotions can be viewed as precursors of self-conscious/moral emotions on which reactions to harm and pleasure are founded. While negative basic emotions may represent survival responses to external harm or danger in the outside world, self-conscious/moral emotions underlie self-respect and they respond to violations of internal oughts and beliefs [40]. It is likely that the proper regulation of basic emotions requires capacities for self-reflection, learning and practice that establish conditions for the subsequent development and management of self-conscious/moral emotions. Thus, one would expect that self-efficacy beliefs related to basic emotions would ultimately be more effective in protecting health and fostering well-being than self-efficacy beliefs related to self-conscious/moral emotions.

Aims and hypotheses

The goal of this study was to further validate the MNESRES by examining how individuals’ beliefs that they are capable of handling negative emotions like anger, sadness, fear, shame and guilt (i.e.: the dimensions of the MNESRES) are related to negative affect and life satisfaction in an adult Spanish population. The study was carried out on a large sample that included males and females of different ages. The size and variation in the sample allowed the evaluation of the extent to which self-efficacy beliefs in terms of the individual’s capacity to manage distinct negative emotions is associated with negative feelings and life satisfaction, across ages and gender.

Based on earlier findings, we first set out to examine the extent to which previous factorial models could be corroborated and then, to ascertain the invariance of the distinct scales across gender and age-groups [19]. In principle, we expected that items designed to assess each scale (SE_Anger, SE_Sadness, SE_Fear, SE_Shame, and SE_Guilt) would rely on separate but related first-order factors. Yet we wondered whether the model positing five oblique factors would fit the data better than alternative models tracing beliefs of mastering negative emotions to a hierarchy of higher-order factors. Likewise, we expected a significant similarity in the factorial structure of the MNESRES scale across age and gender. In accordance with previous findings [41], we expected higher values for men’s self-efficacy beliefs in managing negative emotions than for women. Indeed, women have shown a lower perceived efficacy in managing negative affect than men [13, 42]. Yet women’s perceived efficacy in managing negative emotions has shown to improve with age [41, 42], whereas it has shown to decline in men.

For both men and women, perceived efficacy in managing negative emotions has resulted positively associated to life satisfaction [43] and successful adaptation [37], and negatively associated with depressive feelings, shyness and delinquency [13]. Thus, we expected that perceived self-efficacy in managing negative emotions would result in a negative association with negative affect and in a positive association with life satisfaction. However, in accordance with the aforementioned considerations regarding the impact that basic emotions have on social adjustment and well-being, we expected that self-efficacy beliefs related to the management of basic emotions like anger, sadness and fear would be more strongly associated with both negative affect and life satisfaction than self-efficacy beliefs related to the management of self-conscious/moral emotions.

Materials and methods

Participants and procedures

Data were collected between 2014 and 2017 from a total of 1695 participants (49% males), who ranged from 19 to 94 years of age (M = 43.34, SD = 13.48: see Table 1). The participants were recruited at the Open University in Madrid (UDIMA), and from the relatives and friends of undergraduate Psychology students who received course credits for recruiting them. All the participants were born in Spain, were considered to belong to the middle class, were moderately-to-highly educated, and the majority of them lived in the Municipal area of Madrid.

Table 1

Distribution of the interaction between age and gender in the Spanish sample.

Age group1	N Men	N Women	N Total	Mean	SD
Younger	390	404	794	32.80	5.04
Middle	428	246	674	46.42	4.89
Older	97	130	227	68.87	8.44
Total	915	780	1695	43.34	13.48

1: Younger = 20–40 years old; Middle = 41–60 years old; Older 61–94 years old.

Forward-and back-translation was used to ensure the linguistic and conceptual equivalence of the measures across languages [44]. The Spanish versions of the questionnaires were administered individually to all the participants after obtaining their informed consent. The participants were informed that their responses would be associated with number codes by the research staff and that full confidentiality would be guaranteed throughout all stages of the study. The study and the procedures involved were approved by the Institutional Ethical Research Review Board of the Open University in Madrid (UDIMA), and the data were analysed anonymously.

Measurement scales

Multidimensional Negative Emotions Self-Regulatory Efficacy Scale (MNESRES) [19]

Participants rated their capacity to manage different emotions on a scale of 1–5 (1 = not at all well; 5 = very well) using 3 items for each: anger/irritation (e.g., “How well can you avoid flying off the handle when you get angry?”), despondency/sadness (e.g., “How well can you keep from getting dejected when you are lonely?”), shame/embarrassment (e.g., “How well can you overcome shame when your weaknesses become evident in front of others?”), fear (e.g., “How well can you stay calm in situations in which many others would be fearful?”) and guilt (e.g., “How well can you control feelings of guilt after not fulfilling important commitments and obligations?”). The formulation, selection and phrasing of the challenges presented in each item, carrying different degrees of difficulty, followed Bandura’s guidelines [45].

Negative affect

Participants rated the degree to which they generally experience the ten negative emotional states in the Positive and Negative Affect Scale (PANAS) [46], on a scale of 1–5 (1 = not at all; 5 = very much), with a Cronbach’s alpha coefficient (α) = .83 for negative affect. For the purposes of this research, we used the Spanish version of this scale [47], and the emotions referred to as “distressed”, “upset”, afraid”, “scared”, “nervous”, “jittery”, “hostile”, “guilty”, “irritable” and “ashamed” were the ten items included in this score.

Life satisfaction

Life satisfaction was measured using the 5 items of the Satisfaction with Life Scale (SWLS) [48]. The participants rated the extent to which they felt generally satisfied with life (e.g. “I am satisfied with my life”) using a 7 point Likert scale ranging from 1 (= strongly disagree) to 7 (= strongly agree), and with α = .89. In this study we used the Spanish version of SWLS [49].

Analytic approach

In order to investigate the factorial structure of the MNESRES, in accordance with previous findings [19], we tested four Confirmatory Factor Models (CFA) using a maximum likelihood estimation of the parameters. Model 1, the five types of self-efficacy were considered as separate constructs correlated on the first level. In Model 2, the five types of self-efficacy were modelled as expressions of a second-order latent factor reflecting a general capability to manage negative emotions (SE_Negative Affect). In Model 3, SE_Anger, SE_Sadness and SE_Fear were modelled as expressions of a second-order latent factor reflecting the capability to manage basic negative emotions, whereas SE_Shame and SE_Guilt were modelled as expressions of a different latent second-order factor reflecting the capability to manage self-conscious/moral emotions. In Model 4, SE_Anger and SE_Sadness were grouped as expressions of a second-order latent factor, SE_Fear was left by itself, and SE_Shame and SE_Guilt were grouped together as another second-order factor. In both Model 3 and Model 4, we also included a latent third-order factor of SE_Negative Affect on which the second-order factors loaded. These models were the most competitive of those tested in a previous study, albeit not in a Spanish context [19]. We focused on the indices of the model’s fitness that were least sensitive to sample size, given that obtaining a non-significant χ2 value becomes increasingly unlikely with larger sample sizes [50]. For each model we calculated the Comparative Fit Index (CFI), the Root-Mean-Square Error of Approximation (RMSEA) with its associated Confidence Interval and p value, the Standardized Root-Mean-Square Residual (SRMR), and the Akaike Information Criterion (AIC), a useful index when comparing models that are not nested [51], accepting CFI values greater than .95 [46], RMSEA values lower than .07 [52] and SRMR values lower than .08 [46]. For the RMSEA, a non-significant p-value means the model we hypothesized approximates well to the population. Mplus 7 [53] was used for CFAs.

To test both the gender and age invariance of the MNESRES, we adopted a model-fitting process as indicated elsewhere [54]. We tested three models to examine the gender invariance and three models to test age invariance: configural invariance, the same pattern of fixed and free factor loadings was specified across gender and age-groups, respectively; metric invariance, the same factor loadings for items were specified across gender and age-groups; and scalar invariance, the same intercepts of like item regressions on the latent variables were specified across gender and age-groups. The most frequent additional tests performed were typically those of partial invariance at each step and modification indices for each step were used to refine the structure models [55]. Each model of invariance was nested in the previous model and involved added constraints that built upon previous constraints. Specifically, the model for metric invariance was nested within that for configural invariance and the model for scalar invariance was nested within the one for metric invariance. Difference in the Chi-squared tests were assessed to compare the nested models, adopting a cut-off of p < .01 given that obtaining a significant χ2 value becomes increasingly likely with larger sample sizes [46]. To explore gender and age differences in the dimensions of the MNESRES, we examined gender (male and female) by age (younger, middle and older adulthood) with MANOVA, using the dimensions of the MNESRES as dependent variables. Finally, in order to explore the construct validity of the MNESRES, we examined the correlations among the dimensions of MNESRES, negative affect and life satisfaction. Moreover, having verified the significant associations among the variables, we conducted two hierarchical regression analyses to examine the associations of the dimensions of the MNESRES with negative affect and life satisfaction, respectively. Specifically, the principal effects included age and gender (Step 1), all the dimensions of MNESRES (Step 2), and the interaction of age or gender for each dimension of MNESRES (Step 3). Lower-order and interactive terms were mean-centred in order to facilitate a correct interpretation of the lower-order terms and to decrease non-essential multicollinearity [56]. MANOVA, correlation and regression analyses were performed with SPSS 18.0 software (SPSS Inc.).

Results

Confirmatory factor analyses and reliabilities

The different indices of fit were collected for the four models evaluated: Model 1 χ(80) = 439,196 (p <001), CFI = .94, RMSEA = .05 (.05 - .06), SRMR = .04, AIC = 67413.04; Model 2 χ2(85) = 576,814 (p < .001), CFI = .91, RMSEA = .06 (.05 - .06), SRMR = .05, AIC = 67540.66; Model 3 χ2(86) = 514,599 (p < .001), CFI = .92, RMSEA = .05 (.05 - .06), SRMR = .04, AIC = 67476.44; Model 4 χ2 (83) = 467,502 (p < .001), CFI = .93, RMSEA = .05 (.05 - .06), SRMR = .04, AIC = 67435.35. All models provided a discrete fit to the data according to each index, with Model 1 offering a better fit than the other tested Models according to multiple indices, including the lowest AIC. Although the CFI did not pass the cut-off of .95, it was very close to that threshold, and all the other fit indices were indicative of good fit [35]. Thus, Model 1 appeared to be the model that best represented the structure of the MNESRES in a Spanish context (see Fig 1 for the path diagram of CFA). The alpha coefficients for the five sub-scales designed to assess self-efficacy beliefs regarding anger/irritation (SE_Anger), despondency/sadness (SE_Sadness), fear (SE_Fear), shame/embarrassment (SE_Shame), and guilt (SE_Guilt) management were .55, .54, .71, .74, .67, respectively.

Fig 1

Path diagram of the best factorial solution of the MNESRES in a Spanish context.

The labels in the first order factors refer to the five subscales designed to assess self-efficacy beliefs regarding managing anger/irritation, despondency/sadness, fear, shame/embarrassment, and guilt (SE_AI, SE_DS, SE_F, SE_SE, SE_G, respectively). e1–e15 represent error terms. Standardized factor loadings are shown on the straight arrows. All the paths are significant with p-value < .001.

Gender invariance

Multi-group CFAs confirmed that the model with five oblique factors was replicated consistently in men and women. Specifically, the change in the overall chi-squared value between the configural invariance model (χ(159, N = 915,780) = 591.14, (p < .001), CFI = .92, RMSEA = .06 (.05 - .06), SRMR = .04) and the metric invariance model (χ (174, N = 915 780 = 608.37 (p < .001), CFI = .92, RMSEA = .05 (.05 - .06), SRMR = .05) was not significant (- Δχ (15) = 17.23, p = .31). Likewise, the change in the overall chi-squared value between the metric invariance model and the scalar invariance (Δχ(10) = 12.74, p = .24) was not significant, suggesting that both the metric and scalar invariance persisted across genders. The fit indices for the scalar invariance were: χ(184; N = 915,780) = 621.11, CFI = .92, RMSEA = .05 (.04 - .06), SRMR = .05 (see the first two columns of Table 2 for the unstandardized values of loadings and the intercepts in the five-oblique factor model for men and women). It is important to note that the modification index of all the models evaluated suggested an estimation of covariation between two item errors for males. Our rationale in freely estimating this error covariance was that in the male group, one covariant was estimated freely between item errors 8 (SE_Fear; “Overcome feelings of panic and keeping a clear mind in the presence of very dangerous situations”) and 11 (SE_Shame; “Contain your shame after having made a fool of yourself in front of many people”), both referring to the possibility that events that threaten one’s social status can be considered dangerous. This interpretation is in line with a previous study [57] in which feelings of inferiority and threat are fundamental dimensions that underlie behaviour in a social rank system, especially in men. In summary, the test of scalar invariance provided evidence for full scalar variance for all five sub-scales of MNESRES between males and females.

Table 2

Unstandardized values of factor loadings and their intercepts in the five-oblique factor model through gender and age invariance of the MNESRES in a Spanish context.

		Gender Invariance		Age Invariance
		Factor Loadings	Intercepts	Factor Loadings			Intercepts
	Items	Men-Women	Men-Women	Young	Middle	Late	Young	Middle	Late
SE_Anger	1	.55	3.11	.59			3.11
	2	.59	3.09	.61			3.09
	3	.53	3.26	.54			3.25
SE_Sadness	4	.56	3.29	.59			3.24
	5	.60	3.49	.60			3.43
	6	.44	3.22	.48			3.08 b	3.27	3.27
SE_Fear	7	.61	3.29	.60			3.22
	8	.76	3.26	.76			3.17
	9	.64	3.39	.61	.61	.74a	3.30	3.30	3.42 b
SE_Shame	10	.69	3.33	.69			3.26
	11	.70	3.04	.70			2.98
	12	.69	3.23	.69			3.16
SE_Guilt	13	.65	3.09	.58a	.71	.71	3.00 b	2.99	2.99
	14	.65	2.95	.66			2.87
	15	.57	2.81	.55	.55	.73a	2.72	2.72	2.74 b

Note. All factor loadings and intercepts were significant at p < .001

a factor loading differed significantly from the other two in the same column under Factor Loadings

b the intercept differed significantly from the other two in the same column under Intercepts.

Age invariance

To examine age invariance in the five subscales of the MNESRES, we used the age groups reported in Table 1. There was a significant mean change in the overall χ between the configural invariance model (χ (236, N = 794,674,227) = 589.17 (p < .001), CFI = .94, RMSEA = .05 (.05 - .06), SRMR = .04), and the metric invariance model (Δχ(28) = 53.25, p = .003), suggesting that full metric invariance does not hold good across the three age-groups. We found that the following equality constraints needed to be lifted for the factor loading of item 13 (SE_Guilt; “Control feelings of guilt after not fulfilling important commitments and obligations”) in the younger adults, and item 9 (SE_Fear; “Stay calm in situations in which many others would be fearful) and item 15 (SE_Guilt; “Contain feelings of guilt after having violated very important personal moral principles”) in the older adult group. After releasing these constraints on factor loading scores, the mean change in overall χ between the configural invariance model and the partial metric invariance model (χ (261; N = 794,674,227) = 625.29 (p < .001), CFI = .94, RMSEA = .05 (.05–06), SRMR = .06, SRMR = .06) was no longer significant (Δχ(25) = 36.12, p = .07), suggesting that partial metric invariance held across the three age-groups. We repeated the procedure several times to search for items that were not invariant across the three age-groups based on the modification indices, eventually obtaining a model in which the chi-squared difference between the partial metric invariance model and the partial scalar invariance model was no longer significant (Δχ(16) = 19.39, p = .25). We found that the following equality constraints had to be lifted for the intercepts of item 6 (SE_Sadness) and item 13 (SE_Guilt) in the younger adults, and intercepts of item 9 (SE_Fear) and item 15 (SE_Guilt) in the late adulthood group. In summary, the test of scalar invariance provided evidence for full scalar variance for the factors SE_Anger and SE_Shame, whereas it provided evidence for partial scalar invariance for the SE_Sadness, SE_Fear, and SE_Guilt factors (see the third, fourth, and fifth columns of Table 2 for the unstandardized values of loadings and intercepts in the five-oblique factor model for younger, middle and older adults).

It is important to note that the modification indices estimate the covariation between item errors for some of the groups in all the models evaluated. We estimated five covariations between errors when we had substantive reasons to support such covariation based on the similarities in item content. Our rationale for freely estimating these five error covariances was that one covariation was freely estimated between item errors 4 and 6 in younger adults (both in SE_Sadness), these referring to the possibility that being lonely has something in common with being exposed to strong criticism in the young adult phase of life [58]. Covariation between item errors 5 (SE_Sadness) and 8 (SE_Fear) reflected that both refer to high-risk situations, and that between item errors 8 (SE_Fear) and 11 (SE_Shame) reflected that the possibility that events which threaten ones social status could be regarded as dangerous situations by the middle adulthood group [59]. Covariation between item errors 1 and 2 (both in SE_Anger) refer to a possible attribution bias related to experiencing wrong doing when others give you hard time in late adulthood [60]. Similarly, there was covariation between item errors 8 and 9 (both in SE_Fear), which are both about dangerous and fearful scenarios (further details about the content of the items in Spanish are available from the corresponding author.

Gender and age differences

The overall MANOVA suggested that there are gender [F(5,1681) = 7.5186, p < .001] and age [F(10,3364) = 2.287, p = .01] differences for all five of the MNESRES sub-scales, although the gender by age interaction was not significant [F(10,3362) = 1.469, p = .14]. However, the partial η2 indicated there were small effects in terms of effect sizes [61] for both the gender (partial η2 = .02) and age (partial η2 = .01) differences. Given the significant multivariate results, univariate ANOVAs were used to examine each of the MNESRES sub-scales (see Table 3), in addition to the estimates of the partial η2. Males reported higher self-efficacy beliefs than females when dealing with all the negative emotions examined across all age-groups except for SE_Anger.

Table 3

Gender and age differences in the five subscales of the MNESRES in a Spanish context.

	All		Men			Women			Younger			Middle			Older			Gender effects	η2
	M	SD	M	SD	M		SD	M		SD	M		SD	M		SD
SE_Anger	3.16	0.76	3.16	0.77	3.15		0.75	3.16		0.75	3.18		0.76	3.07		0.76	F(1,1688) = .612, p = .43		.00
SE_Sadness	3.25	0.72	3.33a	0.69	3.16b		0.75	3.25		0.70	3.26		0.71	3.26		0.82	F(1,1688) = 16.41, p < .001		.01
SE_Fear	3.26	0.80	3.31a	0.80	3.20b		0.79	3.23		0.75	3.27		0.81	3.33		0.94	F(1,1688) = 11.02, p = .001		.01
SE_Shame	3.14	0.81	3.20a	0.79	3.07b		0.82	3.13		0.80	3.18		0.80	3.05		0.85	F(1,1688) = 7.30, p = .007		.01
SE_Guilt	2.89	0.76	2.95a	0.74	2.81b		0.77	2.87		0.74	2.90		0.76	2.94		0.81	F(1,1688) = 8.70, p = .003		.01

Note. Different letters indicate significant gender differences. None of the univariate age and interactions effects were significant and thus, they were not reported in this Table.

Relationships of the MNESRES with negative affect and life satisfaction

Correlation

In order to examine the construct validity of the five MNESRES sub-scales, the correlations among the sub-scales were examined, as were the correlations of each sub-scale with negative affect and life satisfaction (Table 4). All correlations were significant at p < .01 and moderate to strong positive correlations emerged among the five sub-scales. In particular, stronger correlations emerged between SE_Anger and SE_Sadness, and between SE_Shame and SE_Guilt, that is among self-efficacy beliefs related to basic emotions and among self-efficacy beliefs related to self-conscious/moral emotions. Low to moderate negative correlations emerged between all five sub-scales and negative affect and in particular, a stronger correlation was evident between SE_Sadness and negative affect. Low to moderate positive correlations emerged between all five sub-scales and life satisfaction, with a particularly strong correlation between SE_Fear and life satisfaction. The correlation between the two outcomes examined was moderate.

Table 4

Correlations among the five sub-scales of the MNESRES, negative affect and life satisfaction.

		(1)	(2)	(3)	(4)	(5)	(6)	(7)
SE_Anger (1)	r	1
SE_Sadness (2)	r	.465**	1
SE_Fear (3)	r	.320**	.362**	1
SE_Shame (4)	r	.303**	.399**	.389**	1
SE_Guilt (5)	r	.274**	.319**	.214**	.447**	1
Negative Affect (6)	r	-.192**	-.238**	-.219**	-.129**	-.062*	1
Life Satisfaction (7)	r	.194**	.267**	.297**	.179**	.083**	-.295**	1

Note. p < .01**.

Regression analyses

The results of the hierarchical regression analyses focused on negative affect as the dependent variable (Table 5, left). In the first step, age and sex did not account significantly for the variance in negative affect, and in the second step, the addition of the five self-efficacy beliefs of MNESRES to the regression equation significantly improved the explanation of the variance. However, only self-efficacy beliefs relating to basic emotions (i.e.: SE_Anger, SE_Sadness and SE_Fear) were associated with significant coefficients. In this latter step, self-efficacy beliefs accounted for 8% of the variance in negative affect. In the third step, the subsequent addition of the interaction terms did not account significantly for the variance in negative affect.

Table 5

Hierarchical regressions analyses (HRA) of sex, age and each of the MNESRES dimensions on negative affect and life satisfaction, respectively.

DV:	Negative Affect			Life satisfaction
First set of HRA	B a	(SE)	ΔR2	B a	(SE)	ΔR2
Step 1			.003			.01**
Sex	.04	.03		.01	.04
Age	-.05	.001		.08**	.00
Step 2			.08**			.12**
SE_Anger	-.09**	.02		.05	.03
SE_Sadness	-.16**	.02		.17**	.03
SE_Fear	-.14**	.02		.21**	.03
SE_Shame	-.01	.02		.04	.03
SE_Guilt	.05	.02		-.05	.03
Step 3			.01			.02**
Sex X age	.001	.002		-.02	.003
SE_Anger X Sex	.03	.04		-.06	.06
SE_Sadness X Sex	-.03	.04		.04	.07
SE_Fear X Sex	.07	.04		-.12**	.06
SE_Shame X Sex	.03	.04		.14**	.06
SE_Guilt X Sex	-.02	.04		-.09	.06
SE_Aner X Age	-.06	.001		.05	.00
SE_Sadness X Age	-.03	.002		.03	.00
SE_Fear X Age	.05	.001		-.05	.00
SE_Shame X Age	-.01	.002		-.05	.00
SE_Guilt X Age	.04	.001		.06	.00
Adjusted R2	.09			.14
F value	9.95**			16.46**

Note.

a Standardized regression coefficients.

** p < .01. Sex was coded as 0 = men and 1 = women.

The hierarchical regression analyses focusing on life satisfaction as the dependent variable were also assessed (Table 5, right) and in the first step, only age significantly and positively predicted life satisfaction. In particular, older people scored higher in life satisfaction than younger ones. In the second step, the addition of the five MNESRES self-efficacy beliefs to the regression equation significantly improved the variance explained. However, significant coefficients were only obtained for the self-efficacy beliefs relating to sadness and fear (i.e.: SE_Sadness and SE_Fear), with self-efficacy beliefs accounting for 12% of the variance in life satisfaction in this step. In the third step, the subsequent addition of the interaction terms only moderately accounted for the variance in life satisfaction. In particular, the interaction of SE_Fear and SE_Shame with sex was associated with life satisfaction. We plotted simple slopes of the aforementioned self-efficacy beliefs using high and low values (above and below the median) and accordingly, we found that the association between life satisfaction and SE_Fear was stronger for men than for women (Fig 2A). In addition, the association between life satisfaction and SE_Shame was stronger for women than for men (Fig 2B).

Fig 2

A: The effect of the interaction of sex with self-efficacy about fear regulation on life satisfaction. B: The effect of the interaction of sex with self-efficacy about shame/embarrassment regulation on life satisfaction.

Discussion

The aim of this study was to examine the degree to which self-efficacy beliefs relating to the management of anger, sadness, fear, shame and guilt are associated with negative affect and life satisfaction, in a population of Spanish adults. To our knowledge, this is the first study conducted on a Spanish population using the MNESRES on a population that breaches at least three generations, young, middle and older adults. CFA showed that the model that best represented the MNESRES structure in accordance with standard criteria is that which proposes five separate but correlated factors. These results differ from previous findings in US and Italian populations of young adults, showing that the best model was that involving a hierarchical structure with a third order latent factor overarching SE_Fear, and two second order factors including SE_Anger and SE_Sadness or SE_Shame and SE_Guilt, respectively. In the present study, we focus on a larger and more diverse sample than that reported previously [19], and the five-oblique-factor solution was that which best fitted the data. Yet one should not forget that alternative models, including that with the best fit previously, were not far from standard criteria of acceptance. In particular, the model that fitted best the data obtained here was one that had previously shown an acceptable fit. Albeit with some caution, it might be concluded that the five-oblique-factor solution is in accordance with the underlying specificity of each emotion, implying that it would be recommendable to focus on each emotion irrespective of its source, manifestations and means of regulation. This recommendation may also be worthy of consideration in terms of self-efficacy beliefs, as corroborated by the five factor solution of the confirmatory factor analysis. However, this does not fully exclude attempting to trace one common factor to all five emotional self-efficacy beliefs, since the statistical divide between the two solutions is minimal. Special attention is merited regarding the distinct assessment of the examined emotions, and their impact on the individual’s well-being and social adjustment. As stated in the premises, we feel that self-efficacy beliefs may be generalized within and across domains as the expression of a common self-system that oversees the relationships of the individual with oneself and the environment. This leads to the view that emotions define a domain where one may examine how the impact of self-efficacy beliefs can be generalized across situations and tasks. Yet the extent to which the regulation of emotions represent a domain of functioning that relies upon the same mental structures and processes requires further study. In this regard, previous findings invite us to address positive and negative emotions as expressions of different systems whose regulation depends upon different self-efficacy beliefs. Thus, we are resistant to embrace the idea of a general emotional self-efficacy, in accordance with previous criticisms of the idea of general efficacy [13].

In addressing negative emotions, we did not exclude that they could be traced to a hierarchical model. However, the present and previous findings invite us to be cautious about the extent to which the regulation of negative emotions can be traced to a common sense of mastery. Currently, the structure of self-efficacy beliefs assessed by the MNESRES parallels the common distinction between basic emotions like anger, sadness and fear, and that of self-conscious moral emotions like shame and guilt. The results of the CFA indicate they are correlated no less than they are distinctive. Thus, one may surmise that the common distinction between basic and self-conscious/moral emotions also holds well for self-efficacy beliefs in terms of the management of those emotions. It is also likely that the regulation and the beliefs of mastery of basic emotions and of self-conscious/moral emotions rely on mechanisms and strategies that differ from one to the other.

Our findings show that self-efficacy beliefs related to basic emotions are more strongly associated with negative affect and life satisfaction than self-efficacy beliefs related to self-conscious/moral emotions. This is consistent with our expectations positing the beliefs people hold about their capacity to manage negative basic emotions as better predictors of well-being than people’s beliefs about their capacity to manage self-conscious/moral emotions. Thus, one might argue that self-efficacy beliefs related to basic emotions may significantly affect and, even subsume, self-efficacy beliefs related to self-conscious/moral emotions. However, further research is required to clarify this hypothesis.

In accordance with the findings from an Italian sample [38], men’s self-efficacy beliefs in dealing with all the negative emotions examined were higher than women’s corresponding beliefs. Although an overall significant age effect emerged, none of the specific MNESRES factors were seen to have age differences in the three groups examined. In addition, no age by gender effect emerged for such factors. In other words, self-efficacy beliefs in dealing with basic and self-conscious/moral emotions were similar among young, middle and older adults. In the Spanish context, and in agreement with studies on adult cohorts from Italy [19], Germany [15] and the United States [39], the beliefs people hold about their ability to manage negative emotions were generally negatively associated with negative affect and positively with life satisfaction. Although correlations do not allow us to establish a causal relationship, it is reasonable to surmise that being confident of one’s own capacity to manage negative emotions contributes to enhance life satisfaction and to counteract negative feelings. This does not rule out that satisfaction with one’s own life may contribute to being confident of one’s own capacity to manage emotions. However, it is likely that the impact of self-efficacy beliefs on positive feelings and satisfaction is reciprocal. In this regard, the social cognitive theory provides guidance as how to enhance self-efficacy beliefs, which in turn can be instrumental in enhancing both positive feelings and life satisfaction.

Finally, the relationships that emerged from the regression analyses between self-efficacy beliefs related to different emotions and negative affect and life satisfaction are worthy of consideration. In general, perceiving themselves capable of managing sadness and fear proved to have a higher impact on sustaining life satisfaction and counteracting negative affect, than perceiving themselves capable of managing anger. Children probably learn how to refrain from anger earlier through socialization practices that sustain empathic concern and that lead them to avoid reactions that may hurt others. By contrast, sadness and fear are insidious experiences that are difficult to avoid and hard to handle throughout life. In this regard, the major impact that self-efficacy beliefs associated with the management of fear has on life satisfaction and negative affect in males rather than females deserves special attention.

The results obtained represent an important contribution to the current literature on emotional regulation as they provide new information regarding the role of self-efficacy beliefs in dealing with basic and with self-conscious/moral emotions. The data corroborate the distinction of self-efficacy beliefs among discrete emotions. Likewise, the data corroborate the validity of the measures that have been developed to study the contribution of self-efficacy beliefs to emotions’ regulation and their impact on different aspects of well-being, like negative affect and life satisfaction. The fact that the main effects of self-efficacy beliefs hold well across gender and age-groups is particularly noteworthy.

That said, we are aware that there are some limitations to our study. These mostly concern the overall representation, the generality of the findings and the direction of the influences among the correlated variables.

The alpha values of the MNESRES subscales for anger and sadness are far from optimal indices and this represents an important limitation. We are inclined to believe that this is due to the limited number of items, although this issue should be addressed in future studies by either enlarging the number of items or through their reformulation.

It is important to note that our sample was a convenience sample and therefore, its generalization is somewhat limited (e.g., the age groups do not match perfectly the age distribution in Spain at the time when the study was carried out). This fact may also reflect the potential dependence in terms of the social relationships among the participants. However, the fact that the major findings were replicated across both sex and gender, and that they corroborate the data from previous studies, are witness to their robustness.

As stated above, the correlations resulting from the cross-sectional design only allow us to guess about causal influences. It is likely, however, that self-efficacy beliefs dealing with emotions have an impact on people’s chronic moods and life satisfaction by setting favourable conditions to learn from experience and to manage them at best.

As effective emotional regulation involves adapting strategies and deploying them effectively, studying the kinds of strategies commonly implemented by self-efficacious people to manage distinct negative emotions is an issue worthy of further study. Since previous study have shown that self-efficacious people tend to use more reappraisal than suppression strategies [15] future studies might examine whether the same patterns occur when dealing with fear, shame, embarrassment and guilt, as assessed by the MNESRES.

Finally, the results presented here highlight the importance of self-efficacy beliefs in terms of the regulation of discrete negative emotions in middle and late adulthood, in addition to what has already been established with young adults in previous studies [14, 19]. These results may set the basis to further investigate the degree to which the strengthening of emotional self-efficacy beliefs can represent a viable strategy to equip people to face major life transitions and successful aging.

Data regarding Tables 1, 2, 3, 4 and 5.

(XLSX)

Click here for additional data file.

Data regarding Fig 1.

(XLSX)

Click here for additional data file.

Data corresponding to Fig 2.

(XLSX)

Click here for additional data file.

(PPTX)

Click here for additional data file.

15 Apr 2020

PONE-D-20-00499

How Mastery Beliefs in Dealing with Negative Emotions Can Counteract Negative Affect and Promote Life Satisfaction Across Gender and Age

PLOS ONE

Dear Mrs Caprara,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

It has been particularly difficult to find reviewers, which may account for the unusual long time between the submission and the decision.

We would appreciate receiving your revised manuscript by May 30 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Delphine Grynberg, PhD

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements:

1.    Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.plosone.org/attachments/PLOSOne_formatting_sample_main_body.pdf and http://www.plosone.org/attachments/PLOSOne_formatting_sample_title_authors_affiliations.pdf

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: I Don't Know

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The manuscript deals with the question of the factorial structure and construct validity of the Multidimensional Negative Emotions Self-Regulatory Efficacy Scale (MNESRES), which measures self-efficacy beliefs in regulating discrete negative emotions. Overall, the manuscript seems scientifically rigorous to me. However, I have some questions and comments.

(1) Clear presentation of the aims, methods, & results

(a) Throughout the manuscript, the writing was not always very clear / concise.

(b) The introduction was too broad for me. The MNESRES, which should be front and center, is only mentioned at the end of the second page. I would have expected an introduction that focusses more on why this particular study is of importance. I would also have liked to have more information on why self-regulatory efficacy may be comparable across age groups and gender. Also, the rationale for choosing the two outcomes (negative affect & satisfaction with life) was not clear to me.

(c) Given that the authors stress the importance of shame and guilt, it seemed somewhat surprising to me that they expected stronger associations with negative affect (NA) and satisfaction with life (SWL) for the basic emotions.

(2) Methods

(a) Sample – The sample consisted of relatives and friends of undergraduate Psychology students. The authors may want to discuss that this is not ideal, given that participants were not independent from one another (e.g., couples, relatives, friends, etc.). Also, middle-aged men were oversampled and the distribution of age in the older age group seems skewed.

(b) MNESRES – I would help my understanding of the concept of self-regulatory efficacy to have a table with the MNESRES items or at least example items for each dimension.

(c) Negative affect – I was wondering why only NA was reported, but not PA. Given that the authors also assessed SWL, it would have made sense to report on subjective well-being (with the facets NA and PA for affective well-being, and SWL as the cognitive facet; see Diener). Alternatively, it would have been interesting to have measures of discrete emotions matched to the dimensions of the MNESRES as outcomes.

(3) Analytic approach

(a) It was unclear to me why the three proposed models were tested. How were these models pre-selected?

(b) Some authors have argued for CFI values to be close to .95 rather than .90 (Hu & Bentler, 1999). To me, .90 seems like a rather low bar – maybe the authors want to discuss this.

(c) It was not clear to me, why the authors did not test three-way interactions (MNESRES x age x sex).

(d) It was not clear to me why the authors tested the different linear regression models the way they did. It seemed to me that they wanted to show that although some of the MNESRES facets did not have predictive value above and beyond the other facets, there were some bivariate associations. Two options occur to me: (1) reporting a bivariate correlation table (which I would suggest anyway), and/or (2) reporting regression coefficients for each individual MNESRES facet (controlling for age and sex in each model) & for all combined facets. Also, I would like to see the full models and not just the interaction effects in Step 3. The presentation of results in Table 5 seems slightly misleading to me as the effect of SE_SE of course turns insignificant when adding the predictors in Step 3 (third set of HRA).

(4) Results

(a) CFA model comparisons: Please note that I am not an expert on SEM. I did not understand why the authors chose Model 2 (correlated factors) instead of a model with a superordinate factor. I see that the AIC was slightly lower in the former model. However, on theoretical grounds, it seems to me that the assumption is that the different factors are part of one underlying construct, namely, self-regulatory efficacy in managing negative emotions. It seems so much more parsimonious when predicting other outcomes to only have one score (i.e., the superordinate score); in particular, when the outcome is as broad as NA (i.e. not at the same level of discrete emotions) or SWL. I would be curious how reliable and predictive an aggregated score would be.

(b) Interaction effects in the prediction of SWL (regression): The interpretation of the interaction effects does not seem quite right to me. The interaction effects show differences between male and female participants. I am not sure how the data was coded, but assuming that men = 0 and women = 1, self-efficacy for regulating fear was more strongly associated with SWL for men than for women, and the other way around for shame. Please check and report the coding in the manuscript.

(5) Discussion

(a) The authors state that they found support for the reliability and validity of the scale. However, while the authors investigate the factorial structure, the reliability of the measure was not tested (in terms of internal consistency or re-test reliability). In terms of validity, the authors tested the predictive validity, but not convergent and discriminant validity, for example. Therefore, the authors may want to be more specific with their conclusions.

(b) I do not understand why the authors conclude that “one may surmise that the common distinction between basic and self-conscious/moral emotions holds good also for self-efficacy beliefs relating to the management of those emotions.” (p. 27). If that was the case, shouldn’t the factor structure be different (i.e., two factors)?

(c) I would have wanted the authors to discuss the fact that guilt and shame had no predictive value above and beyond the other facets. One hypothesis that came to my mind was that these emotions may be relatively rare in healthy adults’ everyday life?

(6) Other comments

(a) Sometimes, sign. results are presented in bold print in the tables, and sometimes not. In Table 4, the estimate for the SE_G X Sex interaction is printed in bold although it is not sign (outcome SWL).

(b) I was not entirely sure whether mastery beliefs and self-efficacy were used synonymous here.

(c) The data (excel) is not sufficiently labeled. Also, labeling in English would be appreciated.

Reviewer #2: The researchers addressed three aims: (1) examine measurement invariance of the MNESRES across gender and age in a Spanish sample; (2) examine gender and age differences in self-regulatory self-efficacy beliefs using the MNESRES; and (3) examine associations between discrete emotion self-regulatory self-efficacy beliefs, negative affect, and life satisfaction. These are interesting and important aims and I commend the authors for this study. However, I think several aspects of the presentation and analyses could be improved. I have outlined my specific concerns and suggestions below.

General Writing:

1. The clarity and organization of the manuscript could be improved. In general, the sentences are very long and the paragraphs are rather short. This made it difficult for me to follow the authors' logic from one idea to the next in many locations.

2. There are grammatical errors and inconsistencies throughout the manuscript that should be addressed. For example, individual's is used where individuals' should be used instead in multiple locations. Hyphens in words like life-satisfaction, age-group, and "life-experience" are unnecessary and used inconsistently.

3. In the abstract and in several locations throughout the manuscript, the authors combine all three research aims into a single sentence or phrase. This makes it difficult to understand the key research questions. Instead, the authors should consider stating the three aims separately.

4. The authors should remove the causal language from the Title ("counteract" and "promote").

5. It is unclear throughout whether the basic emotions versus moral emotions distinction is a key research question/guiding motivation of the present research or if it is a secondary exploratory analysis.

6. The manuscript should be carefully proofread and edited for clarity.

Introduction:

7. The first sentence of the Introduction is much broader than the topic of the paper. An opening sentence focused on self-efficacy beliefs (rather than theories of personality more generally) would be more appropriate.

8. The Introduction could benefit from clearer organization, perhaps with subheadings or at least specific paragraphs focused on the different components of the research question (e.g., self-efficacy beliefs, the structure of emotion regulation self-efficacy beliefs, the role of discrete emotions, associations between self-efficacy beliefs and well-being outcomes).

9. Several assertions in the Introduction lack references. (e.g., "While people display basic emotions like anger and sadness from the early stages of life, self - conscious/moral emotions do not appear unless self-structures and processes develop..." and "A large body of research has documented that the beliefs people hold about their capacity to cope efficaciously with specific challenges and to face demanding situations, exert a pervasive influence over thought, motivation and action across different domains of functioning.").

10. The Introduction focused heavily on scale development and is missing relevant literature to links between emotion regulation self-efficacy beliefs and well-being outcomes. This seems like an important literature to review, given that it is a key focus of the present research. Relatedly, the authors should consider reviewing emotion regulation self-efficacy research beyond the MNESRES (e.g., Emotion Regulation Questionnaire - Self Efficacy; Goldin et al., 2009; beliefs about emotions; Ford & Gross, 2019).

11. The following hypothesis seems much too broad to come from the reviewed evidence: “These findings further corroborate the hypothesis postulating that distinct discrete emotions play a different role with regard to adaptation and, as such, they require a specific consideration regarding their assessment and their treatment.” Is this the authors' hypothesis? If so, it should be significantly narrowed. If it is an existing hypothesis from the literature, a reference should be cited and it should be used to support the researchers' focus on discrete emotions rather than the other way around. In other words, this broad hypothesis can motivate the researchers' more specific research question, however, the researchers' specific research question cannot motivate this broader hypothesis.

Method and Results

12. What does "Within such not previously explored age group," refer to? In the next paragraph, the authors note that the scale was previously validated in an Italian sample of young to older adults.

13. The researchers should list all 10 negative emotional states in the PANAS. This is crucial information for interpreting findings given that the research question is about discrete emotions. Readers shouldn’t have to refer to a separate reference to obtain this information. My key question here is, are basic emotions and self-conscious emotions both represented in the PANAS items? Are the five discrete emotions from the MNESRES represented in the PANAS?

14. The authors need to justify their analytic approach, in particular, the hierarchical regressions.

14a. Why were age AND gender interactions included together? This makes for a more complex model that could obscure significant interactions.

14b. Why were moral emotions and basic emotions included in different steps. Was this based on theory? Was this based on the observed correlations or the result of the first set of regressions? If so, this should be noted.

15. Given the focus on basic versus moral emotions, I am surprised the researchers did not test this distinction in the factor analysis. Was there a reason for not testing a model with a basic emotions factor and a moral emotions factor?

16. The abbreviations for the different discrete emotions are hard to remember. The authors should consider using more descriptive abbreviations or the complete terms (e.g. SE_fear).

Discussion:

17. I was surprised that the authors did not discuss the differential associations of discrete emotions versus moral emotions with well-being, given the focus on this in the Introduction and Results. Could these results be explained simply by the fact that basic emotions occur more frequently and/or are better represented in the PANAS?

Reviewer #3: Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) (Limit 200 to 20000 Characters)

Could you please see my attached review, I have added all my comments to the Author in an attachment.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Review PONE-D-20-00499.docx

Click here for additional data file.

8 Sep 2020

We have incorporated all the reviewers'suggestions into our revision. They were very helpful. Thank you

Submitted filename: Response to reviewers.docx

Click here for additional data file.

8 Oct 2020

PONE-D-20-00499R1

How Self-efficacy Beliefs in Dealing with Negative Emotions are associated to Negative Affect and to Life Satisfaction across Gender and Age

PLOS ONE

Dear Dr. Caprara,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by 28th October. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Paola Iannello

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: (No Response)

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: I read the manuscript for the second time. The revised version is much more concise and much easier to follow. I also very much appreciate that the authors now provided the data with the relevant variables labeled in English. However, I still have a few concerns.

(1) Regulation of basic emotions vs. moral emotions: I still had some difficulty following the argument that the regulation of basic emotions should be more important than the regulation of moral emotions. For parts of the argument, references to relevant research findings seem to be missing. For example, the authors write (pp. 5-6): “Indeed, negative basic emotions like anger, fear and sadness have a pervasive impact over individuals’ life, and their regulation has a primary influence on adaptation when compared to self-conscious/moral emotions like guilt and shame.” I think such a statement (and similar statements in the manuscript) would need a reference.

(2) Reliabilities of the MNESRES subscales were low for anger and sadness (.55, .54). This could be discussed.

(3) Figure 1: Some correlations between factors seem to be missing from the figure (e.g., correlations between SE_G and SE_DS.

(4) Interaction effects/ Figure 2: I still think that the authors’ interpretation of the interaction effects is (a) not really in line with their research question and (b) not 100% correct. With regard to (a), I think what the authors did (and illustrated), is considering the SEs as moderators of the association between sex and satisfaction with life. However, since the research is concerned with associations between the SEs and satisfaction with life, it makes much more sense to me to consider sex as the moderator here. Statistically, there is of course no difference between considering sex or the SEs as moderators. However, the interpretation is slightly different. The results show that there is an association between SE_fear and life satisfaction: Individuals with higher SE_fear have higher life satisfaction. Treating sex as a moderator, the interaction effect (SE_fear x sex, b= -.12) means that for women, the association (semi-partial correlation) between SE_fear and life satisfaction is weaker than for men. Although there is no main effect of SE_shame on life satisfaction, there is an interaction effect SE_shame x sex (b=.14), indicating that for women, SE_shame is related to satisfaction with life, but not for men.

With regard to (b), the authors write: “life satisfaction was particularly higher in women with high SE_Shame than in both men and women with low SE_Shame (Figure 2B), as well as in men with high SE_Shame.” This cannot be concluded from the interaction effect, which only indicates that there are differences in trajectories. One would have to test these specific differences between higher and lower groups post-hoc. In the plots presented in the paper, it is fairly obvious that women high in SE_shame are not more satisfied with their life than men with high SE_shame (the difference is only about 0.1 points on the scale). As sex is a categorical variable here, I also do not think that plotting the data as trajectories in the way presented in the paper is most useful. Therefore, I would strongly suggest to switch to considering sex as the moderator, and changing the interpretation and the plots accordingly.

(5) Discussion: I think that some statements made by the authors in the discussion need further elaboration. I also felt that some conclusions go beyond what was studied.

(a) “Albeit with caution, it may be concluded that the five-oblique-factor solution further corroborates the view that self-efficacy beliefs are associated with the regulation of discrete emotions.” (p.20) Since the questionnaire measures self-efficacy beliefs with regard to regulating discrete emotions, I am not sure how a factor analysis can corroborate the association between self-efficacy beliefs and regulation. Maybe the authors could rephrase?

(b) “However, the present and previous findings invite us to be cautious about the extent to which the regulation of negative emotions can be traced to a common sense of mastery.” (p. 21) – In the present study, a model with one common mastery factor was not tested. The fact that a model with five correlated factors fit the data best does very much invite the idea that there may be a common factor, I think. I still think that the best option here would be to simply test a model with a common factor. As a reader of this paper, I would be very interested in that.

(c) “Thus, one may surmise that the common distinction between basic and self-conscious/moral emotions also holds well for self-efficacy beliefs in terms of the management of those emotions.” (p.21). The reply given by the authors regarding my question in the first review does not convince me. I still fail to see how the present study supports this conclusion. Based on the CFAs, a model with 5 correlated factors fit the data best, which is a model that does not support this distinction. I understand that the fit of alternative models was also not that bad, but I do not think that this is strong evidence to corroborate a distinction. In fact, the correlations between the subscales (Table 4) show that the correlations between SE_shame with SE_sadness and with SE_fear were similar to the correlation between SE_anger and SE_fear, which does not indicate that SE_shame belongs to a different factor than the other SEs. Furthermore, all factors correlated with the outcomes (negative affect and satisfaction with life). The fact that SE_shame and SE_guilt were not uniquely predictive of the outcomes above and beyond all other factors is not indicative of these factors not belonging to the same underlying construct. It may very well be the case that the distinction makes sense conceptually and that this was found in other studies, but in the present study, the data does not support such a conclusion, I would say.

(d) “One might think that most people learn how to refrain from anger earlier” (p.22) – Why might one think that? (see Comment #1

Reviewer #2: The authors addressed the majority of comments raised in the prior round of revision. I appreciated that the authors tempered their conclusions about the factor structure and direction of causality in the Discussion section. I also appreciated that they considered both evolutionary and constructivist views of emotion in the Introduction. Overall, this is a thorough revision of a nice manuscript. I have outlined a small number of minor suggestions below.

1. The rationale for the hypothesis that regulation of basic emotions should be more important for well-being (bottom of page 5, top of page 6) was not clear.

2. Please name all 10 emotions that were included in the negative emotion measure. This information is important for evaluating the degree of overlap between the self-efficacy beliefs measure and the negative emotion measure.

3. It is unclear what is meant by this sentence in the Discussion: "Since the sample was taken mostly from a relatively affluent middle class population, one cannot exclude significant variations due to age and gender, either in the means or in the relationship of the variables examined, particularly among a population whose conditions of life are quite different." I believe the authors are trying to make a point about lack of generalizability beyond middle class population, but it is unclear to me what they are saying about age and gender.

4. The first paragraph includes jargon and complex sentence structures. This may be personal preference and the authors may feel free to leave this paragraph as is, however I found it difficult to understand.

Reviewer #3: The authors have appropriately addressed reviewer comments, and the manuscript is now much stronger. I have no further revisions.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

29 Oct 2020

We are grateful for the constructive comments of the reviewers, which we believe have helped us to present our data more clearly. In the attached cover letter we detail our responses to the reviewers’ comments and we indicate the changes made in the revised manuscript.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

2 Nov 2020

How Self-efficacy Beliefs in Dealing with Negative Emotions are associated to Negative Affect and to Life Satisfaction across Gender and Age

PONE-D-20-00499R2

Dear Dr. Caprara,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Paola Iannello

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

9 Nov 2020

PONE-D-20-00499R2

How Self-efficacy Beliefs in Dealing with Negative Emotions are associated to Negative Affect and to Life Satisfaction across Gender and Age

Dear Dr. Caprara:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Paola Iannello

Academic Editor

PLOS ONE