Who expresses their pride when? The regulation of pride expressions as a function of self-monitoring and social context

ABSTRACT

Pride expressions draw attention to one’s achievement, and therefore can enhance one’s status. However, such attention has been linked to negative interpersonal consequences (i.e. envy). Fortunately, people have been found to regulate their pride expressions accordingly. Specifically, pride expressions are lower when the domain of the achievement is of high relevance to observers. We set out to replicate this effect in a non-Western sample. Additionally, we extended the current finding by investigating the moderating role of self-monitoring, an individual’s ability and willingness to adjust their behaviours under different social contexts to cultivate status. This allows us to explore the previously assumed underlying status motive in regulating pride expressions. Data from two preregistered studies (N₁ = 913; N₂ = 1081) replicated the effect that pride expressions are inhibited when the achievement domain is relevant. A significant main effect of self-monitoring was found, such that high self-monitors express more pride than low self-monitors, consistent with the conceptualisation of self-monitoring as rooted within a status-enhancement motive. The assumed interaction effect between domain relevance and self-monitoring was not significant. Our findings suggest that the effect of domain relevance on pride expression is robust and status driven.

KEYWORDS:

• Pride expression
• pride suppression
• self-monitoring
• achievement

Emotional expressions are often regulated in a strategic manner in order to benefit from social interactions, maintain certain reputations, and sustain relationships (Tamir, 2016). However, positive emotions, compared to negative emotions, are less regulated and often expressed (Nezlek & Kuppens, 2008; Schall et al., 2016) since expressing positive affect has been linked to social-relational benefits (Reysen, 2005). Recent evidence, however, suggests that positive emotional expressions are also subject to display norms, especially when those emotional expressions can have negative interpersonal consequences for the person expressing them (Manokara et al., 2020). In this paper, we study the regulation of pride expressions and whether this regulation is dependent on the relevance of the achievement domain to observers, and the trait to self-monitor.

Pride is an emotion for which the expression may have negative interpersonal consequences, and there is mounting evidence that it may be functional to inhibit expressions of pride as they can unintentionally elicit negative responses from observers of the display. These negative reactions can range from perceiving the outperformer (one who outperformed other or achieved outstanding results in a task or competition) as boastful and arrogant to triggering feelings of envy, which can lead to other social costs such as hostility and resentment toward the envied person (Lange & Crusius, 2015). Interestingly, outperformers seem to anticipate these negative reactions from their audiences and accordingly regulate the expression of positive emotions (happiness and enjoyment) after a victory to avoid backlash (Kalokerinos et al., 2014; Schall et al., 2016). Further, envy seems to be triggered in situations in which the domain of achievement is of personal relevance (referred to as domain relevance) and is important to the outperformed individual who is observing the pride display (Smith & Kim, 2007). Expressions of pride are in fact inhibited or deamplified by outperformers when their achievement is personally relevant to their audience. This effect was observed in self-reported and observed pride expressions, and of medium size (Hedge’s g = 0.50; van Osch et al., 2019).

The assumed explanation for these findings is that pride expressions serve the function of drawing attention to one’s achievements to gain social status (Tracy et al., 2013). However, if observers feel bad after witnessing such an expression, or perceive the pride expresser in a negative light, this function is thought to be interrupted by the associated negative backlash (van Osch et al., 2016; van Osch et al., 2019). Therefore, it is expected that people strategically deamplify their pride expressions in situations where their achievements are of relevance for their audience to prevent potential negative effects (e.g. envy) on their own reputation (e.g. van Osch et al., 2019). This assumed motivation however has never been tested. To explore whether the regulation of pride expression is related to one’s reputation and status, we turn to the trait of self-monitoring.

Self-monitoring is an individual difference that concerns one’s willingness and ability to exert control over expressive behaviours to cultivate a specific appearance (Snyder, 1974). High self-monitors are more attuned and responsive to social and interpersonal cues, and able to adapt and exert more (expressive) control over their behaviours and self-presentation, whereas low self-monitors tend to behave and express themselves in a more consistent manner across situations (e.g. Scott et al., 2012). In addition, high self-monitors seem to be more adept at recognising social cues (Costanzo & Archer, 1989), which provides them with the necessary input to evaluate and determine their desired course of action. Moreover, there is already evidence that high self-monitors adjust their expressions of positive emotions when deemed appropriate. Friedman and Miller-Herringer (1991) showed that high, compared to low, self-monitors were perceived to express less positive affects (e.g. happiness, positive emotion in general) after receiving good performance feedback when with fellow students rather than when alone. In line with these findings, we expect high self-monitors to be better able to identify relevant cues in outperformance situations and adapt their pride expressions to avoid negative social evaluations when necessary.

Apart from a greater sensitivity to social cues, self-monitoring is also assumed to be related to the desire to improve one’s social status (Gangestad & Snyder, 2000). Research has shown that high self-monitors react more positively to image-focused marketing messages (compared to quality-focused messages) and are willing to test and spend more money on products with image-oriented advertisements (Snyder & DeBono, 1985). More importantly, self-monitoring has been shown to correlate positively with the need for social status (e.g. Flynn et al., 2006). Because the function of pride expressions is to communicate success and enhance social status, we expect high self-monitors to express pride more intensely than low self-monitors when there are no expected social costs attached to expressing pride.

In this study, we test the following hypotheses:

H1: Participants in a domain relevant condition express less pride over their achievement compared to those in a domain non-relevant condition.

H2: We expect to observe an interaction effect between domain relevance and self-monitoring, such that in a domain relevant condition, high self-monitors would express less pride than low self-monitors (H2a), whereas in a domain non-relevant condition, high self-monitors would express more pride than low self-monitors (H2b).

The contribution of this paper is twofold. First, we aim to replicate the effect of domain relevance on pride expression (cf. van Osch et al., 2019) in a non-WEIRD sample of Vietnamese university students. As most existing studies were conducted using WEIRD samples, the generalizability of this finding remains unclear (Henrich et al., 2010). Second, we extend the original findings by delving deeper into the previously assumed explanation for the regulation of pride expression by exploring the effect of self-monitoring. If pride is indeed expressed and inhibited strategically, we would expect this to happen to a larger extent among those who care more about their reputation (i.e. high in self-monitoring).

We tested these hypotheses in two pre-registered studies. In these studies, we used vignettes in which participants imagined achieving something and then were asked by others about how they had performed, in other words give them an imagined opportunity to express their pride. Of course, such studies do not capture in vivo pride experiences nor allow us to observe online pride expressions. It is complicated to manipulate and observe pride expressions, especially during COVID-19. One could therefore argue that our studies cannot reveal anything about pride expressions but tap more into ideas about the appropriateness of pride expressions. Although this poses a limitation on the conclusions that our study can draw, there is evidence from follow-up studies of the experiment we are replicating that the observed effects of domain relevance on pride expressions were similar for both self-reported pride expressions in hypothetical situations as well as for actual observed pride expressions in behavioural experiment (Experiment 4; van Osch et al., 2019).

Study 1

Method

Participants

In this pre-registered study (https://aspredicted.org/blind.php?x=jc2yh7) participants were assigned to one of two experimental conditions in which the vignette presents a relevant or a non-relevant achievement to their audience. We conducted an a priori power analysis for multiple linear regression with four predictors to achieve a power of 0.8 with a significance level of 0.025 for an interaction effect of 0.02, and achieved a target sample size of 478. A total of 1912 Vietnamese university students were approached by four contacts who assisted in emailing university students in Ho Chi Minh City over a period of 3 weeks.

Responses in which participants (1) spent less than 10 seconds reading the vignette, (2) completed the entire survey in less than 60 seconds, (3) failed the attention check were excluded based on our pre-registered exclusion criteria. In addition, responses in which participants (4) were younger than 18 years old (3 cases), (5) did not answer all items for the dependent variables, and (6) did not complete the survey within the data collection period were also excluded from our final sample.

After exclusion, our final sample consisted of 913 respondents (19.4% male, 80.2% female, 0.4% other) with an age range of 18–29 (M = 20.27, SD = 1.75).

Procedure and measures

Participants first read a vignette that described them having achieved something that was either relevant or non-relevant to their audience (i.e. classmates). Subsequently, participants answered questions about this vignette concerning their experienced and expressed pride. After a manipulation check, participants then completed the self-monitoring scale, an attention check, and provided demographic information (age, gender).

Domain relevance was manipulated by presenting participants with vignettes in which the domain of the achievement was either relevant or non-relevant to the people they were communicating with (cf. van Osch et al., 2019). Participants were randomly assigned to one of two conditions in which they either achieved academically on an exam that their classmates also partook in (relevant condition) or in a sports tournament that their fellow students did not partake in (non-relevant condition). Subsequently, they were informed that their classmates had asked them how their exam/tournament had been and were asked to communicate to these fellow students. The original non-relevant vignette asked participants to imagine winning a squash tournament. However, as squash is an uncommon sport in Vietnam, the vignette was adapted to ping-pong as it is not commonly played amongst young adults but widely known and identifiable.

Experienced pride was measured using three items: “To what extent do you feel proud/satisfied/happy with your achievement?” (1 = not at all, 7 = very much; Cronbach’s α = 0.81).

Verbally expressed pride was measured using three items: “To what extent would you show your fellow students what you are capable of/share this experience with your fellow students/tell your fellow students about your achievement?” (1 = not at all, 7 = very much; Cronbach’s α = 0.81).

Nonverbally expressed pride was measured using a 6-point visual scale consisting of 6 pictures of a woman expressing her pride ranging from neutral/no pride to very intense pride (van Osch et al., 2019). Participants reported how they would express themselves towards their fellow students by clicking on one of the pictures.

Self-monitoring was measured using the Revised Self-Monitoring Scale (R-SMS, 18-items) developed by Snyder and Gangestad (Gangestad & Snyder, 2000; Snyder & Gangestad, 1986). Participants answered True/False for each item (e.g. I have trouble changing my behaviour to suit different people and different situations.) and a composite sum score was created for self-monitoring. Ten out of 18 items were reverse coded (Snyder & Gangestad, 1986). An exploratory factor analysis (EFA) performed on this scale indicates a 5 factors structure.¹

All materials were translated to Vietnamese using a translation-back-translation procedure by the first author and an independent translator, both native in Vietnamese and fluent in English. Inconsistencies were resolved through discussion. All materials and data can be found on https://osf.io/xc79v/files/.

Results

Descriptive statistics and correlations between study variables can be found in the Supplementary Material (Table S1).

Manipulation check

To test if our manipulation were successful in varying the perceived relevance of the achievement domain, participants answered to what extent they thought their fellow students would also like to achieve the same thing they had achieved (1 = not at all, 7 = very much). The relevance manipulation was successful: Participants reported to perceive their fellow students to want to achieve more in the exam vignette (M = 5.53, SD = 1.70) than in the ping-pong condition (M = 4.30, SD = 1.80), t(912) = −10.57, p < .001 (CI 95% = −1.45; −1.00), Cohen’s d = 0.70.

Additionally, there was a difference in the reported experienced pride across conditions. Participants in the relevant academic condition reported lower levels of experienced pride (M = 5.37, SD = 1.26) to those in the non-relevant sport condition (M = 5.64, SD = 1.13), t(911) = 3.40, p < .001 (CI 95% = .11; .42), Cohen’s d = 0.23.

Pride expressions

Two separate ANOVAs with domain relevance as independent variable and the two types of pride expressions as dependent variables supported H1. Participants reported to verbally express less pride in the relevant (M = 2.70, SD = 1.22) than in the non-relevant condition (M = 3.63, SD = 1.40), F(1, 911) = 111.70, p < .001, η² = .11. Participants also reported to express less pride nonverbally in the relevant (M = 2.51, SD = 1.06) than in the non-relevant condition (M = 2.91, SD = 1.28), F(1, 911) = 26.73, p < .001, η² = .03.

Separate ANCOVAs revealed that experienced pride was a significant covariate. Controlling for experienced pride, the effect of relevance on both verbal and nonverbal pride expressions remained significant; F(2, 910) = 98.91, p < .001, η² = 0.10 and F(2, 910) = 21.19, p < .001, η² = 0.02, respectively.

Self-monitoring

Self-monitoring was mean-centered in all models. Table 1 presents the results of the hierarchical regression models for verbal and nonverbal pride expressions separately. As expected, we found that domain relevance is negatively associated with the expressions of verbal and nonverbal pride. Self-monitoring was positively associated with nonverbal pride expression; however, this effect was not significant for verbal pride expression (p = .05), partially supporting our hypothesis. Lastly, H2a and H2b were not supported since the interaction of domain relevance and self-monitoring was not significant.

Table 1. The effects of domain relevance, self-monitoring on expressed pride controlled for experienced pride in Study 1.

		Verbal pride expression		Nonverbal pride expression
	Predictor	Beta	SE	Beta	SE
Model 1	Domain Relevance	−.82***	.08	−.37***	.08
	Self-Monitoring	.03^+	.01	.05***	.01
	Experienced Pride	.38***	.03	.17***	.03
	F	86.95***		23.74***
	R^2 (Adj R^2)	.22 (.22)***		.07 (.07)***
Model 2	Domain Relevance	−.82***	.08	−.37***	.08
	Self-Monitoring	.02	.02	.06**	.02
	Experienced Pride	.38***	.03	.17***	.03
	Domain Relevance*Self-Monitoring	.02	.03	−.02	.03
	F	65.33***		17.92***
	ΔF	.60		.48
	R^2 (Adj R^2)	.22 (.22)		.07 (.07)

⁺ p = .05, *p < .05, **p < .01, ***p < .001.

Discussion study 1

This pre-registered study was successful in replicating the effect of domain relevance of an achievement on both verbal and nonverbal pride expressions, such that if the domain of one’s achievement is relevant for their audience, achievers express less pride than when their achievement is non-relevant. The study also expanded on the original study by investigating the effect of individual differences in self-monitoring on pride expression. A main effect of self-monitoring was observed, such that people who self-monitor more express more nonverbal pride compared to those who self-monitor less. The expected interaction between self-monitoring and domain relevance was not observed, and thus the data failed to support H2a and b.

Although our study was successful in replicating the main effect of domain relevance on pride expression, our design has limitations. In the original study (cf. van Osch et al., 2019), achievement domain (academic versus athletic) and domain relevance (relevant versus non-relevant) were confounded, and thus this is also a confound in the current study. It could be argued that it is more appropriate to express pride in a sports situation than in an academic situation, especially so for nonverbal pride expressions, and that this difference between the conditions was responsible for the observed effect rather than domain relevance.² Additionally, we also found that participants reported to experienced more pride in the nonrelevant athletic condition compare to the relevant academic condition, suggesting that different domains of achievement (academic vs. athletic) elicit different level of pride experiences.

Study 2

We conducted a second pre-registered study, with a slight but crucially different manipulation to eliminate the confound present in Study 1, in which we hold achievement domain constant (academic achievement), and only manipulate the relevance of the achievement for the audience: the audience consisted of friends who partook in the same exam (relevant condition) or friends who did not partake in the same exam (non-relevant condition).

We also took this opportunity to delve deeper into the role of self-monitoring, given the non-significant interaction effect and the observation that the self-monitoring scale had a multidimensional factor structure. Some researchers have suggested that self-monitoring consists of not one but two independent dimensions: acquisitive versus protective self-monitoring (e.g. Lennox, 1988). Acquisitive self-monitoring (ASM) relates to one’s desire and tendency to “get-ahead”, conceptualised as a more power- and goal-oriented tendency, while protective self-monitoring (PSM) relates to the motivation of “getting-along” with others and persevering the harmony of the groups (Lennox, 1988; Wilmot et al., 2017). Although both high ASM and PSM individuals employ behaviour management tactics, the underlying motivations are different. Based on these conceptualizations, Wilmot et al. (2017) used the original 25-item SMS as their item pools to develop the Acquisitive Self-Monitoring Scale (A-SMS, 6-item) and the Protective Self-Monitoring Scale (P-SMS, 7-item) with good reliability and psychometrics properties (for details, see Wilmot et al., 2017). By measuring ASM and PSM separately, we can more deeply explore the potential motives behind the inhibition of pride expression found in Study 1. More specifically, if people inhibit their pride expressions because they do not want to deal with negative backlash and preserve their status, we expect ASM (status-driven) to be associated with the regulation of pride expressions. If expressers however inhibit because they do not want to hurt other people's feelings by eliciting negative emotions (i.e. envy), especially in domain relevance situations, then PSM (cohesion-driven) should be related to pride expressions.

Method

Participants

In this pre-registered study (https://aspredicted.org/FNX_2XW), participants were randomly assigned to one of two conditions. We conducted an a priori power analysis for multiple linear regression with two predictors using the effect size of Study 1 (partial R² = .02) to achieve a power of 0.8 with a significance level of 0.025, and achieved a target sample size of 518. A total of 2849 Vietnamese university students were approached by three contacts who assisted in emailing university students in Can Tho city over a period of about 4 months. After exclusion (employing the same exclusion criteria as in Study 1), our final sample consisted of 1081 respondents (32.01% male, 66.51% female, 1.48% other), with an age range of 18–39 (M = 20.48, SD = 2.83).

Procedure and measures

The procedure was identical to Study 1, with the exception of the manipulation of domain relevance and the scale used to measure self-monitoring.

Domain relevance was manipulated by presenting participants with vignettes in which the domain of the achievement was either relevant or non-relevant to the people they were communicating with (cf. van Osch et al., 2019). Participants were randomly assigned to one of two conditions in which they had achieved academically (scoring high on an exam). In the relevant condition, their friends had also taken the exact same exam, whereas in the non-relevant condition, their friends had not taken that exam. Subsequently, they were informed that their friends had asked them how their exam had been and were asked to communicate to their friends.

We used identical measures for Experienced pride (α = .85), Verbally expressed pride (α = .80), Nonverbally expressed pride, and our manipulation check as in Study 1, which proved reliable.

In the second study, we adopted the longer version of self-monitoring measurement and used the 25-item scale instead of the 18-item version. The advantage of the larger scale is that it allows us to replicate Study 1 by retaining the 18-item R-SMS, while provides the opportunity to investigate the two subscales, 6-item acquisitive and 7-item protective self-monitoring.³ Again, participants answered True/False for each item, and the additional items added were translated using the same procedure as in Study 1.

Similar to Study 1, an EFA on the R-SMS showed 5 factors with inconsistent factor loadings (see Supplementary Material, Figure S2 and Table S4). In contrast, an EFA with 13 items from the A-SMS and P-SMS scales showed a two-factor structure with the items loading onto their corresponding scales (see Supplementary Material, Table S5). We created sum scores for the A-SMS and P-SMS for our main analyses, and only used the 18-item R-SMS sum score as a comparison to Study 1.

Results

Descriptive statistics and correlations between study variables can be found in the Supplementary Material (Table S2).

Manipulation check

The relevance manipulation was successful: Participants reported to perceive their fellow students to want to achieve more in the relevant (M = 5.38, SD = 1.80) than in the non-relevant condition (M = 4.95, SD = 1.79), t(1078) = −3.9, p < .001 (CI 95% = −.64; −.21), Cohen’s d = 0.24.

Additionally, there was no difference in the reported experienced pride across conditions. Participants in the relevant condition reported similar levels of experienced pride (M = 5.30, SD = 1.35) to those in the non-relevant condition (M = 5.39, SD = 1.37), t(1076.8) = −1.0772, p = .28 (CI 95% = −.07, .25).

Pride expressions

Separate ANOVAs with domain relevance as the independent variable and the two types of pride expressions (verbal and nonverbal) as dependent variables were performed. Participants reported to verbally express less pride in the relevant (M = 2.67, SD = 1.31) than in the non-relevant condition (M = 2.88, SD = 1.37), F(1, 1079) = 6.77, p = .009, η² = .006. However, the effect of domain relevance on nonverbal pride expression was insignificant, F(1, 1072) = 2.568, p = .11, η² = .002. These results partially support H1.

Separate ANCOVAs revealed that experienced pride was a significant covariate. When controlled for experienced pride, the effect of domain relevance on verbal pride expressions remained significant; F(2, 1078) = 51.99, p < .001, ${\eta }_{\mathrm{Condition}}^2$ = 0.006, and the effect on nonverbal pride expression remained insignificant; F(2, 1071) = 16.31, p = .153, ${\eta }_{\mathrm{Condition}}^2$ = 0.002.

Self-monitoring

Self-monitoring was mean-centered in all models. Using the computed score for the R-SMS (18-item) as in Study 1, hierarchical regression model revealed that both domain relevance and self-monitoring significantly affected verbal pride expression: Domain relevance is associated with less verbal pride expressions but no difference in nonverbal pride expression, while high self-monitoring is associated with more verbal and nonverbal pride expressions. Contrary to our predictions, no interaction effect was found between domain relevance and self-monitoring. Table 2 presents the results of the hierarchical regression models for verbal and nonverbal pride expressions separately.

Table 2. The effects of domain relevance, self-monitoring on expressed pride controlled for experienced pride in Study 2.

		Verbal pride expression		Nonverbal pride expression
	Predictor	Beta	SE	Beta	SE
Model 1	Domain Relevance	−.21**	.08	−.11	.07
	Self-Monitoring	.15***	.04	.13***	.03
	Experienced Pride	.27***	.03	.12***	.02
	F	39.58***		15.84***
	R^2 (Adj R^2)	.10 (.10)***		.04 (.04)***
Model 2	Domain Relevance	−.21**	.08	−.11	.07
	Self-Monitoring	.13*	.06	.13**	.05
	Experienced Pride	.27***	.03	.12***	.02
	Domain Relevance*Self-Monitoring	.04	.08	−.01	.07
	F	29.75***		11.87***
	ΔF	0.31		0.01
	R^2 (Adj R^2)	.10 (.10)		.04 (.04)

Note: Self-monitoring score is computed using the 18-item R-SMS, similar to Study 1.

*p < .05, **p < .01, ***p < .001.

Similar analyses were performed with the computed A-SMS and P-SMS scores as predictors. As can be seen in Tables 3 and 4, A-SMS revealed a significant main effect for both verbal and nonverbal pride expressions, whereas for P-SMS no effects were observed. Again, no interactions between either type of self-monitoring and domain relevance were observed.

Table 3. The effects of domain relevance, acquisitive self-monitoring on expressed pride controlled for experienced pride in Study 2.

		Verbal pride expression		Nonverbal pride expression
	Predictor	Beta	SE	Beta	SE
Model 1	Domain Relevance	−.20*	.08	−.10	.07
	Acquisitive Self-Monitoring	.12**	.04	.16***	.03
	Experienced Pride	.27***	.03	.12***	.02
	F	36.92***		18.58***
	R^2 (Adj R^2)	.09 (.09)***		.05 (.05)***
Model 2	Domain Relevance	−.20*	.08	−.10	.07
	Acquisitive Self-Monitoring	.08	.06	.14**	.05
	Experienced Pride	.27***	.03	.12***	.02
	Domain Relevance*Self-Monitoring	.07	.08	.03	.07
	F	27.91***		13.98***
	ΔF	0.91		0.21
	R^2 (Adj R^2)	.09 (.09)		.05 (.05)

*p < .05, **p < .01, ***p < .001.

Table 4. The effects of domain relevance, protective self-monitoring on expressed pride controlled for experienced pride in Study 2.

		Verbal pride expression		Nonverbal pride expression
	Predictor	Beta	SE	Beta	SE
Model 1	Domain Relevance	−.20*	.08	−.08	.07
	Protective Self-Monitoring	.03	.04	−.06	.03
	Experienced Pride	.28***	.03	.14***	.02
	F	33.84***		11.89***
	R^2 (Adj R^2)	.09 (.08)***		.03 (.03)***
Model 2	Domain Relevance	−.20	.08	−.08	.07
	Protective Self-Monitoring	.003	.06	−.07	.04
	Experienced Pride	.28***	.03	.14***	.02
	Domain Relevance*Self-Monitoring	.05	.08	.02	.07
	F	25.46***		8.92***
	ΔF	0.37		0.05
	R^2 (Adj R^2)	.09 (.08)		.03 (.03)

*p < .05, **p < .01, ***p < .001.

Discussion study 2

This study addressed the potential confound in Study 1 and successfully replicated the effect of domain relevance and self-monitoring for verbal pride expressions. However, no effects of domain relevance were observed for nonverbal pride. Hypothesis 1 was thus supported for verbal but not for nonverbal pride expressions. This could have to do with the domain over which pride was expressed. As noted prior and also observed in other studies (e.g. van Osch et al., 2019), in some achievement domains, nonverbal pride expressions are more common and accepted and thus more intense. Displaying the typical nonverbal pride expression (arms up in the air, chest expanded, chin up) is very common and accepted in sports contexts, but maybe less natural in an academic context. Regardless, participants still indicated to express less pride in a verbal manner when their achievement was relevant for their audience than when it was non-relevant, confirming H1.

Although the effect of domain relevance on verbal pride expression from Study 1 was succesfully replicated, the effect size was substantially smaller in Study 2 than that in Study 1. This might be due to the COVID-19 pandemic and its development in Vietnam. Data for Study 1 was collected in June 2020, while Study 2 was conducted between November 2021 and March 2022. Until May 2021, Vietnam was minimally impacted by COVID, with few reported confirmed cases and deaths. However, the country experienced a spike in COVID cases and deaths starting from June 2021 until April 2022 (World Health Organization, 2023). This spike in infections and deaths after a year of relatively effective COVID-19 prevention could have impacted the emotional experiences of participants in Study 2, such as suppressing their experiences of positive emotions (i.e. pride) or their expression of positive emotions (i.e. talking about being proud), reducing the pride-inhibition effect of domain relevance. Additionally, there could be other demographic factors that were not accounted for that could explain the difference in effect sizes observed.

We also replicated the main effect of self-monitoring on pride expressions, confirming H2, using the exact same scale as in Study 1. Like in Study 1, we did not observe the expected interaction effect between self-monitoring and achievement domain. The acquisitive and protective self-monitoring subscales revealed that higher A-SMS scores were associated with more intense pride expressions, while higher P-SMS scores were not. These results do not support hypotheses 2a and 2b, as high self-monitors were not strategically expressing more or less pride given the relevance of the situation. These results do indicate that high self-monitors, especially those high in acquisitive self-monitoring, could employ pride expressions as a means to manage one’s status. Additionally, the pattern of results from the R-SMS, A-SMS and P-SMS scales suggest that the R-SMS is largely reflective of acquisitive self-monitoring, but not protective self-monitoring.

General discussion

Our research provides further support for the effect of domain relevance as an important social clue used by outperformers to regulate their pride expression. As suggested by previous studies (see van Osch et al., 2016, 2019) pride expressers strategically express their pride in a less intense manner when their achievement can be self-relevant to the audience of their expressions. Replicating the effect using both the original and the adjusted vignette (holding domain of achievement constant) in 2 large non-WEIRD samples of Vietnamese university students, we can conclude that the effect of domain relevance on pride expressions is relatively robust. Our expectation that those high in self-monitoring would be most strategic in expressing their pride in non-relevant situations and inhibiting in relevant situations was not confirmed. We failed to observe interaction effects between domain relevance and self-monitoring in both studies, and regardless of how self-monitoring was operationalised. We did observe a main effect of self-monitoring in both studies, specifically for those high in acquisitive self-monitoring, indicating that those who are motivated to get ahead express their pride more regardless of whether their achievements are relevant for their audience or not. This finding seems to contradict with the extant evidence showing that high (acquisitive) self-monitors not only are more willing but also more capable of regulating their behavioural expressions to avoid negative social evaluations (e.g. Friedman & Miller-Herringer, 1991).

There are several possible explanations to this finding. In outperformance situations, high acquisitive self-monitors are faced with two directly opposing interests: the display of pride over an achievement in pursuit of social status, or the inhibition of pride expression to prevent negative social evaluations, both of which are proposed to be integral motives of high acquisitive self-monitors. Our findings suggest that, when faced with a dilemma, high acquisitive self-monitors are more motivated by status and therefore, still express more pride compared to low self-monitors. It is conceivable that during situations of conflicting interest, high acquisitive self-monitors are more capable of managing opposing objectives and therefore still express pride for self-promotion while simultaneously employing other impression management strategies in tandem such as ingratiation (flattery, performing favours) or supplication (showcasing shortcomings) to mitigate negative social consequences of appearing boastful and arrogant (Turnley & Bolino, 2001).

The current study found the R-SMS scale to be non-unidimensional in 2 separate samples. The pattern of results for the R-SMS closely resembled that of the A-SMS subscale, but not of the P-SMS subscale. This suggests that self-monitoring, if measured with the R-SMS scale, largely reflects acquisitive self-monitoring and not protective self-monitoring. The current findings thus endorse the alternative bivariate model of acquisitive and protective self-monitoring by Wilmot and colleagues (2017). Additionally, we also observed that protective self-monitoring was not associated with the inhibition of pride expression in both the domain relevance and domain non-relevance conditions. This indicates that the motivation to get along with others is not a significant predictor of behaviours and expressions in situations of achievement.

Limitations and future research

First, as we already pointed out in the introduction, the current study was based on self-reports of behaviours in hypothetical situations, and thus cannot speak to how individuals would behave in real life situation, and might be more reflective of the expectation of cultural norms or ideas about how pride should be expressed. Again, we do want to note that previous research has shown that findings on self-reported pride expressions align with observed pride expressions (van Osch et al., 2019).

Second, due to the between-subject design, we could only draw conclusions on high self-monitors as a group. A within-subject design would allow us to investigate differences of acquisitive (and protective) self-monitors across different social contexts. Future research can explore this aspect by looking into the variation in behaviours of high and low self-monitors in different social situations with varying degrees of complexity in managing others’ impressions of oneself. Moreover, research into the effectiveness and successfulness of high and low self-monitors when employing different impression-management strategies will also shed more light onto the subject.

Third, as previously discussed, Study 1 and Study 2 were conducted during very different phases of COVID-19 in Vietnam, which might have affected the emotional experiences and expressions of participants differently, leading to a discrepancy in the effect sizes observed between studies. Hence, future research can aim to replicate Study 2 to get closer to the true effect size of domain relevance on pride expression.

Much has been written about display rules or norms for the expression of positive emotions (e.g. Matsumoto et al., 2008), and particularly about norms restraining the expression of so-called disengaging emotions such as pride in Asian contexts (e.g. Kitayama et al., 2006). Based on this literature, which is mute about situation specific aspects, one would expect that there would be no difference between relevant and non-relevant achievements in pride expressions, and that Vietnamese participants would de-amplify/inhibit/suppress expressions of a socially disengaging emotion as pride (Kitayama et al., 2006; Yoshie & Sauter, 2020). The current findings are not in line with such expectations as we observed that the regulation of positive emotions is goal or context dependent in an Asian country. Hence, our findings are more in line with recent research which has emphasise the importance of context in all, not just non-Asian, cultures (e.g. Greenaway et al., 2018; Ma et al., 2018; Manokara et al., 2020; van Osch et al., 2016). We of course acknowledge that the current study is limited as a direct comparison between a Western and Eastern sample is lacking and necessary to check whether we can assume equivalence between measurements. This limitation also holds for all interpretations of the main effect of self-monitoring on pride expressions, which for now was only observed in an Asian sample. Future research could expand on our findings and investigate the relationship between self-monitoring and pride expression in difference samples of different cultures to further shed lights on this association.

In conclusion, pride expressions are sensitive to cues of domain relevance for the audience, also in non-Western contexts. In addition, pride expressions are related to the trait to self-monitor. However, the effect of domain relevance was not dependent on the trait to self-monitor. Overall, we conclude that people are motivated to draw other’s attention to their positive outcomes, especially for those high self-monitoring. However, when pride expressions could lead to negative social costs, people can suppress their expressions accordingly.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.

Notes

1 Factor analysis can be found in the Supplementary Material (Figure S1 and Table S3).

2 The authors of the original study acknowledged this confound and reported an additional study in which this confound was eliminated and results remained the same (see van Osch et al., 2019; Study 2a and 2b).

3 All 6 items of the A-SMS are present in the R-SMS; only 3 out of 7 items of the P-SMS are present in the R-SMS. The rest of the P-SMS items are present in the full 25-item scale.